By Dr. P.H. Mapham, BVSc (Hon) and Dr. J.H. Vorster, BVSc, MMedVet(Path)
According to Dr PG Marais heartwater was first diagnosed in the North-eastern Transvaal in 1838 (now Mpumalanga) and he makes the interesting comment that the actual origin of the disease is unknown and that it is doubtful whether the disease is indigenous to South Africa. Nevertheless, over the years heartwater has caused considerable loss to livestock producers and frustration to veterinarians trying to control it. Recently reports of it occurring where it wasn’t found previously have been emerging and it is appropriate for veterinarians to review this significant disease. Much of the following article was adapted from OIE information
Heartwater is caused by Ehrlichia ruminantium (formerly Cowdria ruminantium) which is of the order Rickettsiales, Family Anaplasmataceae. These are small, gram negative, pleomorphic cocci, which are obligate intracellular parasites. There are diverse strains of E. ruminantium which vary in virulence and have a high level of genomic plasticity. Several different genotypes may coexist in a geographical areas, and may recombine to form new strains according to the OIE. The heartwater organism is extremely fragile and cannot persist outside of a host for more than a few hours.
Heartwater occurs only where its Amblyomma tick vectors are present and infection depends on the epidemiological triad of the tick vector, causative agent, and vertebrate hosts. The tick vector is subject to seasonal, environmental and management variables. The principal method of bringing the disease into an area is by introduction of infected ticks or carrier animals. Ticks become infected by feeding on acutely ill or sub clinically infected animals. According to the OIE it is not known for how long wild or domestic ruminants can be a source of infection for ticks in nature, but it may be many months. Ticks are a robust reservoir of E. ruminantium, and infection can persist in them for at least 15 months.
Of the 13 species capable of transmitting the disease, A. variegatum (tropical bont tick) is by far the most important because it is the most widespread. Other major vector species are the bont tick A. hebraeum (in southern Africa), A. gemma and A. lepidum (in Somalia, East Africa and the Sudan). A. astrion (mainly feed on buffalo) and A. pomposum (distributed in Angola, Congo [Dem. Rep. of the] and Central African Republic) are also natural vectors of the disease.
Four other African ticks, A. sparsum (feed on reptiles and buffalo mainly), A. cohaerans (feed on African buffalo), A. marmoreum (adults occur on tortoises and immature stages on goats) and A. tholloni (adults feed on elephants) experimentally transmit heartwater. Three North American species of Amblyomma ticks also experimentally transmit heartwater: A. maculatum (the Gulf Coast tick), A. cajennense (the Cayenne tick) and A. dissimile, but none of these ticks has been incriminated so far in natural transmission of heartwater. A. maculatum is widely distributed in the eastern, southern, and western U.S., and feeds on ungulates (cattle, sheep, goats, horses, pigs, bison, donkeys, mules, white-tailed deer, sambar deer and axis deer), various carnivores, rodents and lagomorphs, marsupials, birds, and reptiles. A. maculatum was shown to be as efficient as A. hebraeum, and was susceptible to a wide range of E. ruminantium strains. A. cajennense has host preference similar to A. maculatum but is not as widely distributed and is a less efficient heartwater vector. A. dissimile feeds on reptiles and amphibians.
Amblyomma ticks are three-host ticks whose life cycles may take from 5 months to 4 years to complete. Because the ticks may pick up the infection as larvae or nymphs and transmit it as nymphs or as adults, the infection can persist in the tick for at least 15 months. Infection does not pass transovarially. While transmission of heartwater can be by adult and nymphal ticks in the field, in general adults prefer to feed on cattle and nymphs on sheep and goats. Cattle egrets have been implicated in the dispersal of Amblyomma ticks in the Caribbean.
The causative agent, E. ruminantium multiplies in vascular endothelial cells throughout the body to cause vascular compromise with varying virulence. Because of its extreme fragility, it is only able to be transmitted by ticks in the presence of infected animals. Vertebrate hosts include all domestic and wild ruminants, but indigenous domestic and wild ruminants appear to be less susceptible than other domestic ruminants. Heartwater causes severe disease in cattle, sheep, and goats, with milder disease in some indigenous African breeds of sheep and goats, and inapparent disease in several species of antelope indigenous to Africa. Bos indicus (Zebu) cattle breeds are in general more resistant than Bos taurus (European) breeds. Up to 80% of merino sheep may die, but the mortality rate can be only 6% in Persian or Afrikaner sheep.
Angora and Saanen goats are also very susceptible to heartwater, while indigenous goats in certain areas may show a significant level of resistance. Heartwater has caused mortality in the African buffalo (Syncerus caffer) in some situations while deaths in springbok (Antidorcas marsupialis) in South Africa have been attributed to heartwater. Other species that have been shown to be susceptible are the blesbok (Damaliscus albifrons), the black wildebeest (Connochaetes gnu), the eland (Taurotragus oryx oryx), giraffe (Giraffe camelopardalis), greater kudu (Tragelaphus strepsiceros), sable antelope (Hippotragus niger), sitatunga (Tragelaphus spekii), steenbok (Raphicerus campestris), and lechwe (Kobus leche kafuensis). It is believed that these species may serve as reservoirs of heartwater and the disease in these animals is usually mild or undetectable. According to the OIE there is one report of an African elephant dying of heartwater, but this animal was also infected with anthrax.
Many non-African ruminants are experimentally susceptible to heartwater including the Timor deer (Cervus timorensis) and chital (Axis axis) of southern Asia, and white-tailed deer (Odocoileus virginianus) of North America.
Other species suspected to be susceptible to heartwater but lacking in definitive proof are nilgai (Boselaphus tragocamelus), fallow deer (Damadama), Himalayan tahr (Hemitragus jemlahicus), barbary sheep (Ammotragus lervia), mouflon (Ovis aries), blackbuck (Antilope cervicapra), and white and black rhinoceros.
Wildlife used in farming, seem to be the main wild ruminant species in which heartwater can have a significant economic impact. The OIE states that it was once believed that the guinea fowl and leopard tortoises were the nonruminant hosts of E. ruminantium, but recent data have confirmed that these species are not susceptible and do not transmit to vector ticks that usually feed on them. The scrub hare’s susceptibility to infection is also not fully substantiated. Although the striped mouse and the multimammate mouse have been shown to be susceptible to E. ruminantium, they are not hosts of the vector ticks and are not believed to play a role in the epidemiology of heartwater. Some laboratory inbred strains of mice have been shown to be susceptible to E. ruminantium and have assisted in defining disease and immune mechanisms, but these are not indicated as important in disease maintenance.
Heartwater is transmitted trans-stadially by ticks of the genus Amblyomma, which are biological vectors of heartwater. Ticks become infected by feeding on an infected vertebrate host during the febrile reaction. Whole blood or plasma from the vertebrate is infective, but the highest levels of the agent occur during the second or third day of fever.
Heartwater can apparently be transmitted vertically and through colostrum of carrier dams containing infected cells (reticuloendothelial cells and macrophages).
Transmission can also occur by intravenous inoculation of blood, tick homogenates or cell culture material containing E. ruminatium.
Horak et al reported the recovery of adult A. hebraeum from rhinoceroses at four localities in the Free State constituting a new distribution record. They believe that the ticks were introduced into the province on the animals concerned and warn that should A. hebraeum become established in the Free State where the climate and vegetation may be suitable for its survival, it will be nearly impossible to eradicate because of the approximately 15 000 eggs laid by engorged females. Should cattle or antelope species that are asymptomatic carriers of the organism be translocated from regions in which heartwater is endemic to properties in the Free State where A. hebraeum has become established foci of infected ticks may arise with serious consequences.
Figure 1 Madder Horak and Stoltsz
Heartwater occurs in nearly all the sub-Saharan countries of Africa where Amblyomma ticks are present and in the surrounding islands: Madagascar, Reunion, Mauritius, Zanzibar, the Comoros Islands and Sao Tomé. The disease is also reported in the Caribbean (Guadeloupe, Marie-Galante and Antigua), from where it threatens the American mainland.
The average incubation period in natural infections is 2–3 weeks, but can vary from 10 days to 1 month. The incubation period after intravenous blood inoculation is seven to 10 days in sheep and goats, and 10 to 16 days in cattle. However, the incubation is strongly dependent on the inoculated dose of elementary bodies. The outcome can range from 100% death with high doses to 0% death with low doses followed by protection of animals.
Heartwater occurs in four different clinical forms, determined by variations in host susceptibility, agent virulence and infective dose. Peracute disease is usually seen in Africa in non-indigenous breeds of sheep, cattle and goats. Heavily pregnant cows are particularly susceptible to this form. Peracute disease is characterised by sudden death preceded by a brief interval of fever, severe respiratory distress, hyperesthesia, lacrimation and, in some breeds of cattle, severe diarrhoea. Terminal convulsions may be seen. This form of heartwater is relatively rare.
Acute disease is the most common form of heartwater in domesticated ruminants, and is seen in both non-indigenous and indigenous cattle, sheep and goats. Animals with the acute form of heartwater usually die within a week.
Disease begins with pyrexia, which may exceed 41°C within 1–2 days after onset. It remains high for 4–5 weeks with small fluctuations and drops shortly before death. Fever is followed by inappetence, sometimes listlessness, diarrhoea (particularly in cattle), and dyspnoea indicative of lung oedema.
Nervous signs develop gradually, and are generally less pronounced in sheep and goats than cattle. The animal is restless, walks in circles, makes sucking movements and stands rigidly with tremors of the superficial muscles. Cattle may push their heads against a wall or present aggressive or anxious behaviour.
In the terminal stage the animal falls to the ground in lateral recumbency, pedalling and exhibiting opisthotonos, nystagmus, hyperaesthesia, chewing movements, and frothing at the mouth. The animal usually dies during or following such an attack. On rare occasions, heartwater occurs as a subacute disease with prolonged fever, coughing and mild incoordination. CNS signs are inconsistent in this form. The animal either recovers or dies within 1 to 2 weeks.
Mild or subclinical infections may be seen in young calves, lambs or kids; partially immune livestock; some indigenous breeds; and some wild ruminants. The only sign may be a transient fever. Morbidity is highly variable and depends on the degree of tick infestation, previous exposure to infected ticks, and level of acaricide protection.
Once signs of the disease have developed, the prognosis is poor for non-native and exotic sheep, goats, and cattle. The mortality rate in non-native breeds of sheep and goats may be 80% or higher, in contrast to 6% in native breeds. In cattle, mortality of 60–80% is not uncommon. Recovery from heartwater infection usually results in complete immunity against homologous strains, although animals remain carriers of infection.
Pathology and Diagnosis
The pathology caused by heartwater is suggestive but would not allow for a definitive diagnosis. A full set organ samples (brain, lung, heart, liver, spleen, kidneys, lymph nodes, gastrointestinal sections and muscle) should be collected in 10% formalin to confirm the presence of lesions consistent with heartwater.
The macroscopic lesions observed are fairly similar in cattle, sheep, and goats. The accumulation of fluid (straw coloured to more reddish) is typical and carcasses may present with lesions of hydropericardium, hydrothorax, pulmonary oedema, oedema of the mediastinal and bronchial lymph nodes. Congestion may be seen in many organs along with variable degrees of petechiation of organs and serosal surfaces.
Petechiae on the epicardium and endocardium and congestion of the brain may be prominent. Parenchymatous organs may be enlarged and moderate splenomegaly may be seen. Pulmonary oedema may be prominent and froth in the trachea is often seen. Macroscopic lesions are not usually seen absent except for subtle swelling of the brain, which may result in herniation through the foreman magnum. Congestion and/or oedema of the abomasal folds are commonly observed in cattle, but less so in sheep and goats. Nephritis of varying degree, especially in Angora goats, may be seen.
A diagnosis of heartwater may be confirmed following the demonstration of colonies of E. ruminantium in brain smears. The entire brain should preferably be removed and the hippocampus is the preferred region to sample but the cerebral grey matter could also be collected.
Brain tissues are crushed and smears are prepared spreading the crushed brain material thinly and more thickly, at regular intervals, in a single cell layers. Smears are the air-dried, fixed with methanol and stained with Giemsa or Diff Quick to demonstrate the presence organisms in the capillary endothelial cells of the brain. Smears can be fixed and stained in less than one minute. Heartwater colonies are reddish-purple to blue, and seen very often close to the nucleus of the infected endothelial cell. More than one smear should routinely be prepared as organisms may be very few and difficult to find, particularly in peracute cases. However, they should always be present in the brain of an animal that has succumbed to heartwater.
Colonies may still be visible for a period of two days after death in brains stored at tolerable room temperature; and up to 34 days in brains stored in a refrigerator. Organisms could also still be seen in the brains of animals with advanced autolysis. Fixed or unfixed smear may still suitable for a diagnosis for at least one month after having been prepared. Difficulties may be experienced in demonstrating organisms if he animal has been treated with drugs (48 – 60 hours previously) or in peracute clinical cases.
Histology of formalin fixed tissues is often unrewarding but IMP staining on such tissue has been successfully used to detect the presence of antigen in formalin fixed tissue samples collected from typical lesions during post mortem. The brain would be the organ of choice.
Serological tests include indirect fluorescent antibody tests, enzyme-linked immunosorbent assays (ELISAs) and Western blotting. However, cross-reactions with Ehrlichia spp may occur in some these tests. These tests have specific application and may be employed to monitor experimental infections, to check the immune status of immunised animals, and to screen animals prior to importation. As diagnostic test it is of less value.
The MAP1-B ELISA and the MAP1 competitive ELISA have been shown to have a higher specificity for E. ruminantium, and it does not cross-react with antibodies against Anaplasma bovis, Anaplasma ovina and Anaplasma phagocytophila. The MAP 1B ELISA test does, however, detect antibodies to other Ehrlichia spp including E. canis, E. chaffeensis, an unidentified Ehrlichia sp, infecting white-tailed in the south-eastern USA, and E. ruminantium (Omatjenne). The MAP 1B ELISA test works well in sheep and goats, but not cattle. In cattle antibody levels against E. ruminantium can be very low in heartwater endemic areas, even in vaccinated cattle or continuously naturalle challenged by infected ticks. Cattle may become seronegative 14 to 33 weeks after initial exposure. A serum sample, in a red or yellow stopper tube, is required. It is to be stored at 4° C and transported to the laboratory as soon as possible on ice.
PCR assays generally are less suitable for large-scale testing and therefore not likely to replace serological tests in epidemiological studies. PCR assays are useful as confirmatory tests and in some specific instances such as regulatory testing, research on the Ehrlichia genome and for epidemiological studies. Blood samples, in EDTA tubes are suitable. PCR can also be attempted on other target organs, such as brain, lungs, kidneys, and thoracic fluids Samples are to be stored at 4° C and transported to the laboratory on ice, as quickly as possible.
E. ruminantium may be isolated from the blood of an infected host by cultivation on ruminant endothelial cells and the presence of characteristic morulae may be confirmed by immunofluorescence or immunoperoxidase techniques using a specific antiserum. Blood samples, in EDTA tubes are suitable. Samples are to be stored at 4° C and transported to the laboratory on ice, as quickly as possible.
The peracute form of heartwater may be confused with anthrax. While the acute form may resemble rabies, tetanus, bacterial meningitis or encephalitis, babesiosis, anaplasmosis, cerebral trypanosomiasis, or theileriosis. It must also be differentiated from poisoning with strychnine, lead, ionophores and other myocardial toxins, organophosphates, arsenic, chlorinated hydrocarbons, or some poisonous plants. Accumulations of fluid similar to heartwater are also sometimes seen in heavy helminth infestations (haemonchosis).
Treatment and Control
P G Marais reports that several drugs have been used with varying success to treat animals suffering from heartwater. Tetracyclines, especially oxytetracycline, however, are most effective and are widely used for the specific treatment of the disease. Short-acting formulations of oxytetracycline are most commonly used at a dosage rate of 10-20 mg/kg body weight, either administered intramuscularly as a single dose, or half the calculated dose is given intravenously and the other half intramuscularly. This treatment is usually repeated 24 hours later. A long-acting oxytetracycline preparation has given similar results to the short- acting formulations of the drug in the treatment of clinical cases of heartwater.
OIE recommendations include oxytetracycline at 10 mg/kg or doxycycline at 2 mg/kg during the early, febrile stages of this disease, but they suggest that animals often die before treatment can be administered and antibiotic treatment alone is not always successful in later stages. Tick control is imperative because E. ruminantium cannot survive outside a living host for more than a few hours at room temperature and heartwater is usually introduced into free areas by infected animals, including subclinical carriers, or by ticks.
Carrier animals may be detected by serological tests before introduction, but these tests have limitations according to the OIE. Animals moved into endemic areas may be protected by chemoprophylaxis. PG Marais describes this as a procedure by which a series of oxytetracycline injections is used to protect susceptible animals against heartwater when they are introduced into an endemic area. He suggests that the development of such a prophylactic regime holds great promise as a practical and reasonably safe method for the introduction of large groups of animals into endemic areas. However, he says, its success is almost exclusively dependent on all the animals becoming naturally infected with heartwater during the time that they are protected by the oxytetracycline.
During 1972 CJ Mare reported investigating prolonged oral administration of oxytetracycline to manage artificially-induced heartwater in sheep. He fed 10 sheep per group 500g per day of meal containing 0.2, 0.3 or 0.4 kg of Terramycin A/D fortified crumbles per Kg of feed for 25 days. A control group of ten sheep had no antibiotic added to the feed. On the second day of feeding he inoculated all the sheep with 5ml of fresh sheep’s blood infected with the Ball 3 strain of heartwater after which temperatures were taken daily. The sheep were observed for a period of 38 days. All surviving sheep were challenged on the 38th day by the intravenous inoculation of 5 ml of Ball 3 heartwater blood. Eleven susceptible sheep were also inoculated as controls. Only one of the thirty antibiotic fed sheep showed typical heartwater during the 38-day period post inoculation. All the remaining sheep were immune when challenged with heartwater after 38 days.
All ten untreated control animals developed typical heartwater, with febrile responses of up to 42°C, hyperaesthesia, rapid breathing, and finally prostration, galloping movements and foaming at the mouth and nose. Despite efforts to salvage these sheep by the parenteral administration of antibiotics, 7 out of 10 died. Dead sheep were autopsied, and brain smears prepared, stained with Giemsa, and examined for the causal agent of heartwater.
Vaccination currently consists of infection with a live E. ruminantium strain, then treatment with antibiotics when a fever develops. Alternatively, the vaccine may be given to young kids or lambs during their first week of life, or to calves less than 5 to 8 weeks of age; young animals possess a degree of non-specific resistance to infection, and do not always require treatment. Vaccination does not always protect animals from all field strains, and revaccination is risky due to the possibility of anaphylactic reactions. This method is likely to be replaced with inactivated and attenuated vaccines in the future.
A first-generation vaccine consisting of inactivated purified elementary bodies of E. ruminantium emulsified in Montanide ISA 50 adjuvant has given good results in experimentally controlled conditions and has demonstrated significant protection in the field.
Three different isolates (Senegal, Gardel and Welgevonden) have been attenuated and shown to confer good protection, and significant protection has also been obtained using DNA vaccination. However, none of these new experimental vaccines has been fully validated under field conditions. Field trials have revealed that antigenic diversity is important in formulating effective vaccines, and further investigations are critical for the delivery of any vaccine in the field.
Inactivated vaccines using the Gardel strain and subsequently other strains have been produced in bioreactors for an industrial production following the development of the whole production process. The vaccine produced has the same efficacy as the initial laboratory produced experimental vaccine and a production cost calculated at around 0.1 ? per dose which becomes acceptable. The Mbizi strain inactivated vaccine is being developed commercially by Onderstepoort Biological Products in South Africa.
These inactivated vaccines do not prevent infection but do prevent or reduce death of vaccinated animals when exposed to live virulent challenge. They confer protection for more than one year. The advantage is that several field strains can be incorporated to make the vaccine more widely cross-protective.
A major challenge remains the characterisation of the extent of strain diversity in a region to be covered by an appropriate formulation of the vaccine. This knowledge will also be essential for new generation vaccines that will be developed in the future.
Live Attenuated Vaccines
Infection of ruminants with live E. ruminantium strains induces a strong long-lasting protection against an homologous isolate. This is the basis for infection and treatment using virulent isolates. Isolates of attenuated virulence that do not necessitate treatment of animals would be ideal, but a limited number of such attenuated isolates are available.
An attenuated Senegal isolate has been obtained and shown to confer 100% protection against an homologous lethal challenge, but very poor protection against a heterologous challenge. The Gardel isolate, which gives a significant level of cross-protection with several isolates (although far from complete), has also been attenuated.
Recently, a third isolate named Welgevonden from South Africa has been attenuated and shown to confer complete protection against four heterologous isolates under experimental conditions.
The main drawback of attenuated vaccines is their extreme lability, which necessitates their storage in liquid nitrogen and their distribution in frozen conditions. In addition, they have to be administered intravenously.
Several reports show partial protection of mice using map1 DNA vaccination and an improvement of protection by vaccination following a prime (plasmid) – boost (recombinant MAP1) protocol. However, protection of ruminants has never been demonstrated using this strategy.
In opposition, significant protection of sheep was reported against homologous and heterologous experimental challenge following plasmid vaccination using a cocktail of four ORFs (open reading frames) from the 1H12 locus in the E. ruminantium genome.
No further results have been described since then.
Recombinant vaccines will probably not be available soon.
Veterinarians will need to be alert to the spread of this disease into previously uninfected areas as environmental conditions change and free movement of game species and domestic ruminants occurs
1. OIE HEARTWATER – COWDRIOSIS Aetiology Epidemiology Diagnosis Prevention and Control References
2. Marais PG, Lets talk Heartwater (Cowdria ruminantium) Grootfontein Agricultural Development Institute – http://gadi.agric.za/articles/Agric/artikel_hartwater.php
3. Mare CJ, The Effect of Prolonged Oral Administration of Oxytetracycline on the Course of Heartwater (Cowdria ruminantum) Infection in Sheep. Trop. Anita. Hlth Prod. (1972) 4, 69-73
4. Horak, I.G., Jordaan, A.J., Nel, P.J., Van Heerden, J., Heyne, H. & Van Dalen, E.M., 2015, ‘Distribution of endemic and introduced tick species in Free State Province, South Africa’, Journal of the South African Veterinary Association 86(1), Art. #1255, 9pages. http://dx.doi.org/10.4102/jsava.v86i1.1255
5. Madder M, Horak IG, Stoltsz H. Tick identification http://www.itg.be/photodatabase/African_ticks_files/index.html