2. Epidemiology of OPA and JSRV Infection

The first descriptions of OPA were made more than 100 years ago in South Africa and Britain and, since then, the disease has been reported in many countries elsewhere in the world with the exception of Australia and New Zealand. The disease was eradicated in Iceland in the 1950s by drastic slaughter measures^[2].

Under natural conditions, clinically apparent OPA mostly appears in animals 1–4 years old, although the disease can occur at all ages, and there is no clear evidence of sex or breed susceptibility^[1]. The incubation period for the development of clinical disease following natural infection may last from months to years, but may be shorter (6–8 months) in flocks where OPA is not endemic. On the other hand, when very young lambs are experimentally infected, clinical signs tend to appear in 3–6 weeks or even before^[1][2].

The mortality rate in OPA affected flocks varies depending on how long the infection has been present. In the first years after introduction, the infection mortality rates can reach 30–50% (as during the OPA epidemics in Iceland in the 1930s), but rates drop to 1–5% when the disease becomes endemic^[1][3]. The prevalence of OPA appears to vary between countries and is endemic in some of them, such as Peru, Scotland, South Africa and Spain^[2]. A study conducted in a Spanish abattoir recorded visible OPA lesions in 0.3–1.4% of sheep slaughtered in a year^[5], and high figures were also obtained in an abattoir study in Edinburgh^[6], indicating that OPA prevalence may be underestimated. Results from a longitudinal survey in two OPA endemic flocks carried out in Scotland are in accordance with this. Around 30% of the sheep had histologically confirmed OPA lesions, but the annual losses attributable to OPA varied between 2 and 10% ^[2]. These findings were obtained by clinical observation and histopathological analysis and, therefore, do not reflect the prevalence of JSRV infection in OPA-affected flocks. The lack of a detectable specific immune response against JSRV in infected animals^[7][8] prevented the development of serological tests, and the acquisition of certain knowledge about this issue was not possible till the arrival of molecular techniques for the specific detection of JSRV, which demonstrated the tissue distribution of the virus outside the OPA lesion^[9]. In this way, JSRV proviral DNA could be detected by PCR in lymphoid tissues and peripheral blood mononuclear cells (PBMC) in clinically OPA affected sheep, in animals with preclinical OPA lesions and in lesion-free infected animals^{[10][11][12][13][14]}. The PCR test for the detection of JSRV proviral DNA in blood cells has been used in epidemiological studies, pointing to an incidence of JSRV infection in OPA-affected flocks much higher than previously believed^[12][14]. In a study in which a Spanish commercial flock was tested periodically over three years, at the end of the study, the virus had been detected in 50% of the animals^[14]. JSRV could be detected in animals of all age ranges, but the highest incidence (80%) was found in animals that were under one year old when the study began^[14]. Despite this, only a minority of animals that tested positive developed OPA lesions (17%), many of which were subclinical, since 40% of animals with OPA lesions at necropsy were not clinically affected^[14]. This is in accordance with a previous study indicating that many OPA cases can remain subclinical at the end of the sheep’s commercial lifespan, and induction of OPA is not a common outcome of naturally occurring JSRV infection^[12].

It is generally accepted that the respiratory route is the most important natural mode of transmission for JSRV^[1]. In recent years, the contribution of colostrum and milk (C/M) to the spread of infection in commercial sheep farms has also been investigated. Epidemiological studies demonstrated that JSRV infection can occur perinatally or in the first few months of life in lambs^[12][14], and the prevalence of JSRV infection in this age range can be particularly high in commercial flocks, with 30% of lambs destined for replacement found to be JSRV blood positive in one single PCR test ^[14]. In addition, a survey carried out in a flock with a high prevalence of OPA showed that an important reduction in the incidence of the disease was possible by creating a new flock with lambs separated at birth from their mothers and reared artificially, suggesting that C/M could be relevant for the transmission of JSRV under natural conditions^[15]. Further studies demonstrated that colostrum and milk can transmit JSRV to lambs. The presence of JSRV proviral DNA was demonstrated in somatic cells from colostrum in sheep belonging to OPA-affected flocks, and the JSRV provirus was detected in the blood of lambs artificially fed with infected C/M^[16]. Evidence of the relevance of this route of JSRV transmission to lambs under natural conditions has been provided by the detection of JSRV in Peyer’s patches and/or mesenteric lymph nodes in 25% of the lambs (between 12 h and 10 days of live) naturally fed by blood infected but asymptomatic ewes^[17]. To date, no other details of this way of transmission for JSRV are known. Based on current data, this route of transmission should be taken into account in the design of control strategies, although the possibility that C/M transmission of JSRV to lambs can result in OPA development has not been explored yet.

3. Clinical Features and Pathology

OPA clinically affected sheep show signs of progressive afebrile respiratory disease associated with cachexia caused by the growth of lung adenocarcinoma. When the lung tumour is very small, the disease is subclinical, but as the tumour becomes extensive enough to interfere with lung function, dyspnoea and moist respiratory sounds caused by the accumulation of fluid in the respiratory airways are detected^[1]. In the final stages of the disease, variable amounts of frothy sero-mucous fluid (from 10–40 mL to as much as 400 mL)^[1][18] are discharged from the nostrils when the hindquarters are raised (“wheelbarrow” test) or the head is lowered (Figure 2A), which is considered an OPA characteristic sign^[19]. Affected sheep remain alert, afebrile and have a good appetite, but progressive loss of weight is evident, and death inevitably occurs within a few weeks of the start of the clinical disease as a result of compromised respiratory function caused by tumour enlargement. However, the clinical course can be shortened, and fever appears if bacterial infections become superimposed. The concurrence of other diseases can also affect the clinical outcome of the disease. Maedi-visna has been frequently reported, and it results in a worsening of the clinical signs and a precipitation of the disease course^[1].

Two pathological forms of OPA have been described in the literature, classical and atypical^[1]. In the classical presentation, at necropsy, the lungs do no collapse when the chest is opened and are enlarged. The neoplastic lesions can occur in any part of the lungs, but cranio-ventral parts are more frequently involved. They are grey or purple in colour, do no protrude significantly on the surface and have an increased consistency (Figure 2B). The cut surface of the tumour lesion has a granular appearance and is moist, and a frothy fluid pours from the bronchioles and bronchi with slight pressure (Figure 2C). Relatively often, this tumour lesion pattern may be overshadowed by concurrent lesions of bacterial pneumonia, abscesses or maedi. In contrast with the classical form, the atypical presentation tends to be more nodular in both early and advanced tumours. The nodules may be solitary or multiple and mainly located in the diaphragmatic lobes. They are pearly white in colour and have a very hard consistency (Figure 2D). A section of the tumour lesions shows that they are very well demarcated from the surrounding parenchyma, and their surface looks dry (Figure 2E). The tracheobronchial and mediastinal lymph nodes may not show any visible changes or may be slightly or clearly enlarged, and may occasionally present small metastases^[1]. More rarely, metastases in distal organs, such as the liver, kidney, heart, skeletal muscle, digestive tract, spleen, skin and adrenal glands, have been observed^[1][3][20]. Both forms, classical and atypical, may be present in a flock and in individual sheep, and intermediate and mixed forms have been described. Classical and atypical forms may represent two extremes of the disease spectrum, rather than two separate forms. Descriptions of experimentally induced OPA are compatible with the classical forms observed in natural conditions, but atypical forms have not been reported in experimental conditions^[1].

Histological examination of OPA natural cases reveals the presence of neoplastic proliferation foci of epithelial cells in alveolar and bronchiolar areas. These proliferations have a papillary and acinar appearance and expand into adjacent structures. In the alveolar neoplastic regions, cuboidal or columnar cells replace the normal type II pneumocytes, but the structure of the alveolar wall is maintained (lepidic growth) (Figure 2F). Concurrently, polypoid ingrowths arise from the bronchiolar epithelium in affected terminal bronchioles (Figure 2G). The stroma of the tumour is generally thin but may be infiltrated by variable amounts of lymphocytes, plasma cells and connective tissue fibres. Macrophages are consistently found in variable numbers surrounding neoplastic alveoli and affected bronchioles. Neutrophils can also be found, but they are interpreted as indicative of secondary bacterial infections. In some cases, mesenchymal tissue foci (myxoid nodules or growths) have been described admixed with the neoplastic epithelial component^[1][3][21]. The histopathological features of atypical OPA are essentially the same as those of classical OPA, but a large number of inflammatory cells and connective fibres infiltrate the stroma. The histological appearance of experimentally induced OPA closely resembles that of natural cases^[1].

4. Aetiology and Pathogenesis

Jaagsiekte sheep retrovirus is the causative agent of OPA^[22][23]. This retrovirus infects and induces the transformation of secretory epithelial cells of the distal respiratory tract of sheep, and more rarely of goats and wild mouflon^[1]. JSRV is an exogenous retrovirus belonging to the genus Betaretrovirus and is highly related to enzootic nasal tumour virus (ENTV) of sheep (ENTV-1) and goats (ENTV-2), which also causes an adenocarcinoma of secretory cells of respiratory epithelia, but in the upper tract^[24][25]. Interestingly, sheep and goats, and other mammals, contain several copies of nonpathogenic JSRV-related endogenous retroviruses (enJSRVs) integrated in their genome^[26][27]. JSRV has the typical genomic organization of a simple retrovirus, and contains the genes gag, pro, pol and env. These genes, respectively, encode the proteins of the viral core (MA, CA, NC and others), the viral protease (PR), the viral reverse transcriptase (RT) and integrase (IN) enzymes and the glycoproteins of the viral envelope (the surface domain SU which interacts with the cellular receptor and mediates cellular entry, and the transmembrane domain TM). In addition to encoding viral envelope proteins, the env gene of JSRV functions as a dominant oncogene. Its sole expression is sufficient to induce cellular transformation^[28][29], and the cytoplasmic tail of the JSRV TM protein is essential for envelope-induced (Env-induced) transformation^[30]. Apart from these four common retroviral genes, JSRV has a further open reading frame (orf-x) overlapping pol gene, whose role is unknown. Noncoding regions are present at the ends of the genome: U5 is present at the 5′ end, U3 at the 3′ end and the R region is repeated at both. Once JSRV interacts with a specific cellular receptor to enter the cell (HYAL2)^[31], and after reverse transcription of the viral genome into double-stranded DNA, the viral DNA integrates into the host DNA to form a provirus. During the process of reverse transcription, the noncoding regions at the ends of the genome are duplicated and give origin to the viral long terminal repeats (LTRs), which are major determinants of retrovirus tropism. JSRV can infect many cell types; in fact, it establishes a disseminated infection of the lymphoid tissues of OPA affected sheep^[9], but the exogenous JSRV LTRs are particularly active in type II pneumocytes and Club cells of the lung, the cells where the tumour develops^[32].

The mechanisms involved in JSRV Env-induced transformation have not been fully elucidated, but several studies have shown the activation of signalling pathways that control cellular proliferation, including phosphatidylinositol 3-Kinase (PI3K)-Akt and mitogen-activated protein kinase (MAPK)^[33][34], and additional pathways, including the AGR-2-YAPI-AREG axis, may also contribute to oncogenesis in this disease^[34]. Other mechanisms, such as targeted integration of JSRV, cannot be totally excluded ^[33].

OPA has several features in common with lung adenocarcinoma of humans, including a similar histological appearance and activation of common cell signalling pathways, and additionally, the size and organization of human lungs are much closer to those of sheep lungs than to those of mice. This has led to the suggestion that OPA may be a valuable large animal model for the human disease^{[35][36][37][38]}. This model can be informative for understanding cancer in humans and can identify and test the efficacy of new therapeutic interventions in a high reproducible system^[38].

5. Diagnosis

As stated above, diagnosis of clinical OPA is possible by the detection of moist respiratory sounds and the presence of frothy lung fluid emitted from the nostrils. The overproduction of lung fluid is a characteristic clinical sign, and in the final stages of classical OPA, variable amounts of nasal discharge are obtained when the rear limbs are raised (“wheelbarrow” test) or the head is lowered. The diagnosis can be confirmed by the detection of JSRV RNA in lung fluid samples using reverse transcriptase PCR^[9]. However, not all cases of OPA produce this fluid in detectable amounts, such as the early OPA, and also atypical OPA, even in advanced stages. In these cases, post-mortem examination is needed for OPA diagnosis and gross pathology and histopathological changes described above should be observed.

Lesions of OPA can be confirmed by immunohistochemical methods for the detection of JSRV proteins, using antibodies against proteins encoded by gag and env genes^[39][40]. Immunolabelling is associated with the cytoplasm of transformed alveolar and bronchiolar cells (type II pneumocytes and Club cells, respectively) where JSRV replicates actively (Figure 2H). In addition, JSRV proteins have been demonstrated in myxoid nodules and also in the infiltrating lymphoreticular cells of some early OPA lesions^[21].

JSRV proteins can be also detected in tumour homogenates by the Western blotting technique^[41]. In addition, OPA tumours are always positive when tested by PCR techniques for the detection of JSRV genome^[9]. These tests are based on the detection of JSRV RNA by reverse transcriptase PCR, but JSRV proviral DNA can also be specifically detected by PCR. In this case, primers are designed to amplify the U3 region of the JSRV genomic sequence, in which major differences with JSRV-related endogenous retroviruses that the sheep genome contains are located.

However, in vivo identification of OPA preclinical cases and lesion-free infected animals would be vital for the implementation of effective strategies for the control and eradication of the disease. Unlike other ovine retroviral infections, such as visna-maedi virus (VMV), the absence of a specific antibody response in JSRV infected animals^[7][8] has precluded the use of diagnostic serological tests. The design of PCR techniques for the specific detection of JSRV provirus integrated in the sheep genome^[9] revealed the presence of JSRV in lymphoid tissues and PBMC in clinically OPA affected sheep, in animals with preclinical OPA lesions and in lesion-free infected animals^{[10][11][12][13][14]} and opened the door to the development of blood PCR tests for preclinical diagnosis of OPA. Although this PCR blood test is very specific, it has low sensitivity and provides an inconsistent detection of JSRV^{[11][13][14][42]}, probably due to the low proportion of infected cells in the blood^[43]. Therefore, this test is not suitable to test individual animals for accreditation purposes but can be applied for the identification of infected flocks^[13] and has been used in epidemiological studies ^[12][14]. Other tests have been investigated in order to improve the sensitivity in the detection of OPA preclinical cases. The same PCR test on bronchoalveolar lavage samples collected from live animals provides better results than the blood PCR test for the detection of early OPA (early visible lesions and microscopic OPA), and its sensitivity is 89% in comparison to the results of histological examinations^[44]. However, this test does not detect lesion-free infected animals which may develop the disease in the future, and the practical difficulties in collecting the samples prevent its application at the field level. PCR testing of bone marrow aspirates collected in asymptomatic infected sheep has been attempted with negative results, although positively labelled cells were revealed by immunohistochemical methods in bone marrow samples collected at necropsy^[45]. The same PCR when applied to colostrum and milk samples of JSRV blood positive lactating ewes with no signs of OPA disease did not seem to be more sensitive than the blood PCR test, and also provided an inconsistent JSRV detection throughout a lactation period^[17]. Apart from PCR tests, transthoracic ultrasonography of both sides of the chest is another method that has been investigated for the in vivo detection of OPA lesions, in an attempt to eliminate this disease^[46][47]. This test may be very helpful in reducing the OPA prevalence in a flock by identification and culling of affected animals. However, the test currently lacks sufficient sensitivity and specificity for the diagnosis of early stages of the disease. It is not able to detect lesions smaller than 2 cm and cannot clearly discriminate some nodular parasitic lesions or suppurative pneumonias that could be confused with OPA nodules. A negative scan cannot provide a guarantee that the animal is free of JSRV infection nor early OPA, and re-scanning is recommended in a short time, as tumours can develop much faster (a few months) than previously thought^[48]. Therefore, eradication of OPA based only on this method seems to be unlikely. More recently, other approaches to detect preclinical OPA have been tested, such as reverse transcriptase PCR tests on nasal swabs. These seem to be more sensitive than the blood PCR test, but they have been proposed at flock level, not for testing individual animals^[49]. Tests based on biomarkers are also in progress^[49]. In this regard, in a recent study, levels of several tumour markers were found to be significantly higher in the blood of sheep with clinically suspected OPA and lesions confirmed by macroscopic and histopathological examination, than in lesion-free animals^[50]. These tumour markers are thought to facilitate the diagnosis of OPA, but its possible usefulness in the diagnosis of subclinical OPA has not been investigated.