EcoHealth

, 4:346

Towards a Case Definition for Devil Facial Tumour Disease: What Is It?

  • Stephen B. Pyecroft
  • Anne-Maree Pearse
  • Richmond Loh
  • Kate Swift
  • Kathy Belov
  • Nolan Fox
  • Erin Noonan
  • Dane Hayes
  • Alex Hyatt
  • Lingfa Wang
  • David Boyle
  • Jeff Church
  • Debra Middleton
  • Robert Moore
Special Focus: Tasmanian Devil Declines

DOI: 10.1007/s10393-007-0126-0

Cite this article as:
Pyecroft, S.B., Pearse, AM., Loh, R. et al. EcoHealth (2007) 4: 346. doi:10.1007/s10393-007-0126-0

Abstract

In the mid 1990s an emerging disease characterised by the development of proliferative lesions around the face of Tasmanian devils (Sarcophilus harrisii) was observed. A multi-disciplinary approach was adopted to define the condition. Histopathological and transmission electron microscopic examination combined with immunohistochemistry help define Devil Facial Tumour Disease (DFTD) as a neoplastic condition of cells of neuroendocrine origin. Cytogenetic analysis of neoplastic tissue revealed it to be markedly different from normal devil tissue and having a consistent karyotype across all tumours examined. Combined with evidence for Major histocompatability (MHC) gene analysis there is significant evidence to confirm the tumour is a transmissible neoplasm.

Keywords

Tasmanian devilSarcophilus harrisiineoplasmkaryotypefacial tumourneuroendocrine

Introduction

In the mid 1990s a new disease was observed affecting Tasmanian devils (Sarcophilus harrisii) and an increasing numbers of animals were submitted for examination and diagnosis by veterinary pathologists at the Animal Health Laboratories, Department of Primary Industries and Water, Launceston Tasmania. As the prevalence of this disfiguring facial tumour disease increased in many populations throughout the state it became quickly apparent that it was presenting as a new and emerging disease of the iconic Tasmanian devil.

Initially many questions were asked of the investigators as to the cause of this apparently fatal disease. The investigating team that was formed from the state diagnostic labs in collaboration with a number of other centres of animal disease research within Australia and overseas were guided by advice attributed to Hughling Jackson that :
  • “The study of things caused must precede the study of causes of things”

(York and Steinberg, 2002). Within the research team the focus was to identify what was the nature of this new disease or in epidemiological terms formulate a case definition.

Similar to other disease investigations in animals this key initial step would determine the direction of many other research priorities and directions and more importantly control of the disease. A scientific forum determined that the key outcome of research on Devil Facial Tumour Disease (DFTD) was that the “Tasmanian devil be maintained in an ecologically sustainable population in the wilds of Tasmania”. It was therefore significantly important that researchers formulating the case definition be accurate in their assessment.

A multi-disciplinary approach was adopted using standard veterinary pathology techniques and others more commonly utilised in the diagnosis and study of human conditions. The marriage of approaches has been serendipitous in moving towards the prompt resolution of a case definition for DFTD.

Pathology

Using standard pathological techniques of cytology from fine needle aspirates of lesions, histology and transmission electron microscopy, Loh et al. (2006a) found that the cells in the facial masses of DFTD were consistent with an undifferentiated neoplasm. Neoplasms had been described from devils previously (Griner, 1979); however, none had the characteristics of this new disease. Affected devils consistently had disfiguring and debilitating facial tumours (Figure 1) and initially there was very little agreement on the cell type and classification of this tumour as DFTD cells were highly pleomorphic and anaplastic. One consistent finding was that the tumours appeared morphologically similar in affected animals from a wide geographic range in the state.
https://static-content.springer.com/image/art%3A10.1007%2Fs10393-007-0126-0/MediaObjects/10393_2007_126_Fig1_HTML.jpg
Figure 1.

Advanced lesion of Devil Facial Tumour (DFT) producing marked disfigurement of the upper and lower lips and associated peri-oral tissue in an anaesthetised Tasmanian devil (Sarcophilus harrisii).

Further classification of the neoplasm was undertaken using Immunohistochemistry (IHC) techniques and Loh et al. (2006b) describe in detail the challenges presented by adopting techniques to the study of a disease in a dasyurid marsupial that have been develop for the study of conditions in placental mammals.

Despite the challenges DFTD cells tested positive for vimentin, S-100, melan A, neuron specific enolase, chromogranin A and synaptophysin. From this information combined with the morphological and ultrastructural features of DFTD it was concluded that the neoplasm was consistent with a malignant neuroendocrine tumour.

Cytogenetics

In parallel with this work cytogenetic studies were being undertaken to describe the normal karyotype of the Tasmanian devil. Like so many aspects of the Tasmanian devil very little was known of the basic chromosomal characteristics of the species. Standard cytogenetic techniques were adapted to the study of devil tissues and cytogenetic characterisation of this unique neoplasm was established.

The high degree of aneuploidy displayed in DFTD cells was remarkable in the extent and consistency (Pearse and Swift, 2006) (Figure 2a and 2b). Like the morphological descriptions derived from pathological studies the aneuploidy was consistent in tumours from animals in many geographical locations within the state. Rearrangements were identical in male and female animals from a range of ages indicating that cytogenetically, DFTD is relatively stable. Further work has identified three new cytogenetic strains of DFTD. However, these are closely related and appear to be easily derived from the original strain (1).
https://static-content.springer.com/image/art%3A10.1007%2Fs10393-007-0126-0/MediaObjects/10393_2007_126_Fig2_HTML.gif
Figure 2.

The normal karyotypes of the Tasmian devil (Sarcophilus harrisii). Female (A) and Male (B). The karyotype of DFT showing marked changes in chromosomes 2, 4 and 6 and the formation of markers M1, M2 and M3.

The work of Pearse and Swift (2006) identified a characteristic marker karyotype, which is asexual and consistent. During this defining work one animal affected by DFTD was shown to have a constitutional chromosome anomaly that was not exhibited by the cancer cells within the animal. The degree of chromosomal rearrangement that had been observed in all tumours, the consistency of the rearrangements combined with this finding lead researchers to hypothesise that the cancer was acting as a transmissible graft. Spread appears to be from one animal to another during close physical contact and particularly at times of biting. One other case of transmissible neoplasm is described in domestic animals, Canine Transmissible Venereal Tumour (CTVT) and so a precedence for the nature of such tumours had been seen previously (Bloom et al., 1950; Dingli and Nowak, 2006; Murgia et al., 2006; VonHoldt and Ostrander, 2006).

Further Support for a Transmissible Neoplasm

Subsequent unpublished work has given strength to this hypothesis. Molecular genetic studies reveals that DFT has unique Major Histocompatablity Complex (MHC) Class I and II genes that differ from those identified in the host devil (K. Belov pers com).

The application of Amplified Fragment Length Polymorphism (AFLP) technology has also revealed that there is a significant quantitative genetic difference between host cells and DFT cells (N. Fox pers com).

The most significant confirmative indicator that DFT is a transmissible neoplasm has been the successful experimental transfer of DFT cells derived from cultured cell lines and natural tumours to devils with the subsequent variable development of DFTD (Pyecroft et al., in prep). This work has fulfilled elements of Koch’s postulates confirming the transmissible neoplasm hypothesis. The cells of DFT were isolated in culture and then transferred to an unaffected devil. The cells grew to produce a subcutaneous tumour and then went onto produce a fulminating DFTD in the treated animal. Whilst these postulates were originally used in the context of infectious bacterial disease the findings from these studies has shown that DFT has the characteristics of an infectious agent.

Aspects of Pathogenesis and Epidemiology of DFTD

DFTD is a debilitating disease and in wild populations has lead to the reduction in devil populations across the state of Tasmania with the exception of the west coast and pockets in the mid-north coast where trapping has shown no evidence of DFTD. Hawkins et al., (2006) present data from a number of sources that have shown a significant decline in devil numbers as a result of the emergence of DFTD and possibly other causes. Initial work by the DFTD research team described the nature of the disease but the emphasis is now to characterise the pathogenesis of the disease in an individual animal.

The mechanism by which neoplasm cause death in affected animals has been much theorised. In the case of DFTD, it is presumed to be one or a combination of factors. The animals may die as a result of septicaemia from secondary infections, which together with tumour necrosis may release toxins. In advanced cases of the disease the facial tumours enlarge to the point of erosion through the epidermis leaving ulcerative proliferative masses. These lesions are prone to secondary infection by environmental organisms. The large exposed surface of some neoplasms may lead to protein loss due to exudation. In rapidly growing lesions there may be caloric diversion to tumour growth. Many neoplasms produce inappetence factors, which cause appetite loss. Not uncommonly, the DFTD neoplasms cause space-occupying lesions. If these lesions are within the oral cavity, as many of them are, they prevent prehension and mastication of food. They may also interfere with a number of senses utilised for locating food (e.g., there may be occlusion of the eye or damage to whisker beds). In some cases, the metastatic locations of DFTD have been observed to impact on the organs either as space-occupying lesions (e.g., heart or brain) or alter the function of an organ. Pain may also be a factor. Understanding these possible influences of DFTD on the host will allow researchers to better define the disease functionally. The work to investigate these possible mechanisms continues.

Further work is also being undertaken to better define the mechanism of natural transmission between animals. Loh et al. (2006) showed that 62 percent of all discreet non-metastatic lesions observed within the study were on the oral mucosa (Figure 3). Whilst this observation would suggest that the facial area of a devil may be the area of first implant of cells at time of agonistic interaction between devils these tumours may not necessarily develop at the site of inoculation. The seed and soil theory is normally used to explain metastatic locations of tumours and similarly it may also explain the distribution of DFTD. Neuroendocrine cells are widely dispersed throughout the body in many mammals and where they do not form discrete organs, they are normally in especially high concentrations in tactile tissues such as the lips and in the whisker beds (Meuten, 2002). It is possible that DFT cells may “plant” themselves in the more suitable stroma of the facial region after inoculation from a number of areas.
https://static-content.springer.com/image/art%3A10.1007%2Fs10393-007-0126-0/MediaObjects/10393_2007_126_Fig3_HTML.jpg
Figure 3.

Two oral DFT lesions in the roof of the mouth of a young devil.

Oncogenesis and Predisposing Factors

Whilst investigations continue to allow a greater understanding of the nature of DFTD there is a developing curiosity as to what factors may have influenced the development of the neoplasm in the index devil, which has evolved to become DFT. Comparative studies show that genetic factors play a major role in the development of neuroendocrine tumours (NETs) in many species (Gratzinger, 2004; Barakat et al., 2004) and so much focus is being placed in the molecular and genetic aspects of the disease and the devils.

The DFTD research to date has shown that the Tasmanian population of devils carry a gross autosomal polymorphism. Some of the devils are heterozygous for a chromosome 5, which is identical (with G banding) to that found in other Dasyurids such as the Quolls. The C5 commonly found in devils is derived from the ancestral C5 (AC5) by a pericentric inversion (piC5). The prevalence of this anomaly, whilst still not accurately determined, appears to be much higher than could be expected from random mutation. The distribution of AC5 in the devil population suggests that it is a remnant chromosome rather than a reversion as it is found in the geographically isolated populations of devils on the West Coast as well as in contiguous populations. This finding raises a number of important issues relating to the evolutionary status of Sarcophilus. In addition it may also have implications with particular reference to the predisposition of devils to developing a transmissible neoplastic disease. Both questions are being further investigated (Pearse et al, in prep).

There are six viral families known to be associated with cancer, Hepadnaviridae, Polyomaviridae, Papillomaviridae, Adenoviridae, Herpesviridae and Retroviridae. It would therefore make sense to investigate the involvement of viral agents particularly with respect to their possible role in initiating the neoplastic condition that has evolved into DFT.

If an infectious agent is involved in the manifestation of DFTD then it may be detected by approaches ranging from transmission electron microscopy (TEM) to molecular biology involving cDNA subtraction and microarray hybridisation.

Examination of 29 lesions from 12 affected animals by TEM has, to date, failed to detect viruses from any of these or any other viral families (D. Hayes pers com). Additional tissue samples including cultured cells and supernatants from these cell cultures are being analysed for the presence of viral agents. A separate trial looking for the presence of cytopathic effect (CPE) on a number of commercial cell lines after exposure to DFT and DFT cell culture supernatants is also be undertaken. Whilst these studies are progressed, other work involving the use of two-dimensional DNA microarrays for the rapid detection and characterisation of unknown pathogens are being undertaken. These arrays (Palacios et al., 2007) involve the use of 60mer oligonucleotides designed for the detection of all vertebrate viruses for which there is sequence available in GenBank. cDNA subtraction and random sequencing (Murgia et al., 2006) techniques are also being used for the molecular identification and characterization of putative viruses. Results from the three approaches will provide a comprehensive data set for supporting or otherwise the hypothesis that tumours associated with DFTD have a viral aetiology.

Pre-clinical Diagnosis

Other ongoing studies are attempting to identify markers of DFTD disease. Identification of bio-markers for use in the detection of pre-clinical disease is of critical importance for the current and future management of DFTD. Whether DFTD is caused by an infectious agent or by direct transfer of cancer cells, there is a possibility that there will be a host response. Such biomarkers could be antigens, disease-specific antibodies and/or changes in tissue structure (e.g. hair and whiskers, Fraser et al., 1964; Church, et al., 1997) resulting from the disease. Whilst these markers may be of host origin they will still add to the characterisation of DFTD.

Investigations into the use of X-ray diffraction and infrared spectroscopy for the pre-clinical diagnosis of DFTD have commenced. Initial X-ray results found no detectable ordered keratin in the whisker samples. Initial infrared investigation has been undertaken using attenuated total reflectance spectra (Lyman and Murray-Wijelath, 2005) obtained from greasy fibres. Chemometric data analysis (Woodhead et al., 1997) was employed to enhance the detection of differences between fibres derived from animals with and without overt signs of DFTD. Results from analysis of a very limited data set is shown as Fig. 4 and suggests the possibility of discrimination between diseased animals shown as dashed clusters and control animals.
https://static-content.springer.com/image/art%3A10.1007%2Fs10393-007-0126-0/MediaObjects/10393_2007_126_Fig4_HTML.gif
Figure 4.

The PCA (Principal component analysis) scores-scores plot obtained using the 3600–2700 and 1800–750cm-1 spectral regions of the ‘greasy’ fibre spectra. The data obtained from the diseased animals form the clusters circled with dashed lines.

Concurrent molecular studies including the use of random phage display peptide libraries (Wang and Yu, 2004) are also being used to assess the presence of disease-specific proteins which can ultimately be used for large-scale screening of Sarcophilus harrisii for the presence of DFTD. These studies like many listed above are in progress and will add further definition to the case definition that is DFTD.

To date DFTD is described as being a moderate morbidity and high mortality disease caused by a transmissible neoplastic cell line, which is derived from neuroendocrine tissue. The neoplasm is an undifferentiated subepithelial sarcoma that has a high likelihood of metastasis particularly to regional lymph nodes (Loh et al., 2006a). The work that is progressing will endeavour to give a better understanding of the pathogenesis of the disease in an attempt to control it better in the wild populations of Tasmanian devils.

Acknowledgments

The authors would like to thank all team members of the DFTD research group within DPIW for their assistance as well as those in the AHL, DPIW Tasmania and AAHL and TFT, CSIRO Geelong, who have contributed to diagnostic results. Sections of the work were undertaken under DPIW AEC Certificate # 33/2004-05 and 33/2005-06.

Copyright information

© Ecohealth Journal Consortium 2007

Authors and Affiliations

  • Stephen B. Pyecroft
    • 1
  • Anne-Maree Pearse
    • 1
  • Richmond Loh
    • 1
  • Kate Swift
    • 1
  • Kathy Belov
    • 2
  • Nolan Fox
    • 1
  • Erin Noonan
    • 1
  • Dane Hayes
    • 1
  • Alex Hyatt
    • 3
  • Lingfa Wang
    • 3
  • David Boyle
    • 3
  • Jeff Church
    • 4
  • Debra Middleton
    • 3
  • Robert Moore
    • 3
  1. 1.Department of Primary Industries and WaterAnimal Health Laboratories, Diagnostic Services BranchKings MeadowsAustralia
  2. 2.Centre for Advanced Technologies in Animal Genetics and Reproduction, Faculty of Veterinary ScienceThe University of SydneyNew South WalesAustralia
  3. 3.Australian Animal Health LaboratoryCSIRO Livestock IndustriesGeelongAustralia
  4. 4.CSIRO Textile and Fibre TechnologyClaytonAustralia