Tasman-PCR: a genetic diagnostic assay for Tasmanian devil facial tumour diseases

Tasmanian devils have spawned two transmissible cancer clones, known as devil facial tumour 1 (DFT1) and devil facial tumour 2 (DFT2). DFT1 and DFT2 are transmitted between animals by the transfer of allogeneic contagious cancer cells by biting, and both cause facial tumours. DFT1 and DFT2 tumours are grossly indistinguishable, but can be differentiated using histopathology, cytogenetics or genotyping of polymorphic markers. However, standard diagnostic methods require specialist skills and equipment and entail long processing times. Here, we describe Tasman-PCR: a simple polymerase chain reaction (PCR)-based diagnostic assay that identifies and distinguishes DFT1 and DFT2 by amplification of DNA spanning tumour-specific interchromosomal translocations. We demonstrate the high sensitivity and specificity of this assay by testing DNA from 546 tumours and 804 normal devils. A temporal–spatial screen confirmed the reported geographic ranges of DFT1 and DFT2 and did not provide evidence of additional DFT clones. DFT2 affects disproportionately more males than females, and devils can be co-infected with DFT1 and DFT2. Overall, we present a PCR-based assay that delivers rapid, accurate and high-throughput diagnosis of DFT1 and DFT2. This tool provides an additional resource for devil disease management and may assist with ongoing conservation efforts.


Introduction
Tasmanian devils (Sarcophilus harrisii) are marsupial carnivores endemic to the Australian island of Tasmania. This species is affected by two transmissible cancers, known as Tasmanian devil facial tumour 1 (DFT1) and Tasmanian devil facial tumour 2 (DFT2). DFT1 and DFT2 manifest as facial tumours collectively known as devil facial tumour disease (DFTD) and are transmitted between animals by the direct transfer of living allogeneic cancer cells, probably by biting [1,2]. Thus, DFT1 and DFT2 are two independent malignant clones, one of which arose from the somatic cells of an individual female devil (DFT1), the other from a male (DFT2), and which both now propagate through the devil population via transfer of contagious cancer cells [2 -4]. DFT1 was initially reported in 1996 in northeast Tasmania and is now widespread across the island, causing significant declines in devil populations [5][6][7]. DFT2, on the other hand, which was first observed in 2014, has been reported in only five animals, all located in the Channel, an approximately 550 km 2 peninsula in Tasmania's southeast [2]. DFTD tumours caused by DFT1 and DFT2 are grossly indistinguishable, but are histologically, cytogenetically and genetically distinct [2,8]. Differential diagnosis of DFT1 and DFT2 is routinely performed with histopathology and immunohistochemistry [2,9 -11]. Diagnosis can also be confirmed using cytogenetics, as DFT1 and DFT2 have markedly different karyotypes [1,2]. Genetic methods, based on microsatellite and major histocompatibility complex (MHC) genotypes, have also been used to distinguish DFT1 and DFT2 [2]. However, the presence of host tissue within DFT1 and DFT2 tumour biopsies, as well as allelic variation across tumour and host populations, often leads to ambiguity with the interpretation of such genotypes.
DFT1 and DFT2 are foreign tissue grafts in their hosts and must escape the allogeneic immune system. DFT1 cells down-regulate MHC class I, perhaps contributing to this clone's low immunogenicity [12]; the mechanism of DFT2 immune avoidance is unknown. Interestingly, however, whereas DFT1 equally affects males and females [9], all five devils diagnosed with DFT2 have been male [2]. DFT1 was first detected in the Channel Peninsula in 2012, shortly before the discovery of DFT2 there in 2014 [2]. Thus, DFT1 and DFT2 have overlapping host ranges within the Channel Peninsula, and in the period from December 2012 to June 2015, seven cases of DFT1 and five of DFT2 were reported in the area [2]. While it is known that a devil can carry two strains, or subclones, of DFT1, derived from different infections [4], it remains unknown whether a host can possess both DFT1 and DFT2 tumours simultaneously.
Transmissible cancers have been rarely observed in nature. There are currently eight known naturally occurring transmissible cancers, five of which affect various species of bivalve molluscs [2,13 -15]. In 1996, when DFT1 was first observed, only one other example was known, a sexually transmitted venereal tumour in dogs [15]. Furthermore, there was no evidence of devil tumours similar to DFTD prior to 1996 [5,9]. Thus, the discovery of DFT2 was surprising, and suggests that Tasmanian devils may be particularly vulnerable to the emergence of transmissible cancers [8]. There is a possibility that additional DFT clones may be present or may have previously existed in the devil population, but escaped detection. Moreover, the extent of the DFT2 range beyond the Channel Peninsula is not yet confirmed.
The identification of two transmissible cancers within an interval of 18 years raises important questions about Tasmanian devils and their susceptibility to transmissible cancers. Here, we describe the design and validation of a genetic diagnostic test for confirmation of DFT1 and DFT2 in devils. We use this 'Tasman-PCR' test to examine the range of DFT2 and to confirm the absence of additional DFT clones within a cohort of 546 tumours collected between 2003 and 2016 from more than 69 locations around Tasmania. DFT1 and DFT2 are both clonal lineages, hence somatic variants that were acquired early during each lineage's history may present useful markers for polymerase chain reaction (PCR)-based diagnostic assays. Interchromosomal translocations are particularly effective markers, as the presence or absence of an amplification product is the screening endpoint. We assessed a set of interchromosomal structural variants that had been identified and validated in two DFT1 and in two DFT2 genomes, respectively, but not found in 34 normal devil genomes [8]. Of these, we selected two putative DFT1-specific markers (DFT1-A and DFT1-B) and two putative DFT2-specific markers (DFT2-A and DFT2-B) for further screening (table 1). A triplex PCR assay, named Tasman-PCR, was then optimized, containing primers specific for DFT1-A (231 base pairs, bp), DFT2-A (321 bp), as well as positive control primers amplifying the endogenous RPL13A locus (520 bp) [12] (figure 1).

Normal devil screen
Interchromosomal translocations are more common in cancer than in germline DNA. However, it is possible that DFT1-A and/or DFT2-A were in fact inherited polymorphisms present in the constitutive genomes of the individual devils that spawned the respective clones, and hence may also be found in the genomes of some normal devils in the population. In order to assess this, we screened Tasman-PCR across normal DNA from 804 devils sampled between 2003 and 2016 from more than 65 locations in Tasmania, including both non-diseased individuals as well as devils infected with DFT1 and/or DFT2 (table 2; electronic supplementary material, table S1). DNA from three devils (0.4%) was faintly positive for DFT1-A, and no devils amplified DFT2-A (table 2; electronic supplementary material, table S1). We further assessed the three DFT1-A-positive DNA samples by amplifying a panel of microsatellite length polymorphisms. All three samples amplified three or more alleles at one or more loci, suggesting DNA contamination (electronic supplementary material, table S2A). Furthermore, all three DNA samples which were positive for DFT1-A also amplified DFT1-B (table 1), further suggesting the possibility that these normal samples may have been contaminated by DFT1 DNA. Altogether, these results suggest that the tumour DNA markers used in our triplex assay are highly specific for DFT1 and DFT2, and are likely to be somatic translocations that arose subsequent to neoplastic transformation in these two clones.

Analysis of confirmed DFTD tumours
We used the triplex assay to screen 344 confirmed DFTD (DFT1 and DFT2) tumours (table 2;  electronic supplementary material, table S3A). The tumours were collected between 2003 and 2016 from 56 locations in Tasmania, and DFTD status was confirmed through histopathology, cytogenetics and/or microsatellite genotyping [9,10,16]. In total, 331 of 344 (96.2%) confirmed DFTD tumours were positive for either DFT1-A or DFT2-A (321 DFT1-A positive, 10 DFT2-A positive), and no tumours were positive for both DFT1-A and DFT2-A (table 2; electronic  supplementary material, table S3A). We considered the following three explanations for the 13 remaining confirmed DFT1 tumours which did not amplify DFT1-A (table 2; electronic supplementary material, table S3A): (i) they belong to a DFT1 subclone which diverged from a clonal ancestor prior to DFT1-A being acquired; (ii) they belong to a DFT1 subclone which lost DFT1-A, or acquired mutations within the DFT1-A primer binding sites or (iii) the biopsy used for DNA extraction included only non-neoplastic host tissue, or included DFTD cells at levels that were undetectable under our PCR and gel electrophoresis conditions. To distinguish between these three possibilities, we first screened the 13 tumours with DFT1-B and DFT2-B (table 1); none of these samples amplified either marker. Next, we screened the 13 samples, together with their matched hosts, across a panel of four polymorphic microsatellites (electronic supplementary material, table S2B). None of the 13 samples showed the characteristic microsatellite profiles of DFT1 or DFT2 [2], suggesting that explanations (i) and (ii) above are unlikely to apply. Furthermore, all 12 of these samples for which matched host tissue was available showed microsatellite profiles which were identical to those of their hosts (electronic supplementary material, table S2B). Thus, it appears likely that the biopsies used for DNA extraction carried only host DNA, or that DFTD DNA was present at a low level and was undetectable under our assay conditions (explanation (iii) above). To further assess this possibility, we performed an additional DNA extraction on one of these tumours and confirmed DFT1-A positivity on this second attempt. Thus, variation in the contribution of neoplastic cells to tumour tissues should be considered in the event of an unexpected negative finding.

Tumour screen
We used Tasman-PCR to screen DNA from 175 suspected DFTD tumours collected from 114 devils (table 2; electronic supplementary material, table S3B). In these cases, DFTD was suspected due to gross morphology and clinical presentation, but no records of confirmed diagnosis based on  ). This suggests that these samples may be DFTD tumours from which only non-neoplastic host tissue was sampled for DNA extraction (explanation (iii) above); alternatively, it is possible that these samples may be non-DFTD host-derived lesions. We next used Tasman-PCR to screen DNA from 27 confirmed or suspected non-DFTD lesions. These lesions included cutaneous lymphomas, papillomas, squamous cell carcinomas and other carcinomas (table 2; electronic supplementary material table, S3C). None of the lesions amplified either DFT1-A or DFT2-A. Microsatellite alleles were analysed in the 27 non-DFTD samples, together with their matched hosts when available. Several samples had one or two microsatellite alleles which were not detectable within the tissues of the matched host (electronic supplementary material, table S2D). However, non-DFTD lesions from different individuals had different genotypes. This suggests that the tumour-unique microsatellite alleles arose via somatic mutation of matched host alleles and does not support the possibility that any of these 27 samples represent additional transmissible cancers.

Distribution of DFT1 and DFT2
Of the 494 confirmed or suspected DFTD tumours which amplified a DFT1 or DFT2 tumour marker in Tasman

DFT1 and DFT2 host gender
Our screen detected 11 animals with DFT2 tumours, including the five previously reported cases [2]. Of these, nine (82%), were male (Fisher's exact test for no difference in DFT2 host gender, p ¼ 0.033) (figure 3). By contrast, DFT1 tumours were equally distributed between male and female hosts (of the 345 animals with one or more DFT1 tumours with known gender, 165 (47.8%) were male and 180 (52.2%) were female (Fisher's exact test for no difference in DFT1 host gender, p ¼ 0.451), as previously reported [9] (figure 3). This skew towards male hosts in DFT2 suggests that this clone may preferentially affect males, although larger numbers of DFT2 cases will be required to confirm this.

Co-infection with DFT1 and DFT2
It has previously been noted that individual devils can simultaneously harbour tumours belonging to two distinct strains, or subclones, of DFT1 [4]. As the DFT1 and DFT2 ranges overlap in the Channel Peninsula, we investigated the possibility of co-infection with DFT1 and DFT2. We identified two individuals, Devil 812 and Devil 818, which carried both DFT1 and DFT2 tumours ( figure 4). Devil 812 was a 4-year-old female with three facial tumours, denoted 812T1, 812T2 and 812T3. 812T2 and 812T3 were both DFT1, whereas 812T1 was DFT2 ( figure 4). Devil 818, a 2-year-old male, carried a large DFT2 tumour, 818T1, ventral to the left ear, as well as a DFT1 tumour on the hard palate, 818T2 (figure 4). Devil 818 also had a DFT2 mass in the left pre-auricular lymph node (818T3) (figure 4), probably a metastasis from 818T1.

Discussion
We present a PCR-based genetic diagnostic test for DFTD, Tasman-PCR, which complements existing diagnostic methods to provide rapid and cost-effective confirmation of DFT1 and DFT2. The test uses amplification of tumour-specific interchromosomal translocations to identify DFT1 and DFT2, and to distinguish between DFTD (DFT1 or DFT2) and non-DFTD tumours in devils. We used the assay to screen 546 devil tumours collected from more than 69 locations in Tasmania between 2003 and 2016.
Although the assay that we describe provides a useful method for diagnosis of DFT1 and DFT2, it has some limitations. Users should be aware of the potential for DNA cross-contamination to produce While very little input material is required for PCR, which allows for minimally invasive small biopsies, it can also increase the likelihood of obtaining biopsies that contain no neoplastic cells, leading to false negative results. Moreover, although we did not find evidence of germline DFT1-A or DFT2-A in a screen of 804 normal devils, we cannot exclude the possibility that these markers are rare polymorphisms in devil populations; we suggest that users analyse matched normal tissues alongside tumours to control for this possibility. Additionally, although our screen did not identify any clear cases of DFT1 or DFT2 tumours negative for DFT1-A or DFT2-A, respectively, it is possible that such lineages may have previously arisen or may occur in future. Users encountering unexpected negative results should perform PCR for additional markers, such as DFT1-B and DFT2-B (table 1) as well as repeating DNA extraction and PCR from a separate biopsy of the same tumour. Importantly, this assay is designed to be complementary with other established DFT1 and DFT2 diagnostic methods, such as histology and cytogenetics, while being rapid, scalable and minimally invasive. Furthermore, it would be possible to develop this method to obtain a real-time diagnosis in the field with the use of portable PCR laboratories [17,18]. The discovery of two transmissible cancers in Tasmanian devils raises the possibility that additional, so far undetected, transmissible cancer clones exist within devil populations. In this study, we screened 546 Tasmanian devil tumours, collected between 2003 and 2016 from more than 69 locations in Tasmania (table 3). In total, 479 (87.7%) were DFT1 and 15 (2.7%) were DFT2; the remaining 52 (9.5%) were either confirmed or suspected host-derived non-DFTD lesions (27, 4.9%) or biopsies that shared identical microsatellite profiles with their matched hosts (as well as four biopsies for which matched host tissue was unavailable or microsatellites could not be amplified) (25, 4.5%) (table 3). Thus, this screen has not detected evidence for additional transmissible cancers in Tasmanian devils. However, the method provides a useful tool for devil tumour monitoring and surveillance.
DFT1 had been detected in the Channel Peninsula prior to the discovery there of DFT2, and previous studies have shown that different DFT1 strains, or subclones, can co-infect one host [4]. Here, we confirmed that DFT1 and DFT2 can also co-infect the same devil. Co-infections of these two phenotypically similar DFTs highlight the importance of careful monitoring, as competition for a limited host devil population may influence the evolution of virulence [19].  T1  T2   T1  T2 T3  T4  H  D1 D2  L   T3   T2   T1   T1  T2  T3  H  Of the 11 DFT2 cases that we detected, nine involved a male host (Fisher's exact test for no difference in DFT2 host gender, p ¼ 0.033). This apparent preference for male hosts may reflect underlying transmission dynamics. Alternatively, it is possible that females are less susceptible to DFT2, perhaps due to the recognition of Y chromosome-derived allogeneic antigens [8]. Nevertheless, we detected two females with DFT2 tumours, confirming that this disease does not exclusively affect males. The gender bias observed in DFT2 may impact devil population structures by reducing overall male devil abundance; over time, this may decrease reproductive success and increase inbreeding by limiting mating opportunities. Furthermore, if DFT2 is largely excluded from female hosts, this may affect the future potential for this clone to survive, perhaps selecting for the emergence of DFT2 subclones with balanced host gender proclivity through Y chromosome loss [8].
Overall, this study presents Tasman-PCR, a PCR-based assay for rapid, high-throughput diagnosis of DFT1 and DFT2 in Tasmanian devils. By screening 546 tumours, we have documented the distributions of DFT1 and DFT2, and identify two devils co-infected with both clones. Our analysis does not provide evidence for additional DFTD clones beyond DFT1 and DFT2. The co-occurrence of two transmissible cancer clones in Tasmanian devils presents unique challenges for the conservation of this iconic species. The genetic diagnostic screening tools described here provide important additional resources for population monitoring and disease surveillance.

Sample collection and DNA extraction
Biopsies were collected from wild or captive Tasmanian devils or from animals whose carcasses were found as roadkill. Tumour and host (ear, liver, spleen, blood, gonad or submandibular lymph node) tissue biopsies were collected into either RNAlater or ethanol. Fine needle aspirates collected into a 1 : 1 solution of Buffer AL (Qiagen) : PBS were sampled from tumours that were too small to biopsy. A subset of tumours were previously confirmed as DFT1 or DFT2 using histopathology, cytogenetics or microsatellite genotyping [9,10,16]. Genomic DNA was extracted using the Qiagen DNeasy Blood and Tissue Kit. Genomic DNA was amplified using the Illustra GenomiPhi V2 DNA Amplification Kit (GE Healthcare) prior to PCR amplification.

Selection of markers
Two markers for DFT1 (DFT1-A and DFT1-B) and two markers for DFT2 (DFT2-A and DFT2-B) were evaluated (table 1). All four markers amplified products that were DFT1/DFT2specific in a preliminary screen. DFT1-A and DFT2-A were selected for development of the triplex assay; DFT1-B and DFT2-B can be used to provide additional diagnostic confirmation if required. Single (non-triplex) PCRs, including those involving DFT1-B and DFT2-B, were performed with the same conditions, with primers at a final concentration 0.4 mM. Gel electrophoresis was performed with a 3% agarose gel and products were detected using ethidium bromide. Samples with weak amplification bands were identified, and PCR was repeated with increased DNA input.

Microsatellite genotyping
Microsatellite amplification and fragment size detection was performed across four loci (C, D, N and F) as previously described [2] and analysed using Fragman [20]. and drafted the manuscript. Y.M.K. and M.R.S. prepared figures. All authors reviewed and gave final approval to the manuscript.
Competing interests. We declare we have no competing interests. Funding. This work was supported by grants from the Wellcome Trust (102942/Z/13/A), the National Science Foundation (DEB-1316549), the Australian Research Council (DE170101116), the University of Tasmania Foundation (Eric Guiler Tasmanian Devil Research Grants) and a Philip Leverhulme Prize awarded by the Leverhulme Trust. Y.M.K. is supported by a Herchel Smith Postgraduate Fellowship and M.R.S. is supported by a scholarship from the Gates Cambridge Trust.