Detection and analysis of enzootic nasal tumor virus 2 in China

Abstract

Enzootic nasal tumor virus 2 (ENTV2), the etiologic agent of enzootic nasal adenocarcinoma (ENA) in goats, is highly prevalent in China and causes significant economic losses to the goat industry. Here we describe the occurrence of ENA on a Dazu black goat farm in Chongqing City. At autopsy, nasal cavity masses were observed within the nose of an affected goat; histologically, the tumor was a nasal adenocarcinoma. The qPCR results demonstrated unequivocally that ENTV2 was the primary pathogen responsible for the tumor in this goat. We also collected nasal swab samples from all 180 goats on the farm; 9 goats tested positive for ENTV2. We generated the sequence of the full-length genome of ENTV2 (named ENTV2CQ, GenBank OR024676) with 7,469 nucleotides from nasal tumors from our case. ENTV2CQ shared the highest nucleotide identity with a previously sequenced isolate, ENTV2FJ (GenBank MK559457.1). ENTV2CQ and ENTV2FJ are located in the same major phylogenetic branch, mainly related to isolates from China from 2015 to 2022, and their phylogeny may be clustered geographically.

Enzootic nasal tumor virus (ENTV; Retroviridae, unclassified Betaretrovirus) ¹⁰ and jaagsiekte sheep retrovirus (JSRV; Retroviridae, Betaretrovirus ovijaa) ²³ are associated with neoplastic transformation of secretory epithelial cells in the respiratory system. ENTV is the causative agent of enzootic nasal adenocarcinoma (ENA), with ENTV1 occurring in sheep, and ENTV2 in goats; both viruses are closely related to JSRV, which causes ovine pulmonary adenocarcinoma (OPA). ²¹ Despite being very similar, ENTV and JSRV are different viruses and cause disease in different anatomic locations of the respiratory tract. ENTV induces the transformation of epithelial cells of the ethmoid turbinates in sheep and goats^6,10,18; JSRV transforms epithelial cells of the distal lung, namely alveolar epithelial type II cells in the alveoli and club cells in the bronchioles.^{3,11,20,21,35} We retrieved no cases of coinfection between ENTV1 and ENTV2, or between ENTV2 and JSRV, in a search of Google, PubMed, CAB Direct, Web of Science, and Scopus, using the search term “ENTV2 co-infection”, suggesting that this condition has not been reported in goats and sheep. Interestingly, coinfection by ENTV1 and JSRV has been reported in naturally infected sheep. ¹⁹

Like JSRV, ENTV1 and ENTV2 are single positive-stranded RNA viruses, and they have characteristics of both D- and B-type viruses. These 3 retroviruses share the common structural basis of all retroviruses, which consists of 5′-LTR-gag-pro-pol-env-3′-LTR (long terminal repeat), without any additional viral oncogenes^6,18; env acts as the viral oncogene. ¹² By analogy with JSRV, the viral envelope is likely to be responsible for cell transformation, ¹⁷ although this has not been proved formally for ENTV2.

ENA occurs naturally worldwide and has been reported in many countries, except for Australia and New Zealand ³⁰ ; the number of reports in China has gradually increased. In China, the first report of ENA in goats was in Inner Mongolia in 1995, and thereafter, the disease spread rapidly to many regions of China, including East, Central, South, Southwestern, and Northwestern China. ³⁶ ENTV induces a prolonged latent infection, and clinical signs may take some time to be observed. ⁷ More significantly, the difficulty in distinguishing between animals with a latent infection and those that are healthy provides an opportunity for the virus to spread through a herd. The diagnosis is usually determined through histologic examination of the tumor, coupled with confirmation of the viral pathogen via molecular techniques.

Here we report a case of ENTV2 infection from Chongqing, China, with distinct genome characteristics. We also analyzed the genetic evolution of this virus to trace infection sources and disease development. Our analysis will be helpful for understanding the genetic diversity of ENTV2 in China and opens the way to further studies to elucidate the epidemiology and oncogenic mechanisms of this virus.

Materials and methods

Animals and clinical samples

In October 2022, several goats developed signs of respiratory disease at a farm in Youyang County, Chongqing City, China, that housed 180 Dazu black goats. The affected goats initially had copious serous nasal discharge (Fig. 1A), followed by dyspnea and anorexia with severe weight loss. To identify potential infectious agents that could be the cause of the disease, nasal swab samples were collected from 180 goats. After clinical examination, an autopsy was performed on a severely affected goat. Tumor samples were collected and fixed in 10% neutral-buffered formalin; lung, kidney, spleen, liver, and blood samples were also collected for subsequent research.

Figure 1.

Gross photograph of enzootic nasal adenocarcinoma–affected goats. A. Copious nasal discharge. B. Unilateral mass (arrow) in the caudal nasal cavity. C. The tumors removed from the nasal cavity of the affected goat are soft and gelatinous.

Histologic examination and RNA in situ hybridization

The formalin-fixed tissues were processed routinely, and sections were stained with H&E and examined with a light microscope. We also performed RNAscope in situ hybridization (ISH) on formalin-fixed, paraffin-embedded (FFPE) tissues according to a method described previously ³² by designing a probe targeting the env gene.

Nucleic acid extraction and qPCR detection

DNA/RNA extraction from samples was performed (TIANamp genomic DNA/RNA kit; Tiangen Biotech) following the manufacturer’s protocol. Following initial presumptive diagnosis, we used the following primers to detect ENTV2 in collected samples, as described previously ² : (ENTV2up: 5′-CCTAACCTTCATTCRTTATGGCARAGT-3′; ENTV2do: 5′-CACCGGATCCTTAYGTAATCRGATTTCCTG-3′) and probe (ENTVpr: FAM-TGTTTAGTTCCTTGCCTCCTTCGTGG-IBFQ). In addition, we determined the potential viral concentrations in the tumor, lung, kidney, spleen, liver, and blood using the same method. Briefly, total RNA was extracted from 200 μL of blood or 100 mg of tissue samples (TRIzol; Life Technologies) according to the manufacturer’s instructions. All nucleic acids were stored at −80°C until tested. One sequenced amplicon from a China ENTV2 isolate was cloned (pEASY-T1 cloning vector; TransGen Biotech). The recombinant plasmid was linearized, and then transcribed in vitro. The concentration of an RNA standard was measured with a spectrophotometer, and the copy numbers of the RNA standard were calculated as described previously. ³³ The in vitro transcribed RNA was 10-fold serial diluted and used to establish the standard curve. Each dilution was tested in quadruplicate. Subsequently, viral RNA was used as template for the analysis of the reverse-transcription quantitative real-time PCR (RT-qPCR) assay. All samples and controls were run in triplicate. Nucleic acids isolated from tumor tissue from an affected goat and nasal turbinate epithelial tissue of a healthy goat were used as positive and negative controls, respectively.

Samples from the 180 nasal swabs were investigated, using methods described previously, to rule out other common goat respiratory pathogens, namely, caprine parainfluenza virus 3 (CPIV3), peste-des-petits-ruminants virus (PPRV), Pasteurella multocida (Pm), Mycoplasma capricolum subsp. capricolum (Mcc), Mycoplasma capricolum subsp. capripneumoniae (Mccp), and Mycoplasma ovipneumoniae (Mo).^1,13,25,37 We used the positive nucleic acids of these pathogens stored in our laboratory as positive controls and RNase-free water as the negative control.

To obtain the full-length sequence of ENTV2, we designed and synthesized 6 pairs of primers (Table 1) based on ENTV2 sequences available in GenBank (MK559457.1). The specificity of the primers was validated in NCBI. Viral RNA extracted from the tumors was used as a template for reverse-transcription PCR (RT-PCR) to amplify overlapping segments to determine the complete viral genome of ENTV2 in our case. RNA isolated from ethmoidal epithelial tissue of the nasal passage of a healthy goat was used as a negative control to confirm that the primers did not amplify endogenous goat sequences. One-step RT-PCR was performed (Evo M-MLV One Step RT-PCR kit; Accurate Biology); the reagents and cycle conditions were the same for every PCR: 25 μL of 2× One-Step reaction solution A, 2 μL of One-Step enzyme mix, 1 μL of each primer (0.4 μM each), 1 μL of template, and 20 μL of RNase-free water. Cycling conditions were as follows: 50°C for 30 min, 94°C for 2 min, 35 cycles of 94°C for 30 s, 56°C for 30 s, and 72°C for 1.5 min; and a final extension of 5 min at 72°C. PCR products were purified (MiniBEST agarose gel DNA extraction kit v.4.0; TaKaRa).

Table 1.

Primers for amplifying the complete genome of enzootic nasal tumor virus 2.

Primer	Primer sequence (5′–3′)	Location	Product length, bp
1F	ACAAGGCATCAGCCATTTTGGTC	1–23	1,131
1R	GTAAATGGAGCGGTAGGACCATATTGTG	1131–1104
2F	ACAACCAGCGTTATTATGAATCACTGCC	1041–1068	1,363
2R	TTGCACCCAATAAGCGTCAGAT	2403–2382
3F	GCGGGACAGCGTATAGCTCAACTCC	2290–2314	1,693
3R	GCCTCATGACGTCCCTTCGCCACAA	3982–3958
4F	GCACCCACAGGGGTTCTCTATCAAG	3855–3879	1,395
4R	TGGTTTTTTGCAGGAGGATCGCT	5249–5227
5F	TAGCCCTTCACCACATAATGCCTTG	5126–5150	1,422
5R	CAGAAGCAGTGACAGCAGTTGCTAT	6547–6523
6F	TAGCCGCAAGGAAAGAAGTGTGAGC	6462–6486	1,008
6R	GCTGATGCCTTGTTGTTTTATTGT	7469–7446

Sequencing and phylogenetic analysis

The 6 purified PCR products were cloned (pMD-18T vector; TaKaRa) according to the manufacturer’s instructions; 1 independent clone of each of the 6 positive recombinant plasmids was Sanger-sequenced (Sangon Biotech). The nucleotide sequences obtained were used for BLAST analysis to confirm that they were derived from ENTV2. The full-length genome was assembled using sequence analysis software (Lasergene v.7.1; DNAstar). Phylogenetic analysis of the nucleotide sequences of ENTV2 was performed by the neighbor-joining method using MEGA 7 (https://www.megasoftware.net/) with 1,000 bootstrap replicates.

Results

Cauliflower-like neoplastic lesions in the right nasal cavity of the affected goat were closely connected with the ethmoid mucosa of the turbinates caudal to the nostrils of the goat (Fig. 1B, 1C). No lesions were observed in the lungs or any other organs. Histologically, the tumor tissue had significant structural heterogeneity; gland-like structures were tightly arranged (Fig. 2). Both serous glands and mucous glands were clearly present within the tumor tissue. Tumor cells were arranged in short columnar or cuboidal structures, with low cellular atypia and no mitotic figures observed. A few lymphocytes infiltrated the interstitium, accompanied by mild congestion in the marginal interstitium (Fig. 2A, 2B). The histologic features were compatible with nasal adenocarcinoma. In RNAscope in situ RNA hybridization, the probe targeting the env gene generated a very strong positive hybridization signal in the tumor. In contrast, there was no staining in the negative control (no target probe; Fig. 3).

Figure 2.

Histopathology of nasal tumor from the infected goat. A. The tumor has structural heterogeneity, appearing glandular and tightly arranged (★), with both serous glands (▲) and mucinous glands (↑) visible. B. The tumor cells are uniform, well-differentiated, short columnar and cuboidal (★). Both serous glands (▲) and mucinous glands (↑) are visible.

Figure 3.

Detection of enzootic nasal tumor virus 2 (ENTV2) RNA via RNAscope in situ hybridization in the tumor from the ENTV2-infected goat; env gene stains red and DAPI stains blue. DAPI = nuclear staining.

Nasal swabs from 9 of 180 goats tested positive by PCR for ENTV2; all samples were negative for CPIV3, PPRV, Pm, Mcc, Mccp, and Mo. Only mild clinical signs were observed in these 9 animals, and they did not have more advanced clinical signs, such as exophthalmos or skull deformities. One of the goats that tested positive for ENTV2 was only 3-mo-old. The nasal tumor, lung, blood, and spleen of the affected goat were positive for ENTV2 by qPCR analysis (Fig. 4). No virus was detected in the liver, kidney, or control group. A high ENTV2 viral load was suggested in the tumor by an individual cycle threshold (Ct) value of 18 compared to Cts of 27 in lung, 32 in blood, and 30 in spleen. Using a 10-fold dilution series to generate a standard curve, the copy number equivalent was estimated as 6.82 × 10⁸, 1.22 × 10⁶, 1.44 × 10⁵, and 1.50 × 10⁴ per 1 g of tumor, lung, and spleen tissue or per 1 mL of blood, respectively (Fig. 4).

Figure 4.

Genomic copies of enzootic nasal tumor virus 2 (ENTV2) in various samples. A. The standard curve was constructed with serial 10-fold dilutions of the RNA standards. B. Genomic copies of ENTV2 per gram of tissue or per mL of blood.

Assembly of the 6 overlapping sequences from the PCR products resulted in a full-length genome sequence of 7,469 nucleotides, which we named ENTV2CQ and deposited in GenBank (Fig. 5; OR024676). ENTV2CQ aligns well with other known ENTV2 sequences, with a structure of 5′-UTR-gag-pro-pol-env-3′-UTR. The genome’s open reading frames (ORFs) for the gag, pro, pol, and env genes comprised 1,839 nt (257–2,095), 927 nt (1,930–2,856), 2,613 nt (2,823–5,435), and 1,869 nt (5,323–7,191), respectively, encoding proteins of 612, 308, 870, and 617 aa. In a BLASTn search against available sequences of ENTV2 from China, the sequence that we obtained had the highest nucleotide identity (99.4%) with a Chinese ENTV2FJ isolate (MK559457.1), and the lowest nucleotide identity (86.4%) with a Chinese GDZJ2022 strain (ON843769.1) isolated in 2022. The LTR of ENTV2CQ is 412 bases long and is flanked by the inverted-repeat sequences GCAG (nt 124–127) and CTGC (nt 7,172–7,175). Notably, ENTV2CQ gene mutations are primarily in the U3 and U5 regions compared to the ENTV2 reference strain (AY197548.2). Significant differences were observed in the LTR region when comparing ENTV2 with ENTV1 (GenBank GU292317.1) and JSRV (AF105220.1). The ENTV2CQ LTR shared the highest nucleotide sequence identity (99.5%) with the LTR of ENTV2FJ (MK559457.1). Compared with the gag gene of other ENTV2 strains, we found that ENTV2CQ had the highest nucleotide sequence identity (99.6%) with Chinese reference strains LC762617.1, LC762616.1, MT254061.1, and MT598195.1.

Figure 5.

The 6 fragments of enzootic nasal tumor virus 2 that were amplified by specific RT-PCR. Lanes: M = molecular weight marker; 1, 3, 5, 7, 9, 11 = the RT-PCR products of each fragment amplified from the sample containing RNA from nasal tumors; 2, 4, 6, 8, 10, 12 = RNA from ethmoidal epithelial tissue of the nasal passage of a healthy goat used as a negative control.

At the amino acid level, the Gag protein sequences of ENTV2CQ had the highest identity (99.4%) with ENTV2FJ (MK559457.1), and there were only 4 aa differences (E 38 V, W 55 R, D 159 G, L 239 S). Overall, the pro region of ENTV2CQ remained stable and had little nucleotide or amino acid variability compared to all reference strains. The pol gene of ENTV2 contained an additional ORF, orf-x. A DNAMAN alignment of the orf-x amino acid sequences indicated that the ENTV2CQ orf-x starts at the same methionine as the JSRV orf-x but truncates at amino acids 43 and 168, unlike the ENTV2 orf-x, which ends at 166 (Fig. 6; AY197548.1). Comparison of the env gene revealed the highest nucleotide sequence identity (99.8%) to Chinese reference strains KU980912.1, KU980910.1, KU179192.1, and MT254061.1. The Env protein sequences of ENTV2CQ had the highest similarity (99.7%) with the Shaanxi isolate KU179192.1, and the difference was S 161 P and T 368 I. Like JSRV and ENTV1, the ENTV2CQ strain has the YXXM motif. Studies of JSRV have shown that the YXXM motif in the cytoplasmic tail of the envelope transmembrane (TM) protein is necessary for transformation.^5,16

Figure 6.

Alignment of the predicted amino acid sequence of orf-x from ENTV2CQ isolates, ENTV2, ENTV1NA1, and JSRV-UK. Dots indicate identity. JSRV21 (top line) was used as the reference sequence.

The gag and env genes had good conformity with the ENTV2 sequences available in GenBank, particularly compared to isolates originating from China. The gag gene had 85.4–99.6% sequence identity with other reference strains of China; compared with isolates from outside China, it had the highest identity of 90.4% with 2 Spanish isolates (NC_004994.2, AY197548.2). For the env gene, ENTV2CQ was >86.4% identical at the nucleotide level with all Chinese strains but had the highest identity (88.6%) with the Spanish NAOM-HU3118124 among foreign strains (LC570918.1).

To comprehend the genetic evolution of ENTV2CQ, we employed neighbor-joining trees for phylogenetic analysis based on the gag and env genes. ENTV2CQ clustered within the ENTV2 branch and had higher homology with Chinese strains predominantly circulating in 2015–2022 than with strains from other countries (Fig. 7). Our strain has a nucleotide sequence identity of 95.0% with the env gene of Chongqing CQ1, which was collected in 2018. In the phylogenetic tree based on the env gene (Fig. 7A), ENTV2CQ is associated with the CQ1 strain (MK164400.1), but the genetic relationship is distant.

Figure 7.

Phylogenetic analysis based on A. the gag gene, and B. the env gene of ENTV2 with ENTV1 and JSRV. ENTV2CQ described in our study is marked with a solid black circle.

Discussion

The nasal tumor in our case was locally invasive; pathoanatomic appearance and histopathologic findings were in accordance with previously described ENA cases in goats, predominantly of low grade. ⁹ ENTV can induce unilateral or bilateral neoplastic growth of the mucosal nasal glands from the ethmoidal area in the nasal cavity, ¹⁰ and ENA is classified as a low-grade adenocarcinoma. The occurrence of unilaterally arising tumors may prevail over bilateral ²⁷ ; information regarding the prevalence of unilateral tumors for the right or left nasal cavity is lacking. In our case, the tumor was unilateral, right-sided.

Our RNAscope ENTV2 assay detected the ENTV2 env mRNA transcripts in situ in FFPE tumor tissues. RNAscope ISH might be a valuable tool for detection of ENTV2 in archival samples and for confirming infection. ⁴ Using qPCR technology, we confirmed the presence of ENTV2 in the affected goats; no other respiratory pathogens were detected. Based on these results, we demonstrated that ENTV2 was the cause of ENA in the goat farm. However, more prominent clinical signs described in other studies (e.g., ocular protrusion, facial swelling) did not develop within the 6 mo during which the farm was monitored.^8,34 Mild clinical signs are likely due to smaller ENA tumors.

ENA has been reported to mainly affect goats >6-mo-old. ²⁸ We detected ENTV2 in a 3-mo-old goat. Combined with a previous study, ³⁶ we speculate that the ENTV2 strain may cause acute infection, similar to that caused by ENTV1. ²⁹ Unfortunately, the owner reported that the dam died shortly after the kid was born, with persistent seromucous nasal discharge; an autopsy was not performed. Goats are capable of carrying the virus and infecting other animals via the airborne route as well as via nasal secretions. Usually, due to the slow course of the disease, most clinical cases appear in adult goats. Our report of infection in young goats supports the possibility of vertical transmission. In addition, the impact of the virus on kids and even fetuses should be assessed.

By PCR, we generated a full-length sequence that corresponded to published ENTV2 genomes, and compared the genome with other ENTV2 isolates, as well as ENTV1 and JSRV. This is crucial for understanding the genetic heterogeneity of ENTV2. The LTR of ENTV2CQ was similar to ENTV2 (AY197548.2). The U3 region of LTR contains cis-acting sequences necessary for viral replication and regulatory signals for retroviral transcription. Thus, an understanding of transcriptional control of ENTV may also contribute to elucidating the mechanism of viral oncogenesis. However, the role of the ENTV2 LTR in pathogenesis has yet to be determined. We found most of the amino acid differences in the LTR; tissue tropism appears to be the same as for other ENTV2 strains even though researchers have shown that JSRV and ENTV tissue tropism is orchestrated by the combined effort of the envelope and the U3 promoter. ²⁴ Hence, these mutations do not affect the tissue tropism of ENTV2CQ.

Orf-x is the most genetically diverse protein coding sequence of ENTV2. On the other hand, ENTV2CQ orf-x is truncated by 2 stop codons at residues 43 and 168. This is similar to ENTV1. The 2 stop codons in the orf-x ORF are characteristic of ENTV and differentiate it from the JSRV orf-x, which does not have this stop codon. ²⁶ The stop codons in orf-x make it unlikely that the orf-x protein is important for ENTV2 replication. Reports suggest that a complete orf-x is not required in the pathogenesis of JSRV. ⁵ The large genetic changes of orf-x in ENTV2CQ may indicate that orf-x does not play a significant role in the pathogenesis of ENA. Despite the apparent dispensable nature of orf-x, it is conserved in all exogenous JSRV isolates, and the orf-x transcript is produced in OPA lung tumors. ²² It is not known whether the orf-x transcript is produced from ENTV2-infected cells.

The cytoplasmic tail (CT) of the envelope TM protein is necessary for transformation by JSRV, because the YXXM in the JSRV CT binds to PI3K leading to the activation of downstream signaling pathways in cellular transformation. ¹⁶ Mutations in the YXXM motif can eliminate or reduce cell transformation.^14,15 The YXXM motif is consistently present in the env genes of all currently discovered ENTV2 strains, including the ENTV2CQ in our study. The YXXM motif is a reliable molecular marker for the infectious exogenous virus. However, further research is needed to determine whether transformation by ENTV2, as by JSRV, requires a conserved YXXM motif. An infectious molecular clone has been reported for ENTV1 ³¹ but not yet for ENTV2, which would help determine the pathogenicity of the virus in goats.

ENTV2CQ clusters within the ENTV2 branch, exhibiting a higher degree of homology with ENTV2 strains prevalent in China than with those from other countries. ENTV2CQ is associated with the 2018 CQ1 strain (MK164400.1), albeit with a distant homology. Notably, most of the strains from China were prevalent in 2015–2022. The isolation from different continents probably leads to the relatively large difference between domestic and foreign sequences. The ENTV2CQ strain is in the same major branch as the strains collected from Shaanxi, Guangxi, Fujian, Sichuan, and Anhui provinces in China. The geographic proximity of these provinces, and the rapid expansion of the logistics industry, which has facilitated the increasingly frequent movement of live goats and mutton products, may have accelerated the spread of ENTV2 among them.