PCR-based clonality analysis of antigen receptor gene rearrangements in canine cutaneous plasmacytoma

The Veterinary Journal 241 (2018) 31–37

Contents lists available at ScienceDirect

The Veterinary Journal journal homepage: www.elsevier.com/locate/tvjl

PCR-based clonality analysis of antigen receptor gene rearrangements in canine cutaneous plasmacytoma M. Takanosua,* , K. Okadab , Y. Kagawab a b

Nasunogahara Animal Clinic, 2-3574-98, Asaka, Ohtawara, Tochigi 324-0043, Japan North Lab, 2-8-35, Hondori, Shiroisi-ku, Sapporo, Hokkaido 003-0027, Japan

A R T I C L E I N F O

A B S T R A C T

Article history: Accepted 18 September 2018

Plasmacytomas are discrete, B cell-derived, round cell tumours that sometimes are difficult to distinguish from canine cutaneous histiocytomas or T cell lymphosarcomas (lymphomas). Diagnosis of plasmacytomas relies on morphological observations and immunohistochemistry for multiple myeloma oncogene-1 (MUM-1) and cluster of differentiation 3 (CD3). Clonality testing often is used as an adjunct diagnostic tool to examine lymphoproliferative diseases. In this study, the sensitivity of PCR-based clonality analysis of antigen receptor gene rearrangements in canine cutaneous plasmacytomas was determined. Formalin-fixed paraffin-embedded sections of 29 canine plasmacytomas, 23 diffuse large B cell lymphomas (DLBCLs) and 23 lymph nodes without lymphoma were used for clonality analysis. New oligonucleotide primers for the framework (FR)2 and FR3 regions of the immunoglobulin heavy chain (IGH) V gene subgroup 3 were designed and used with previously reported FR3 primers. Although plasma cells are of B cell lineage, the detected frequency of IGH clonality in plasmacytoma was 0–34.5% with the seven primers used, whereas in DLBCLs it was 8.7–78.3%. In 23 lymph nodes without lymphoma, IGH clonality was detected in only one case with two out of the seven primers used. Sequence analysis of PCR products from plasmacytomas revealed mismatches in the annealing region of the FR3 primers. The sensitivity of detecting IGH clonality in canine plasmacytomas was lower than in DLBCLs. The low detection rate of IGH clonality in canine plasmacytoma may be due to somatic hypermutation of the variable region. © 2018 Elsevier Ltd. All rights reserved.

Keywords: Canine Clonality Immunoglobulin heavy chain Plasmacytoma Variable region

Introduction Plasmacytomas are B cell-derived tumours that occur at various sites in dogs, including the skin, internal organs and bones (Gupta et al., 2014; Mikiewicz et al., 2016). Diagnosis of plasmacytomas primarily involves histomorphology and immunohistochemistry for the multiple myeloma oncogene-1 (MUM-1) protein (RamosVara et al., 2007). In the skin, plasmacytomas are manifested as discrete round cell tumours and may be misdiagnosed as canine cutaneous histiocytoma, epitheliotropic T cell lymphosarcoma (lymphoma) or other entities that are similar in appearance (Pa zdzior-Czapula et al., 2015). Clonality testing can be used as a tool to provide evidence in support of a diagnosis of extramedullary plasmacytoma (Velez et al., 2007; Van Wettere et al., 2009; Rout et al., 2017). Clonality testing also is used as an adjunct tool for examining lymphoproliferative diseases (Keller et al., 2016). The sensitivity of clonality

* Corresponding author. E-mail address: [email protected] (M. Takanosu). https://doi.org/10.1016/j.tvjl.2018.09.010 1090-0233/© 2018 Elsevier Ltd. All rights reserved.

testing in neoplastic lymphocytes has been reported as 67–91% (Burnett et al., 2003; Thalheim et al., 2013; Waugh et al., 2016). Previous reports have described clonality testing based on the immunoglobulin heavy chain (IGH) gene in canine and feline plasmacytomas (Burnett et al., 2003; Werner et al., 2005; Waugh et al., 2016). However, the sensitivity of clonality testing for plasmacytoma has not been examined in a large series of cases. To use clonality testing as an adjunct tool for plasmacytoma diagnosis, it is important to investigate the sensitivity for this tumour type. This study was designed to examine whether clonality testing is useful for diagnosing canine plasmacytomas in addition to lymphomas. Materials and methods Specimens used for clonality analysis Clonality analysis was performed using genomic DNA samples extracted from 29 canine cutaneous plasmacytomas, 23 diffuse large B cell lymphomas (DLBCLs) arising in peripheral lymph nodes and 23 lymph nodes without lymphoma. Tissues were submitted by veterinary animal clinics to a private veterinary pathology diagnostic centre (North Lab, Sapporo, Japan). Out of an

32

M. Takanosu et al. / The Veterinary Journal 241 (2018) 31–37

Table 1 Oligonucleotide primers used for PCR in this study. Primer name

Direction

Target

Primer sequence (50 -30 )

Reference

CB-1 CB-2 CB-3 Tamura-V Tamura-J FR2-6 FR3-1 Sigmf1 Srm3 ATP7B-F ATP7B-R SOD1-F SOD1-R

Forward Reverse Reverse Forward Reverse Forward Forward Forward Reverse Forward Reverse Forward Reverse

IGHV3 (FR3) IGHJ1-5 IGHJ6 IGHV3 (FR3) IGHJ1-5 IGHV3 (FR2) IGHV3 (FR3) IGH constant region IGH constant region ATP7B ATP7B SOD1 SOD1

CAGCCTGAGAGCCGAGGACAC TGAGGAGACGGTGACCAGGGT TGAGGACACAAAGAGTGAGG ACACGGCCVTGTATTACTGT TGAGGAGACGGTGACC AAGGGGCTGCAGTKGGTCGCA GATTCACCATCTCCAGAGACA TTCCCCCTCATCACCTGTGA GGTTGTTGATTGCACTGAGG CTCGCAGCTATAAGAAGCCCGA CTTGTCGGACTTCAGGGAGGAC GCTTTTCTTTGACTGAAGGGAAG TGGATCATTTCCCTAAGGCTG

Burnett et al. (2003) Burnett et al. (2003) Burnett et al. (2003) Tamura et al. (2006) Tamura et al. (2006) Present study Present study Burnett et al. (2003) Burnett et al. (2003) Present study Present study Present study Present study

FR1

FR2

FR3

CDR

CB1 TamuraV FR3-1

FR4 CB2/CB3 TamuraJ

FR2-6

CB1/CB2, TamuraV/TamuraJ

80~120 bp

FR3-1/CB2

150~170 bp

FR2-6/CB2

220~240 bp

Fig. 1. Schematic representation of the IGH gene and primers used in this study. Each primer set and PCR product size is shown in the diagram. CB1 and TamuraV are located downstream of FR3. CB2/CB3 and TamuraJ are located within FR4. The novel primers FR2-6 and FR3-1 were designed specifically for this study and target FR2 and FR3, respectively. IGH, immunoglobulin heavy chain; FR, framework.

initial set of 30 tumours diagnosed as plasmacytomas on the basis of histological morphology, one was excluded from the study subsequently because it was CD3 positive, suggestive of a MUM-1 positive T cell lymphoma or a plasmacytoma that aberrantly expressed CD3 (see below). The 23 lymph nodes without lymphoma were biopsied for evaluation of metastasis of nonlymphoid malignant tumours, including mammary tumours (n = 15), squamous cell carcinoma (n = 2), thyroid carcinoma (n = 1), undetermined epithelial tumours (n = 5). Additional lymph nodes with no evidence of lymphoma were also included to evaluate the specificity of each oligonucleotide primer set. Tissues removed surgically were fixed in 10% phosphate buffered formalin, dehydrated through a graded series of increasing ethanol concentrations, embedded in paraffin and sectioned at 4 mm thickness. Staining with haematoxylin and eosin (H&E), and immunohistochemistry using mouse monoclonal anti-human MUM-1 (clone MUM1p; Dako Cytomation), rabbit polyclonal anti-human cluster of differentiation 3 (CD3; Dako Cytomation) and rabbit polyclonal anti-human CD20 (Thermo Fisher Scientific) antibodies were performed as described previously (Kagawa et al., 2011; Takanosu and Kagawa, 2015). Immunoreactivity was scored as: +++, 50–100% positive cells; + +, 10–50% positive cells; +, <10% positive cells. As negative controls, sections were incubated with normal mouse or rabbit serum instead of the primary antibody. Extraction of genomic DNA and PCR for verification of DNA integrity Four micrometre thick sections were mounted on glass slides, deparaffinised twice in xylene for 5 min, passed through a graded series of ethanol solutions from 100% to 70% and then hydrolysed in 10 mM Tris-HCl buffer (pH 8.0). The tumour area of each section was scraped from the glass slides with a 23 Ga needle and genomic DNA was extracted using the High Pure PCR Template Preparation Kit (Roche Diagnostic). DNA concentrations were determined using a spectrophotometer (NanoDrop-1000, Thermo Scientific). Genomic DNA integrity was verified by amplifying portions of the IGH constant region gene (130 base pairs, bp; Burnett et al., 2003), the ATP7B gene (195 bp) and the SOD1 gene (266 bp). The sequences of the primers used in this study are shown in Table 1. PCR was performed as described previously (Takanosu and Kagawa, 2015) using 50 ng genomic DNA in a total volume of 20 mL containing 1
PCR buffer, 0.2 mM deoxynucleotide triphosphates, 0.2 mM

of each forward and reverse primer, and 0.5 units of Taq DNA polymerase (BlendTaq plus, Toyobo). The thermal cycling conditions consisted of 95  C for 1 min and 35 cycles of 94  C for 30 s, annealing at the specified temperature (60  C, 65  C and 62  C for the IGH constant region, ATP7B and SOD1, respectively) for 20 s and 72  C for 30 s. PCR products were analysed using a capillary electrophoresis system with a highresolution cartridge (Qsep-100, BiOptic). FR2 and FR3 primer design The target region of each primer on the rearranged V/J gene is shown in Fig. 1. In addition to using the existing primer sets CB1/CB2 and CB1/CB3 (Burnett et al., 2003) and TamuraV/TamuraJ (Tamura et al., 2006), we designed forward primers for FR2 and FR3 of the V gene (Table 1). The IGH V genes were obtained with the BLAST search programme (National Centre for Biotechnology Information, NCBI)1 using the CB1 primer and the canine Non-RefSeq RNA data base. Of the 100 sequences obtained from the data base, 23 randomly chosen sequences were aligned using ClustalW,2 and the FR2 and FR3 primers were designed against the conserved region and designated FR2-6 and FR3-1, respectively. These primers were compared to the V gene sequences published by Bao et al. (2010). The FR2-6 and FR3-1 primers completely matched 51 and 15 V genes belonging to IGHV gene group 3. These primers were used with the CB2 or CB3 reverse primers targeting IGHJ1-5 or IGHJ6, respectively. The forward FR2-6 and FR3-1 primers were designed to amplify genomic DNA extracted from formalin-fixed paraffin-embedded (FFPE) sections, so that the size of the PCR product, when combined with reverse primers, was <250 bp (Fig. 1).

1 See: Canis lupus familiaris (dog) Nucleotide BLAST. https://blast.ncbi.nlm.nih. gov/Blast.cgi?PAGE_TYPE=BlastSearch&PROG_DEF=blastn&BLAST_PROG_DEF=megaBlast&BLAST_SPEC=OGP__9615__10726. 2 See: Multiple Sequence Alignment by CLUSTALW. https://www.genome.jp/ tools-bin/clustalw.

M. Takanosu et al. / The Veterinary Journal 241 (2018) 31–37

33

Clonality analysis and sequencing Seven primer sets, namely CB1/CB2, TamuraV/TamuraJ, FR3-1/CB2 and FR2-6/CB2 (targeting IGHV3/IGHJ1-5), and CB1/CB3, FR3-1/CB3, and FR2-6/CB3 (targeting IGHV3/IGHJ6), were used for clonality analysis. All reactions were performed in duplicate to exclude pseudoclonality. The thermal cycling conditions were as described above, except that the annealing temperature was 61  C for the primer pairs CB1/CB2, CB1/CB3, FR2-6/CB2, FR2-6/CB3, FR3-1/CB2 and FR3-1/CB3, and 56  C for TamuraV/TamuraJ. The PCR products were analysed using a capillary electrophoresis system with a high-resolution cartridge (Qsep-100, BiOptic). Interpretation of electrophoretic profiles was performed as described by Keller et al. (2016). Clones with one or two distinct and reproducible peak(s) that were two-fold higher than the polyclonal base were defined as monoclonal or biclonal, respectively. Clones with several distinct and reproducible peaks were defined as oligoclonal. Clones with distinct and non-reproducible peak(s) were defined as pseudoclonal. Clones showing a Gaussian distribution were defined as polyclonal. The absence of a PCR product was defined as ‘no amplification’. Sequencing analysis was performed as described by Takanosu et al. (2016). PCR products were treated with Exo-SAP IT (Thermo Fisher Scientific), then sequenced on an ABI 3730xl Analyzer (Applied Biosystems) using forward (FR2-6 or FR3-1) and reverse (CB2) primers.

Results Immunohistochemistry Positive immunoreactivity for MUM-1 was observed in all 30 cases diagnosed as plasmacytoma on the basis histomorphological features that were initially included in the study, with 50–100% of tumour cells being positive (Fig. 2; Table 2). After excluding one tumour which was also CD3 positive, 29 cases diagnosed as plasmacytomas on morphological and immunohistochemical grounds were used for subsequent clonality analysis.

Fig. 2. Haematoxylin and eosin staining (A) and anti-MUM1-immunoreactivity (B) of a case of canine plasmacytoma (case 27). (A) Round cells are distributed diffusely through the tumour mass. (B) Most round cells exhibit positively immunolabelling for MUM1. Bars = 50 mm.

Table 2 Results of clonality analysis and immunohistochemistry. Case

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29

Clonality analysis

Immunohistochemistry

CB1/CB2 IGHV3/J1-5

CB1/CB3 IGHV3/J6

TamuraV/J IGHV3/J1-5

FR3-1/CB2 IGHV3/J1-5

FR3-1/CB3 IGHV3/J6

FR2-6/CB2 IGHV3/J1-5

FR2-6/CB3 IGHV3/J6

NA C NA NA C NA C NA C NA NA NA NA NA NA NA NA C C NA C NA C NA NA NA C C NA

NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA

NA NA NA NA NA NA C NA C NA NA NA NA NA NA NA NA NA C NA C NA NA NA NA NA NA NA NA

NA C NA C C NA C NA C NA NA NA NA NA NA NA NA NA C NA C NA C NA NA NA C C NA

NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA

NA C NA NA NA NA NA NA C NA NA NA NA NA NA NA NA NA C NA NA NA C NA NA NA C C NA

NA NA NA NA NA NA NA NA NA NA NA NA C NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA

C, clonal; NA, not amplified; +++ 50–100% cells positive;  all cells negative.

MUM1 +++ +++ +++ +++ +++ +++ +++ +++ +++ +++ +++ +++ +++ +++ +++ +++ +++ +++ +++ +++ +++ +++ +++ +++ +++ +++ +++ +++ +++

CD3                             

34

1000

130

20

M. Takanosu et al. / The Veterinary Journal 241 (2018) 31–37

IGH constant region

sensitivity than with primers IGHV3/IGHJ1-5 (Table 3). In each case, the signal from TamuraV/TamuraJ primers exhibited a lower fluorescence intensity than signals from the other primer sets (Fig. 4; case 21). Case 4 exhibited monoclonality with the FR3-1/ CB2 primer set, but with none of other six primer sets (Fig. 4). Case 13 exhibited clonality only with the FR2-6/CB3 primer set (Fig. 4). Case 18 exhibited clonality only with the CB1/CB2 primer set (Fig. 4). Twelve out of 29 cases (41.4%) exhibited clonality with at least one of the primer sets used.

195

Sequence analysis of primer sites

ATP7B

266

SOD1

PCR amplicons from the cases showing amplification with the FR2-6/CB2 primer set (cases 2, 9, 19, 23, 27 and 28), and case 4 with PCR product amplified using the FR3-1/CB2 primer set (but not with the CB1 primer set), were sequenced to analyse the primer regions on FR3 and to examine the primer-annealing sites on the V gene (Fig. 5). Of the six cases with PCR products amplified using the FR2-6/CB2 primer set, one to two mismatches were detected in five cases in the FR3-1 region. Mismatches were observed most frequently in the middle and in the 50 -end of FR3-1. Cases with more than two mismatches in FR3-1 primer region were not observed. Two to three mismatches were detected scattered throughout the CB1 primer region in 5/7 cases. In the TamuraV primer-binding region, 6/7 cases showed one to three mismatches that were scattered within the entire primer region. Clonality analysis of lymph nodes with and without lymphoma

Fig. 3. Control PCR runs performed to evaluate the integrity of genomic DNA at three different loci (IGH constant region, ATP7B and SOD1); a single peak representing each locus can be visualised at 130, 195 and 266 base pairs (bp), respectively. Two peaks at the start and end illustrate 20 and 1000 bp markers.

Clonality analysis of plasmacytomas Genomic DNA extracted from FFPE sections was of sufficient integrity to amplify control PCR amplicons of 130, 195 and 266 bp in all 29 cutaneous plasmacytoma samples, 23 DLBCL samples and 23 lymph node samples without lymphoma (Fig. 3). The results of the cutaneous plasmacytoma clonality analysis are shown in Fig. 4 and Tables 2 and 3. The detection frequencies with primers IGHV3 and IGHJ1-5 were 13.8–34.5% (Table 3). The most sensitive primer sets were CB1/CB2 and FR3-1/CB2. The detection frequencies with the IGHV3/IGHJ6 primer set were 0–3.4%, indicating a lower

The results of clonality analysis of lymph nodes with DLBCLs and lymph nodes without lymphoma are summarised in Table 3. Out of 23 DLBCL cases, the frequencies of detection of clonality were 65.2–78.3% with the IGHV3/IGHJ1-5 primers and 8.7-21.7% with the IGHV3/IGHJ6 primers. Twenty-one of 23 DLBCL cases (91.3%) exhibited clonality with at least one of the primers used in this study. In contrast, of the 23 cases of non-lymphoma lymph nodes, the frequencies of detection of clonality were 0–4.3% with the IGHV3/IGHJ1-5 primers and 0% with the IGHV3/IGHJ6 primers. One non-lymphoma exhibited clonality with both the CB1/CB2 and FR3-1/CB2 primer sets targeting IGHV3 (Fig. 6); no evidence for lymphoma was found on histopathological evaluation. The specificities of these primers on the basis of testing non-lymphoma lymph nodes were 95.7% for the CB1/CB2 and FR3-1/CB2 primer sets, and 100% for the CB1/CB3, TamuraV/J, FR2-6/CB2, FR2-6/CB3 and FR3-1/CB3 primer sets. Discussion In this study, the sensitivity of IGH clonality testing for canine cutaneous plasmacytoma was lower than that for DLBCL. It was anticipated that plasmacytomas, which are derived from B cells, would have a similar frequency of detection of IGH clonality to

Table 3 Frequency of detection of B cell clonality (percentages are given in parentheses). Rearrangement

Gene group recombination

Plasmacytoma n = 29

DLBCL n = 23

LN n = 23

B-major B-minor B-Tamura FR3-1/CB2 FR3-1/CB3 FR2-6/CB2

IGHV3/J1-5 IGHV3/J6 IGHV3/J1-5 IGHV3/J1-5 IGHV3/J6 IGHV3/J1-5

10 (34.5) 0 (0) 4 (13.8) 10 (34.5) 0 (0) 6 (20.7)

15 (65.2) 4 (17.4) 17 (73.9) 18 (78.3) 5 (21.7) 18 (78.3)

1 0 0 1 0 0

DLBCL, diffuse large B cell lymphoma; LN, lymph nodes biopsied for evaluation of metastasis of non-lymphoid tumours.

(4.3) (0) (0) (4.3) (0) (0)

M. Takanosu et al. / The Veterinary Journal 241 (2018) 31–37

35

Fig. 4. Representative data from clonality analysis of plasmacytomas (cases 4, 13, 18 and 21). Case 4 exhibited clonality only with the FR3-1/CB2 primer set. Case 13 exhibited clonality only with the FR2-6/CB3 primer set. Case 18 exhibited clonality only with the CB1/CB2 primer set. Case 21 exhibited clonality with the CB1/CB2, TamuraV/J and FR31/CB1 primer sets. The signal obtained using the TamuraV/J primers was less intense than the signals obtained using the other primer sets. The two peaks at the start and end depict the 20 and 1000 base pair (bp) markers, respectively.

other mature B cell lymphomas. Unexpectedly, clonality was detected in only 0–34.5% of canine cutaneous plasmacytoma cases with the seven primer sets used in this study. Clonality was detected less frequently in plasmacytomas than in B cell lymphomas for all primer sets. In addition to previously reported forward primers targeting IGHV3 (CB1 and TamuraV), new primers mapping to the FR2 and FR3 regions of the IGHV3 gene-homology region were designed on the basis of IGH cDNA sequences obtained from the canine NonRefSeq RNA data base in NCBI and were designated FR2-6 and FR31, respectively. When primers FR2-6 and FR3-1 were compared with the canine V gene sequences published previously (Bao et al., 2010), they completely matched 51 and 15 out of 76 V gene subgroup 3 sequences, respectively. Both primers included

IGHV3-19*01 and IGHV3-38*01, which are preferentially used for IGH rearrangements (Bao et al., 2010). Primers CB1 and TamuraV matched 27 and 23 V gene segments, also including IGHV3-19*01 and IGHV3-38*01. Thus, the newly designed primers FR2-6 and FR3-1 may be used for IGH clonality analysis, since they have comparable sensitivity to CB1 and TamuraV primers when applied to DLBCLs. However, these primers, which were located upstream of the regions recognised by CB1 and TamuraV, did not improve the sensitivity of detection of clonality in canine plasmacytomas. Another reverse primer (CB3) did not improve the frequency of detection of clonality in plasmacytomas. In addition, primer sets containing CB3 in all cases had lower frequencies of detection of clonality in both DLBCLs and plasmacytomas than achieved with

36

M. Takanosu et al. / The Veterinary Journal 241 (2018) 31–37

LN_1: CB1/CB2

LN_1: FR3-1/CB2

LN_19 : CB1/CB2

LN_19 : FR3-1/CB2

Fig. 5. Alignment of the sequences obtained from PCR products amplified using primers FR2-6/CB2 (cases 2, 9, 19, 23, 27 and 28) or FR3-1/CB2 (case 4). Shaded areas indicate matched codons within the primer sequence. The positions of the primers FR2-6, FR3-1, CB1 and Tamura-V are indicated. The number on the left of each sequence is the case number (Table 2). Most mismatches were present in the 50 end to the middle of primer FR3-1. Mismatches were scattered throughout the CB1 and TamuraV primer-binding regions.

primer sets containing CB2. Since the CB2 and CB3 primers anneal to IGHJ1-5 and IGHJ6, respectively (Rout et al., 2018), the difference in the target of CB2 and CB3 might explain differences in the frequency of detection. Frequent somatic hypermutation, leading to primer mis-annealing, is a possible explanation for the low sensitivity of the clonality assay; a similar mechanism has been reported in human (Van Dongen et al., 2003) and feline (Werner et al., 2005) clonality assays. In feline lymphoma, several somatic hypermutations in the primer-annealing region made clonality undetectable (Werner et al., 2005). In our study, mismatches suspected to be result of somatic hypermutation were detected in the forward primer-annealing regions. Thus, somatic hypermutation may explain the low sensitivity of detection of clonality in canine plasmacytomas. To overcome this limitation, other target loci need to be identified for clonality assays. The kappadeleting element could provide an additional target to improve the sensitivity of clonality testing (Keller et al., 2016). An additional cause of low sensitivity may be that some V genes, which are not recognised by the primers used in this study, are rearranged in plasmacytomas. The primers used in this study annealed to the V genes belonging to subgroup 3, but not to those belonging to subgroups 1 and 4 (Rout et al., 2018). The primers available to date have been designed against conserved regions based on alignment of V genes (Burnett et al., 2003; Tamura et al., 2006). The primers constructed in the present study were also designed on the basis of sequences retrieved from the non-RefSeq RNA data base using the CB1 primer sequence as query. Consequently, all retrieved sequences represented V gene subgroup 3; hence, all newly designed primers only recognise V gene subgroup 3. Therefore, the forward primers used in this study do not cover every possible V gene, especially those in subgroups 1 and 4. Thus, primers constructed specifically using these subgroups may increase the sensitivity of detecting IGH clonality.

Contamination of samples used for genome extraction with non-neoplastic lymphocytes may decrease the apparent sensitivity of detection in clonality assays (Burnett et al., 2003; Keller et al., 2016). In the present study, staining with H&E staining and immunohistochemistry for MUM1 revealed that the tumour masses were occupied by neoplastic plasma cells, since 50–100% of the cells in the mass were positive for MUM-1. Furthermore, a fine needle was used to excise only the region of tumour from glass slides for genomic DNA preparation. No polyclonal signals were observed in cases that did not exhibit a PCR signal in the clonality assay, suggesting that the amount of non-neoplastic lymphocytes in the analysed samples was very low. In non-lymphoma lymph nodes, only one case exhibited clonality when the CB1/CB2 and FR3-1/CB2 primers were used. The reason for clonality was not elucidated, as follow-up data for this animal were not available. The specificity of these primers using lymph node samples was 95.7%. No cases exhibited clonality with the FR2-6/CB2, FR2-6/CB3, FR3-1/CB3 and TamuraV/TamuraJ primer sets. High specificity of clonality testing was also reported previously (Burnett et al., 2003; Schöpper et al., 2016). Conclusions Clonality testing based on the IGH locus for cutaneous plasmacytoma had low sensitivity and high specificity. In comparison with the results obtained for DLBCLs, PCR using the IGH locus had lower sensitivity in detecting clonality in canine cutaneous plasmacytomas. This finding might be due to frequent somatic hypermutation of the IGH locus and insufficient coverage of the primers used in this study. However, it is important to note that negative clonality testing results do not exclude lymphoid neoplasia. Further analysis should be conducted to explore the

M. Takanosu et al. / The Veterinary Journal 241 (2018) 31–37

37

Acknowledgements This work was supported by JSPS KAKENHI grant number 16H00481. References

Fig. 6. Representative data from clonality analysis of lymph nodes without lymphoma. LN_1 and LN_19 represent the case number of the lymph node samples. A polyclonal peak in each primer set (CB1/CB2 and FR3-1/CB2) can be seen in LN_1. Monoclonal peaks were seen with the CB1/CB2 and FR3-1/CB2 primer sets at 110 and 177 base pairs (bp), respectively.

possibility of using other targets, such as the kappa chain, in clonality testing for canine plasmacytomas. Conflict of interest statement None of the authors of this paper has a financial or personal relationship with other people or organisations that could inappropriately influence or bias the content of the paper.

Bao, Y., Guo, Y., Xiao, S., Zhao, Z., 2010. Molecular characterization of the VH repertoire in Canis familiaris. Veterinary Immunology and Immunopathology 137, 64–75. Burnett, R.C., Vernau, W., Modiano, J.F., Olver, C.S., Moore, P.F., Avery, A.C., 2003. Diagnosis of canine lymphoid neoplasia using clonal rearrangements of antigen receptor genes. Veterinary Pathology 40, 32–41. Gupta, A., Fry, J.L., Meindel, M., Gumber, S., 2014. Cutaneous extramedullary solitary digital plasmacytoma in a dog. Journal of the American Veterinary Medical Association 244, 163–165. Kagawa, Y., Yamashita, T., Maetani, S., Aoki, Y., Sakaguchi, K., Hirayama, K., Umemura, T., Taniyama, H., 2011. Cutaneous lymphoplasmacytic lymphoma with systemic metastasis in a cat. Journal of Veterinary Medical Science 73, 1221–1224. Keller, S.M., Vernau, W., Moore, P.F., 2016. Clonality testing in veterinary medicine: A review with diagnostic guidelines. Veterinary Pathology 53, 711–725. Mikiewicz, M., Otrocka-Domagała, I., Pa zdzior-Czapula, K., Gesek, M., 2016. Morphology and immunoreactivity of canine and feline extramedullary plasmacytomas. Polish Journal of Veterinary Sciences 19, 345–352. Schöpper, I., Ohmura, S., Rütgen, B., Tusjimoto, H., Weber, K., Hirschberger, J., 2016. Melting curve analysis in canine lymphoma by calculating maximum fluorescence decrease. Veterinary and Comparative Oncology 15, 563–575.   Pazdzior-Czapula, K., Rotkiewicz, T., Otrocka-Domagała, I., Gesek, M., Smiech, A., 2015. Morphology and immunophenotype of canine cutaneous histiocytic tumours with particular emphasis on diagnostic application. Veterinary Research Communications 39, 7–17. Ramos-Vara, J.A., Miller, M.A., Valli, V.E., 2007. Immunohistochemical detection of multiple myeloma 1/interferon regulatory factor 4 (MUM1/IRF-4) in canine plasmacytoma: Comparison with CD79a and CD20. Veterinary Pathology 44, 875–884. Rout, E.D., Shank, A.M., Waite, A.H., Siegel, A., Avery, A.C., Avery, P.R., 2017. Progression of cutaneous plasmacytoma to plasma cell leukemia in a dog. Veterinary Clinical Pathology 46, 77–84. Rout, E.D., Burnett, R.C., Labadie, D.J., Yoshimoto, A.J., Avery, A.C., 2018. Preferential use of unmutated immunoglobulin heavy variable region genes in Boxer dogs with chronic lymphocytic leukemia. Public Library of Science One 13, e0191295. Takanosu, M., Kagawa, Y., 2015. Comparison of primer sets for T-cell clonality testing in canine intestinal lymphoma. Journal of Veterinary Diagnostic Investigation 27, 645–650. Takanosu, M., Amano, S., Kagawa, Y., 2016. Analysis of c-KIT exon 11 mutations in canine gastrointestinal stromal tumours. The Veterinary Journal 207, 118–123. Tamura, K., Yagihara, H., Isotani, M., Ono, K., Washizu, T., Bonkobara, M., 2006. Development of the polymerase chain reaction assay based on the canine genome database for detection of monoclonality in B cell lymphoma. Veterinary Immunology and Immunopathology 110, 163–167. Thalheim, L., William, L.B., Borst, L.B., Forgle, J.E., Suter, S.E., 2013. Lymphoma immunophenotype of dogs determined by immunohistochemistry, flow cytometry, and polymerase chain reaction for antigen receptor rearrangements. Journal of Veterinary Internal Medicine 27, 1509–1516. Van Dongen, J.J., Langeral, A.A., Brüggemann, M., Evans, P.A., Hummel, M., Lavender, F.L., Delabesse, E., Davi, F., Schuuring, E., García-Sanz, R., et al., 2003. Design and standardization of PCR primers and protocols for detection of clonal immunoglobulin and T-cell receptor gene recombinations in suspect lymphoproliferations: report of the BIOMED-2 concerted action BMH4-CT983936. Leukemia 17, 2257–2317. Van Wettere, A.J., Linder, K.E., Suter, S.E., Olby, N.J., 2009. Solitary intracerebral plasmacytoma in a dog: microscopic, immunohistochemical, and molecular features. Veterinary Pathology 46, 949–951. Velez, D., Hinojar-Gutierrez, A., Nam-Cha, S., Acevedo-Barbera, A., 2007. Laryngeal plasmacytoma presenting as amyloid tumour: a case report. European Archives of Oto-Rhino-Laryngology 264, 959–961. Waugh, E.M., Gallagher, A., Haining, H., Johnston, P.E., Marchesi, F., Jarrett, R.F., Morris, J.S., 2016. Optimisation and validation of a PCR for antigen receptor rearrangement (PARR) assay to detect clonality in canine lymphoid malignancies. Veterinary Immunology and Immunopathology 182, 115–124. Werner, J.A., Woo, J.C., Vernau, W., Graham, P.S., Grahn, R.A., Lyons, L.A., Moore, P.F., 2005. Characterization of feline immunoglobulin heavy chain variable region genes for the molecular diagnosis of B-cell neoplasia. Veterinary Pathology 42, 596–607.