One method to establish Epstein‐Barr virus‐associated NK/T cell lymphoma mouse models

Abstract Novel nude mice model of human NK/T cell lymphoma were established by subcutaneously injecting two NK/T cell lymphoma cell lines into the right axillary region of mice and successful passages were completed by injecting cell suspension which was obtained through a 70‐μm cell strainer. These mice models and corresponding cell clones have been successfully developed for more than 8 generations. The survival rates of both resuscitation and transplantation in NKYS and YT models were 90% and 70% correspondingly. Pathologically, the tumour cells in all passages of the lymphoma‐bearing mice and cell lines obtained from tumours were parallel to initial cell lines. Immunologically, the tumour cells expressed the characteristics of the primary and essential NK/T lymphomas. The novel mice models maintained the essential features of human NK/T cell lymphoma, and they would be ideal tools in vivo for further research of human NK/T cell lymphoma.


| INTRODUCTION
Extranodal natural killer (NK)/T cell lymphoma (ENKTCL) is an aggressive malignancy of putative NK cell origin, with a minority deriving from the T cell lineage. Pathologically, the malignancy occurs in two forms, extranodal NK/T cell lymphoma, nasal type (NNKTCL); and aggressive NK cell leukaemia. 1 Although chemotherapy and radiotherapy can improve the disease outcome, the prognosis of NKTCL is poor, and no targeted therapy is currently available. [2][3][4][5][6] The pathogenic mechanisms of NK/T cell lymphoma remain largely unclear, and further investigation may aid in improving the accuracy of lymphoma diagnosis and gene therapy. 7 Meanwhile, lymphoma is one immune-related disease, 8 the microenvironment in vivo matters a lot in the mechanism of lymphoma. Complex biological processes often require in vivo analysis, and many important research advances have been made using mice as a model for the study of various biological systems. Mouse models are irreplaceable tools for the study of carcinogenesis. The mouse shares anatomical, immunological, and genomic similarities with humans and is the most accessible model system. 9 To clarify the molecular pathogenesis of NKTCL, we performed establishment of mouse models in NKTCL, because animal models enable elucidation of the pathological and physiological mechanisms underlying lymphomas and preclinical test- is not only an efficient in vivo platform to investigate drug efficacy and resistance studies in various cancers, but likely to be applied to lymphoma therapy. There is a growing need for mouse models that have stable characteristics to mimic in vivo studies of NK/T cell lymphomas. Here we introduce two NK/T cell lymphoma mouse models we have established successfully.

| Nude mice model establishing
Three-to four-week-old female BALB/c (nu/nu) nude mice and NOD/ SCID mice were bought from Beijing Vital River Laboratory Animal Technology Co., Ltd. (Beijing, China), weighed approximately 15-20 g, fed in the specific pathogen-free (SPF) barrier system and provided sterile food and water. All procedures were performed under aseptic condition. Before animal experiment, mice were fed adaptively for 1 week in the new circumstances and cyclophosphamide (CTX) for 2 days before cell suspension injection and transplantation.
Cells were washed repeatedly in Hank's liquid, resuspended in 0.2 mL RPMI1640 culture media and subcutaneously injected into the right axillary region of mice (NKYS cells in NOD/SCID mice and YT cells in nude mice). The NKYS models were injected with 20 000 IU/kg rhIL-2 once a week intraperitoneally. The general condition and the appearance changes of the lymphoma bearing mice were observed every day, and the size of tumour was measured in length and width three times a week. As soon as the tumour reached about 2000 mm 3 in size or 1.0-2.0 cm at length, or the animal showed distress, the mouse was killed. A part of the tumour tissues was used for continuous transplantation, the remaining part was divided into two parts, one for routine morphologic observation and the other preserved in liquid nitrogen for further use.

| Bioluminescent imaging
To investigate whether the tumour existed in other part of mice, tumour development was quantified by in vivo imaging using an IVIS Lumina III system (PerkinElmer, Waltham, MA, USA). Mice were anaesthetized with isoflurane, injected i.p. with 150 mg/kg body weight Luciferin (PerkinElmer) and imaged 5 minutes after Luciferin injection. Total flux values, which correlate directly with tumour mass, were measured. Images were analysed with Living Image 3.0 software (PerkinElmer).

| Passage
A period of time after cell suspension injection (at least 35 days in NKYS cells and 45 days in YT cells), the tumour-bearing mice (denoted F0) were killed. We took the passage in three ways initially.
Inserting the small block from the tumour into the next mice is the first. However, both the tumorigenecity and growth rate were quite low, so lastly we abandoned this method. A part of tissues excised from the tumour were minced and strained through a 70-μm cell strainer (Becton Dickinson, San Jose, CA, USA). One part of the cells obtained through the strainer was injected in the same manner as for the establishment of first mice model. 10 Others were cultured and began to proliferate. In the case of engraftment, 1.5 × 10 7 to 5 × 10 7 tumour cells were serially subcutaneously injected into the right axillary region of other mice. Stable mouse models were considered to be established based on the success of over six serial passages of the tumours (F5). Currently, the tissues and cells obtained from F0, F4, and F8 mice were used in the subsequent analyses.
Notably, cells were injected with matrigel in the ratio of 1:1.

| Time course engraftment analyses after transplantation
To investigate the pattern of engraftment and proliferation of the lymphoma cells in the mouse models, time-lapse engraftment analyses were performed. After subcutaneous injection of 2.0 × 10 7 tumour cells in the right axillary region of mice, tumour engraftment, and proliferation patterns were analysed every 3 days.

| In situ hybridization
Epstein Barr virus (EBV) encoded EBV-encoded RNA in situ hybridization (RISH) was performed to detect RNA transcripts of EBV using RISH Epstein Barr encoded RNA probe kit (ZSGB-BIO, Beijing, China, ISH-6022-Y, 170510). Tissue sections were deparaffinized, rehydrated, and digested with proteinase K enzyme. After pretreatment in 10 mmol L −1 sodium citrate (pH 6.0), sections were hybridized with digoxygenin labelled RISH EBER probe by incubation at 55°C for 1 hour. Following hybridization, sections were washed with tris buffer (pH 7.6) and hybridized probe are detected and diaminobenzidene treatment followed by counterstaining with haematoxylin. Staining was observed under light microscope. Dark brown staining of the tumour nucleus was considered as positive for EBER transcripts expression.

| Statistical analysis
Two-tailed t tests of equal variance were used to analyse flow cytometry data. Bioluminescence was evaluated using unpaired t tests. Statistical significance was defined as a P value of less than or equal to 0.05. All statistical analyses were done by using Prism 5.  Figure S1. The combined data suggest that, at least in these models, the tumour cells were not disseminated (data not shown). Each passage of tumour tissue suspension and/or cells was injected into the next mice with matrigel. The pieces of fresh transplanted tumour samples in 7th passage were taken from the mice, and transplanted directly into the right axillary region of 10 mice respectively. The NKYS and YT transplanted tumours had appeared in all of the 9 NOD/SCID mice, but YT in all of 7 nude mice.
The growth rate is not different in every passage (data not shown) and the growth curve in the last passage is shown in Figure S1.

| Tissue pathologic features
The cells in the process of passage had the same feature with the original cell lines. In YT and its serial passage mice models tissues, the normal architecture was effaced, heterogeneous lymphocytes diffused into the plate, with big volume and abundant cytoplasm.
F I G U R E 1 In YT and its serial passage mice models tissues, the normal architecture was effaced, heterogeneous lymphocytes diffused into the plate, with big volume and abundant cytoplasm. The karyotype in cells is irregular and the chromatin is coarse. apoptotic and necrotic tumour cells were found XUE ET AL.

| 1511
The karyotype in cells is irregular and the chromatin is coarse. apoptotic and necrotic tumour cells were found (Figure 1). The tumour tissues in YT F0, YT F5 and YT F7were positive for CD56, Granzyme B, Perforin (Figure 2) but negative for TiA1 (Table 1).

Sections of the biopsy from serial NKYS cell mice models
showed similar morphological features with YT mouse model tissues ( Figure 3). Immunohistochemical staining showed the large atypical cells were positive for CD56, Granzyme B, Perforin, TiA1 ( Figure 4 and Table 1).

| EBV infection
EBV was strongly detected in situ hybridization for EBV RNA using the EBER probe in YT, NKYS, and later passage tissues (Figures 7   and 8; Table 1).

| DISCUSSION
In vitro studies have always been a powerful tool to characterize many cancer-associated processes. However, the tumours comprise not only the malignant cells but also a plethora of nonmalignant cells and extracellular matrix-the tumour microenvironment, 11 that's why numerous aspects of neoplastic growth are difficult to model in vitro alone. In vitro approach has limitations when we extrapolate to the more complex in vivo scenario. 12 In vivo strategies are expected to  preserved the original characteristics of EBV-associated nasal angiocentric T/NK cell lymphoma. 28 These two cell lines are successfully representative of the typical NK/T cell lymphoma occurring in the world. In our mouse models, there is no significant difference in growth rate, histopathological, and immunohistochemical features among different passages of YT and NKYS models, which largely proves that we have got stable mouse models. More importantly, we invent a method to establish the NK/T mice model. CTX is necessary before injecting and implanting so that the immune system of mice can be inhibited more thoroughly. However, NKYS model is still hard to establish. In terms that NKYS cell line is dependent on IL-2, We hypnosis that the tumour need IL-2 to grow. It is critical that the tumour itself need the same environment with the corresponding cell line. In the future, the mouse models will facilitate the preclinical research of NK/T cell lymphoma.
However, using cell lines as inoculum an IL-2 support does barely reflect "real world" inter-patient heterogeneity of the lymphomas because the subcutaneously injected cells did not disseminate beyond the injection site. That is a world-wild question and there is a long way to go in the research on establishing mouse models that can mimic the real tumour environment in patients.