Expression of the Bcl-2 apoptotic marker in cats diagnosed with inflammatory bowel disease and gastrointestinal lymphoma

Abstract

Immunolabeling for the critical lymphocyte survival factor, Bcl-2, of intestinal biopsies from cats with histologic evidence of inflammatory bowel disease (IBD) or gastrointestinal (GI) lymphoma was evaluated to determine if expression differed significantly between these two disease processes. Immunolabeling for Bcl-2 was performed on small intestinal endoscopic or full thickness biopsy sections from 55 cats. Diagnosis of IBD, T-cell lymphoma or B-cell lymphoma was established previously. The percentage of infiltrating lymphocytes that were positively labeled for Bcl-2 was subjectively determined for each case. Eight cats were diagnosed with IBD and 47 cats with lymphoma. A significantly higher percentage of cells were positively immunolabeled for Bcl-2 in cats with GI lymphoma [median (range); 90 (5–95)%] compared with cats with IBD [60 (15–95)%] (P = 0.029). However, the overall degree of positive immunolabeling in both groups tended to be high. This over-expression of Bcl-2 may prove useful as a therapeutic target for IBD and GI lymphoma in cats.

Introduction

Lymphoma is the most common malignant neoplasm in domestic cats, and the gastrointestinal (GI) tract is the most commonly affected primary site.^1–5 Accurately differentiating feline GI lymphoma from inflammatory bowel disease (IBD) by non-invasive methods can be challenging. Even in advanced disease, primary lymphomas of the alimentary tract of cats have a tendency to remain localized to the lamina propria before invasion of the muscularis or even ulceration of the mucosa occurs.^6,7 Within the mucosa only the degree of epitheliotropism has been shown to help differentiate lymphocytic inflammation from small cell intestinal lymphoma. ⁷ Regardless, as endoscopic mucosal biopsies rarely contain submucosa and do not contain muscularis, evaluation of the mucosal lymphocytic infiltrate without significant epitheliotropism by histology alone can be of limited diagnostic value and misdiagnoses are all too common.^1,7,8 Moreover, the ileocecocolic junction is a common site for GI lymphoma; this region is not reliably accessible at every diagnostic facility and percutaneous diagnostic methods are not consistently employed.^8,9

Abdominal ultrasound examination may be helpful in the diagnosis of many cases of infiltrative GI disease and is considered more sensitive than routine radiography. ¹ In a recent study of 142 cats, lymph node enlargement was associated with both GI lymphoma and IBD; thus, the presence of lymphadenopathy does not differentiate between these diseases. Furthermore, whilst thickening of the muscularis propria may suggest GI lymphoma, it is not considered to be pathognomonic and diagnostic errors may occur. ¹⁰

Non-Hodgkin’s lymphoma of the gastrointestinal tract is the most common extra-nodal lymphoma observed in humans. ¹¹ The B-cell lymphoma gene-2 (bcl-2) encodes an integral inner mitochondrial membrane protein, Bcl-2, which is a critical cellular survival factor. Up-regulation of Bcl-2 expression prevents activation of the intrinsic cell death pathway by limiting translocation of cytochrome c from the inner mitochondrial membrane, thereby limiting activation of death executioner caspases.^12–14 Bcl-2 protein expression is up-regulated in the majority of human follicular lymphomas and is considered a key etiologic feature of mucosa-associated lymphoid tissue lymphomas of the GI tract in human patients. Bcl-2 expression is noted in a proportion of nodal B-cell non-Hodgkin’s lymphomas and many primary cutaneous T-cell lymphomas of humans.^12,15 In contrast to human GI lymphoma, the B-cell follicular phenotype is rarely identified in feline GI lymphoma. In a study of 29 cats diagnosed with lymphoma, 21/29 (72%) lymphomas expressed Bcl-2, as determined by immunohistochemistry. Seventeen of 20 (85%) T-cell lymphomas demonstrated Bcl-2 immunoreactivity, whereas 5/9 (56%) B-cell lymphomas demonstrated Bcl-2 immunoreactivity, which seems contradictory to the role of bcl-2 in regulation of human B-cell lymphomas. ¹⁶ Furthermore, recent evidence suggests that Bcl-2 is also expressed in normal intraepithelial lymphocytes in humans. In contrast, Bcl-2 expression is decreased in lymphocytes from patients with type II refractory celiac disease, a clinicopathological entity considered an intermediary step between celiac disease and enteropathy-associated T-cell lymphoma. ¹⁷ These data suggest Bcl-2 as a potential diagnostic marker to differentiate IBD from intestinal lymphoma, as well as a predictive marker for using Bcl-2 inhibitors for therapy.

Although the majority (90%) of feline nodal and mediastinal lymphomas are classified as intermediate or high grade by histologic assessment, multiple reports have found that most small intestinal lymphomas are comprised of small lymphocytes.^6,7,18,19 Thus, on the basis of cell morphology alone, it may be extremely difficult to distinguish IBD from GI lymphoma, even when full thickness biopsies are obtained are obtained. Immunohistochemical labeling of biopsy samples may help confirm or redefine a diagnosis of GI lymphoma in some cases, but, in others, polymerase chain reaction (PCR) for clonality may also be needed to differentiate between an inflammatory T-cell population and a neoplastic one.^7,20 The purpose of this study was to determine whether Bcl-2 immunolabeling of intestinal biopsies from cats with IBD differed significantly compared with those with gastrointestinal lymphoma. Our hypothesis was that Bcl-2 expression would be more prevalent in gastrointestinal lymphoma specimens than in those from inflammatory bowel disease cases.

Materials and methods

Formalin-fixed, paraffin-embedded tissue blocks were retrieved for this study. All specimens had been previously diagnosed as either IBD or GI lymphoma based on a combination of histopathology [hematoxylin and eosin (H&E)], immunohistochemistry (CD3e and CD79a) and PCR clonality for B- and/or T-cell gene rearrangement (HotStarTaq Master Mix Kit; Qiagen), as described previously. ⁷ These specimens were full thickness or endoscopic small intestinal biopsy samples collected between 2002 and 2005 and submitted to the Diagnostic Center for Population and Animal Health from either the Michigan State University Veterinary Teaching Hospital or two Michigan feline specialty practices. Specimens that had insufficient remaining tissue to perform Bcl-2 immunolabeling were excluded.

Bcl-2 immunohistochemistry

Serial sections from paraffin-embedded tissue blocks were cut as 3 μm sections. These specimens were immunolabeled with a mouse monoclonal anti-human Bcl-2 antibody (Novocastra) at a dilution of 1:100 (Leica automated stainer) following antigen retrieval with a high pH antigen retrieval buffer for 60 mins. The entire tissue section was scanned microscopically and the percentage of infiltrating round cells that were positively labeled for Bcl-2 was subjectively determined and recorded for each case. The degree of labeling present in the epithelium compared with the deeper mucosa was not differentiated. Positive and negative immunohistochemical controls were performed using normal feline lymph node and omission of the primary antibodies, respectively.

Statistical analysis

Student’s t-test was completed for comparative analysis using commercial software (GraphPad 4.1) and significance was set at P = 0.05.

Results

A total of 55 tissue blocks from cats with a clinical history of chronic diarrhea, vomiting and/or weight loss were included in this retrospective study. Samples included full thickness (n = 44) or endoscopic (n = 11) small intestinal biopsy specimens. The endoscopic tissues consisted of mucosa and submucosa.

Based on the previously described algorithm by Kiupel et al, 8/55 cats were diagnosed with IBD and 47 cats were diagnosed with GI lymphoma. ⁷ Forty cats were diagnosed with T-cell lymphoma and seven cats with B-cell lymphoma. Domestic shorthair was the most common breed in each group; no purebred cats were enrolled in the study. Gender distribution between the groups was unremarkable; however, cats with lymphoma were older than IBD cats (mean 12 years vs 7.3 years; P = 0.028). The increased age of cats diagnosed with GI lymphoma compared with IBD in our study suggests the possibility of progression of IBD to lymphoma. Furthermore, a significantly increased percentage of cells were positively immunolabeled for Bcl-2 in cats with GI lymphoma [median (range); 90 (5–95)%] compared with cats with IBD [60 (15–95)%] (P = 0.029). Thus, a small number of GI lymphoma specimens had low levels of Bcl-2 expression whilst the majority of specimens contained a large number of positive cells. T-cell lymphomas in particular were characterized by a high percentage of Bcl-2-labeled cells [90 (40–95)%] (Figure 1). B-cell lymphomas varied widely in the number of positively-labeled cells [90 (5–95)%]; while the median was 90%, this median was generated from only seven samples. Figure 2 demonstrates a case with a histologic appearance of IBD with a low percentage of positively-labeled cells. Eight of the T-cell lymphomas were endoscopic biopsies and three of the IBD cases were endoscopic biopsies. All of the endoscopic biopsies evaluated had at least 50% Bcl-2 labeling.

Figure 1

Feline small intestine sample infiltrated by malignant T-cells. At least 95% of the transmurally infiltrating round cells demonstrated positive red nuclear immunolabeling for Bcl-2

Figure 2

Feline small intestine sample demonstrating inflammatory bowel disease. Approximately 50% of the infiltrating round cells in the mucosal lamina propria displayed red nuclear immunolabeling for Bcl-2. Occasional intraepithelial lymphocytes within the villi also demonstrate positive labeling

Discussion

The oncogene bcl-2 is a key regulator of apoptosis at the level of cytochrome c release. Whilst typically associated with gene rearrangements in human B-cell lymphoma, Bcl-2 family proteins have also been shown to be expressed in benign and malignant cutaneous T-cell infiltrates. Bcl-2 expression has been correlated with the pathogenesis and response to therapy of peripheral T-cell lymphomas in humans.^21,22 Interestingly, when the 40 cats of this study diagnosed with T-cell lymphoma were evaluated separately, the median Bcl-2 expression level appeared higher than for IBD specimens.

A statistically significant difference was noted in the overall percentage of cells that positively immunolabeled for Bcl-2 between cats with GI lymphoma and IBD. However, the overall degree of positive labeling in both groups tended to be high. To be precise, 50% or more of the infiltrating lymphocytes throughout the specimens expressed Bcl-2 in 43/47 (91.5%) of the cats with GI lymphoma and 7/8 (87.5%) of the cats with IBD. The small number of specimens diagnosed with IBD may have altered the significance of our results; further evaluation of Bcl-2 expression in cats with IBD is warranted.

In most cases of human leukemogenesis, up-regulation of bcl-2 gene expression is not sufficient to stimulate tumor expansion in vivo, but disease progression may occur when bcl-2 cooperates with other oncogenes. ¹³ The high expression levels of Bcl-2 in reactive, as well as neoplastic, tissue populations of lymphocytes may be explained by a necessary cooperation of bcl-2 with other oncogenes in the early pathogenesis of lymphoma.

Although it has been speculated that there is a continuum from IBD to GI lymphoma, limited supporting evidence for this phenomenon has been published.^1,23 The increased age of cats diagnosed with GI lymphoma compared with IBD in our study suggests the possibility of progression from IBD to lymphoma. It is unknown if any of the cats with IBD in this study developed lymphoma at a later point in time. However, cats that had been diagnosed with IBD had significantly longer survival times than the cats with lymphoma, which is illustrated by a Kaplan-Meier survival graph in the earlier study. ⁷ A temporal association may exist between the up-regulation of Bcl-2 in IBD to facilitate transition to GI lymphoma. This Bcl-2 up-regulation may be instigated by prolonged, focal, lymphocyte survival and low-grade clonal proliferation.

Five of the seven cats diagnosed with B-cell lymphoma in this study had neoplastic populations consisting of large B-cells in the small intestine. The median Bcl-2 expression in this group was 30%(range 5–95%). Two cats with small cell B-cell lymphoma had Bcl-2 expression in 95% of the infiltrating cells. Expression of Bcl-2 by small, as well as large, cell B-cell lymphomas in this study differs from observations in human lymphomas and may represent a species-specific difference. However, the small number of B-cell lymphomas in this study precludes further conclusions. Owing to the small number of available specimens with B-cell lymphoma, statistical analysis was not performed for these cases. However, a future goal is to assess whether differential expression of Bcl-2 is a feature of small cell versus large B-cell lymphoma in cats.

In diffuse large B-cell lymphoma in human patients, a different prognostic impact is noted for bcl-2 gene rearrangement versus Bcl-2 protein expression. As genetic mutation does not always correlate with protein expression, rearrangements in one study of 156 human patients did not affect survival. However, Bcl-2 protein over-expression in this group was associated with poor disease-free survival and overall survival. ²⁴ Another future goal is to evaluate for Bcl-2 protein expression levels in feline benign and malignant lymphoid infiltrates, and to compare these results with disease-free survival and overall survival data.

Despite the significant difference noted in the percentage of cells that positively immunolabeled for Bcl-2 between cats with GI lymphoma and IBD, the relatively high degree of positive immunolabeling in both groups may prove more important in a clinical, rather than a diagnostic setting. Glucocorticoids are a mainstay of treatment for GI lymphoma and IBD.^25–27 However, the long-term use of glucocorticoids in cats is not without risk of iatrogenic complications.^28,29 If a therapeutic modality could be identified to specifically down-regulate bcl-2 expression or interfere with Bcl-2 activity, relatively lower doses or shorter durations of treatment with glucocorticoids may be necessary for therapeutic management of both diseases. Furthermore, populations of therapy-resistant lymphocytes with over-expression of Bcl-2 could be specifically targeted for molecular intervention. Clinical trials of Bcl-2 inhibitors are being conducted in human oncology and cats may benefit from these new therapeutic advances. ³⁰