Warning! You won't be able to use the quotation basket until you enable cookies in your Web browser.
Warning! Your Web browser is no longer supported. Please upgrade to a modern browser.

The Utility & Assessment of RNA ISH Probes for EBV, Kappa and Lambda on BOND™

Epstein-Barr virus (EBV) is a ubiquitous herpesvirus that infects and establishes persistent infection in a host. Latent EBV infection has been linked to a wide range of benign and malignant lesions. This study compares the sensitivity of automated IHC and ISH staining to detect latent EBV infection.

Introduction

Epstein-Barr virus (EBV) was first discovered in 1964 and is a ubiquitous herpesvirus that infects and establishes persistent infection in a host. Initial contact with EBV usually occurs in the first 10 years of life and results in asymptomatic infection. Carriers secrete the virus in their saliva, allowing transmission through the oral route. Clinical implications of primary infection in adolescence or adulthood can range from a mild self-limited infection in children to infectious mononucleosis in adolescents and adults. After active infection the virus is thought to reside in a latent form with B cells providing the main cellular reservoir.1 This latent infection has been linked to a wide range of benign and malignant lesions such as angioimmunoblastic lymphoma2, PTLD3, nasopharyngeal carcinoma4 and Burkitt’s lymphoma5 amongst others.

Latent infection by EBV can occur in three forms; latency I, latency II and latency III. Each latency form is marked by a differing viral gene expression profile. The expression profiles can be seen in Table 1.

Routine detection of EBV in formalin fixed, paraffin embedded tissue is achieved at Guy’s and St Thomas’ NHS Foundation Trust (GSTT) by the use of immunohistochemistry using a primary antibody directed against latent membrane protein 1 (LMP-1). However as can be seen in Table 1 not all stages of EBV latency express LMP-1 and questions have been raised about the sensitivity of the technique.

In situ hybridisation (ISH) detection of EBV using probes directed against Epstein-Barr encoded RNA (EBER) has been in use experimentally for approximately 16 years and more recently has seen widespread use diagnostically. EBER actually consists of two small EBV latency transcripts of 166 and 172 bases respectively, called EBER-1 and EBER-2.6 These transcripts however are non-polyadenlylated and therefore not translated into proteins and cannot be detected by IHC. They are naturally amplified and present in all forms of latent infection at high levels, making them ideal targets for ISH, which is widely considered the gold standard for the detection of EBV latent infection in paraffin embedded, formalin fixed tissue.7

Latency 1

Latency 2

Latency 3

EBER

EBER

EBER

EBNA-1

EBNA-1

EBNA-LP

 

LMP-1

EBNA-1

 

LMP-2

EBNA-2

 

 

EBNA- 3a

 

 

EBNA- 3b

 

 

EBNA-3c

 

 

LMP-1

 

 

LMP-2

Key: EBNA= Epstein- Barr Nuclear Antigen, LMP= Latent Membrane Protein

Table 1. Viral genes expressed in different latent infection programs of EBV.

Kappa and Lambda

The demonstration of kappa and lambda light chains in formalin fixed paraffin embedded tissue can be used to assess clonality of a lesion of B cell origin. That is, are all the cells from a single clone (malignant) or from multiple clones (benign). In a normal population of B cells there will be a mixture of cells either expressing kappa light chains or lambda light chains on their surface. The normal ratio of B cells expressing kappa to those expressing lambda is reported to be anywhere between 2:1 and 6:1.8 Reactive lymphoid lesions contain a mixture of cells; the B cells within the legion express Kappa and Lambda in a normal ratio (polyclonal). B-lymphoid neoplasms however, consist of expansions of a single clone of cells which express only one type of light chain (monoclonal). If this lesion is then stained for kappa and lambda then the expected ratio will be shifted one way or the other. This lesion is then termed monoclonal and is neoplastic in nature. The situation is slightly different in plasma cells which are cells that are actively producing light chains within their cytoplasm. However each plasma cell will only produce either kappa or lambda light chains in a ratio of 2:19, so staining for kappa and lambda light chains can still demonstrate clonality, however the staining pattern will be cytoplasmic, not perinuclear membranous.

Routinely at GSTT demonstration of kappa and lambda light chains in formalin fixed paraffin embedded tissue is again achieved by the use of IHC. However cross-reaction of the primary antibodies with other immunoglobulins in the sample can cause problems of high non-specific background staining. This is a particular problem in bone marrow trephine (BMT) specimens. Bone marrow trephine biopsy is a technically diff-icult and traumatic procedure and usually results in the release of serum into the interstitial spaces between cells. The serum immunoglobulins then cross react with the primary antibodies directed against kappa and lambda causing very high back-ground. This can result in both positive and negative staining being masked by the high background, causing difficulty in interpretation and considerable pathologist frustration. In the worst cases, repeats of the IHC staining are ordered by the pathologist, that may cause 24 or more hours delay in diagnosis. The problems are often exacerbated in outside referral cases where the fixation and processing techniques can vary widely, leading to suboptimal immunohistochemistry. High levels of background can also affect the objectivity of the reporting of cases. Reducing background and therefore improving staining quality should reduce this inter-observer variability.

In situ hybridisation using probes against kappa and lambda mRNA can also be used to demonstrate clonality. It has been shown to have much lower levels of background staining and higher sensitivity than IHC.10,11 However ISH detection mRNA is only likely to be diagnostically useful when high levels of kappa or lambda mRNA are present in the cell. Mature B cells do not actively produce kappa and lambda light chains, they are only expressed as proteins on the cells surface, therefore will not stain with ISH. However B cells which terminally differentiate into plasma cells which actively produce kappa and lambda will be detected by ISH, as high levels of kappa and lambda mRNA will be present in these cells.

Aims

  • To compare the sensitivity of an automated ISH system and an automated IHC system for the detection of EBV latent infection in formalin fixed paraffin embedded tissue.
  • Within EBV positive cases, to assess the number of EBV positive cells detected by ISH and IHC.
  • To examine the reduction in background staining by using ISH for kappa and lambda staining in assessing B cell clonality in comparison to IHC.

Method

All tests were carried out on the Bond system (Leica Microsystems). All tissue sections were cut at 3 microns. For the IHC positive and negative control sections were also used. For the ISH mRNA negative and positive probes were also run to ensure the validity of the technique. Tissue blocks were selected for the study from the diagnostic archive at GSTT to reflect the types of cases testing for EBV, kappa and lambda were normally carried out upon. The study was approved by the Trust’s ethics and R&D committees.

Results Assessment

EBV

Firstly the slides were assessed either positive or negative for EBV latent infection within the cells of the lesion of interest (i.e. any positive cells not involved in the lesion were not taken into account and put down to background ubiquitous infection of the population with EBV). The cases that were assessed positive were then scored on a scale of 1+ to 3+ dependent upon the number of positive cells, with 1+ being 3 10% of cells positive, 2+ 3 25% of cells positive and 3+ > 25% of cells positive.

Kappa and Lambda

The slides were scored as to the amount of unwanted background staining 0=no/negligible non-specific staining, 1+ being peripheral non-specific staining only, 2+ being light interstitial non- specific staining and 3+ being heavy interstitial non-specific staining. Whether the staining was uniform or variable was also noted.

Results

EBV

Overall 18/54 (33.3%) of cases were positive for EBV by IHC, whereas 28/54 (51.85%) were positive by ISH. This was found to be significantly different at p=0.05 using chi-squared (X2=8.33). The ratio of numbers of positive cells was also seen to be significantly different by chi-squared (X2=42.48), a X2 which is even significantly different at p= 0.001. The figures can be seen in Table 2.

Kappa and Lambda

In the kappa and lambda arm of the study the slides were scored on the amount of background encountered on a scale of 0 to 3+. With 0 = no/negligible non-specific staining; 1+ = peripheral non-specific staining only; 2+ = light interstitial non-specific staining and 3+ = heavy interstitial non-specific staining.

In this study no background was observed in any of the 46 cases stained for Kappa light chain mRNA and only one slide stained for Lambda light chain showed background, which consisted of a granular artefact.

Testing Method

IHC

ISH

EBV Negative

36

26

EBV Positive (Total)

18

28

EBV Positive 1+

12

7

EBV Positive 2+

3

10

EBV Positive 3+

3

11

Total

54

54

Table 2. Numbers of cells EBV positive by IHC and ISH.

Conclusion/Discussion

Kappa and Lambda

The main aim of the study was to assess the amount of back-ground staining encountered when using ISH to detect kappa and lambda light chain mRNA in comparison to the high levels of non specific background staining encountered using IHC to detect kappa and lambda light chains. The results confirm that kappa and lambda light chain mRNA probes, as expected, have no cross reactivity with serum immunoglobulins known to cause such a problem with high background BMTs especially. In this study 20/46 (43.5%) cases scored 2+ (light interstitial non-specific staining) or 3+ background (heavy interstitial non-specific staining) with IHC for both kappa and lambda light chains with IHC. These figures reflect the problems generally encountered with high background staining using IHC for kappa and lambda light chains and would contribute to non-objective reporting and could cause high inter-observer variability.

 

However ISH for kappa and light chain mRNA does not detect the kappa and lambda protein expressed on the surface of B cells in most stages of their maturation. Only when the B cells terminally differentiate into plasma cells is ISH for kappa and lambda light chain mRNA able to demonstrate clonality.

EBV

Overall in this study automated ISH using probes for EBER on the Bond-max system proved to be more sensitive for the detection of latent EBV infection in formalin fixed paraffin embedded tissue. This result can be explained best by considering the different programs of gene expression in latent EBV infection. As seen in Table 1, latent EBV infection has three different programs that differ according to which viral genes (and therefore proteins) are expressed by the infected cell. EBER transcripts are expressed in all latency programs, whereas LMP-1 is only expressed in latency II and III. This would explain why ISH for EBER is more sensitive for the detection of EBV because it will detect cells with all programs of latent infection where IHC for LMP-1 will only detect cells with latency programs II and III infection.

However, as the investigation into the role of virally encoded genes in lymphoma genesis continues and treatments are directed against the proteins expressed it may be necessary to perform IHC in the future to assess more closely the latency programs present in a given lesion.

Figures

Figure A. Shows an example of a BMT case stained for Kappa light chains by IHC at 10x power. Note the mild diffuse background staining encountered.
An example of a BMT case stained for Kappa light chains by IHC at 10x power. Note the mild diffuse background staining encountered.
Figure B. Shows the same case stained for Kappa light chain mRNA using ISH, none of the background staining seen before is noted, leaving clear demonstration of plasma cells only.
The same case stained for Kappa light chain mRNA using ISH, none of the background staining seen before is noted, leaving clear demonstration of plasma cells only.
Figure C. Shows another BMT case stained for Kappa light chains by IHC at 2x showing extensive edge artefact.
Another BMT case stained for Kappa light chains by IHC at 2x showing extensive edge artefact.
Figure D. The same case stained for Kappa light chain mRNA by ISH shows no such edge artefact.
The same case stained for Kappa light chain mRNA by ISH shows no such edge artefact.
Figure E. Shows excellent demonstration of EBV infected cells by the detection of EBER by ISH (10x mag.).
An excellent demonstration of EBV infected cells by the detection of EBER by ISH (10x mag.).
Figure F. Shows the same case stained for EBV by IHC for LMP. The number of positive cells is less, as is the intensity of the staining.
The same case stained for EBV by IHC for LMP. The number of positive cells is less, as is the intensity of the staining.

References

  1. Khan G, Miyashita EM, Yang B, Babcock GJ, Thorley-Lawson DA. Is EBV persistence in vivo a model for B cell homeostasis? Immunity. 1996;5:173-9.
  2. Zhou Y, Attygalle AD, Chuang S, Dis T, Ye H, Liu H et al. Angioimmunoblastic T-cell lymphoma: histological progression associates with EBV and HHV6B viral load. Br J Haematol. 2007;138:44-53.
  3. Lim WH, Russ GR, Coates PT. Review of Epstein-Barr virus and post-transplant lymphoproliferative disorder post-solid organ transplantation. Nephrology (Carlton). 2006;11(4):355-66.
  4. Takeuchi H, Kobayashi R, Hasegawa M, Hirai K. Detection of latent Epstein-Barr virus (EBV) DNA in paraffin section of nasopharyngeal carcinomas expressing no EBV-encoded small RNAs using in situ PCR. Arch Virol. 1997;142:1743-56.
  5. Kelly G, Bell A, Rickinson A. Epstein-Barr virus associated Burkitt lymphomagenesis selects for down regulation of the nuclear antigen EBNA-2. Nat Med. 2002;8:1098-1104.
  6. MacMahon E M E and Ambinder R F. EBER in situ hybridisation: Sensitive detection of latent Epstein-Barr virus in individual cells. Rev Med Virol. 1994;4:251-260.
  7. Gulley M.L. Molecular Diagnosis of Epstein-Barr Virus-Related Diseases. J Mol Diagn. 2001;3(1):1-9.
  8. Chizuka A, Kanda Y, Nannya Y, Oshima K, Kaneko M, Yamamoto R, et al. The diagnostic value of kappa/lambda ratios determined by flow cytometric analysis of biopsy specimens in B-cell lymphoma. Clin Lab Haematol. 2002;24: 33-6.
  9. Magro C, Crowson A.N, Porcu P, Nuovo G.J. Automated kappa and lambda light chain mRNA expression for the assessement of B-cell clonality in cutaneous B-cell infiltrates: its utility and diagnostic application. J Cutan Pathol. 2003;30(8):504-11.
  10. Pringle J H, Ruprai A K, Primrose L. In situ hydridiazation of immunoglobulin light chain mRNA in paraffin sections using biotinylated or hapten-labelled oligonucleotide probes. J Pathol. 1990;162:197-207.
  11. Sumiyoshi Y, Kikuchi M, Yamashita Y, Maekawa T, Shirakusa T. Analysis of immunoglobulin light chain mRNA in gastric malignant lymphoma using new highly sensitive in situ hybridisation method. Gastric Cancer . 1999;2:8-13.