Keratinocyte apoptosis following bone marrow transplantation: evidence for CTL-dependent and -independent pathways

Abstract

Summary:Acute graft-versus-host disease (GVHD) is a complication of bone marrow transplantation (BMT). The histopathologic features used to diagnose GVHD are nonspecific, and may be secondary to chemotherapy or irradiation given before BMT. The presence of apoptotic keratinocytes or activated CTL may distinguish GVHD from conditioning effects. This study investigated the relationship in BMT recipients between keratinocyte apoptosis and the effects of conditioning regimens or immune-mediated GVHD. Inflammatory cells, apoptotic keratinocytes, and CTL expressing TIA-1 (a molecule associated with the lytic granules of CTL) were quantitated in allogeneic and autologous recipients. Allogeneic recipients could exhibit keratinocyte apoptosis secondary to a combination of conditioning effects and immune-mediated GVHD. In contrast, autologous recipients should show conditioning effects only. 'Capped' TIA-1-positive lymphocytes and apoptotic keratinocytes were much more frequent in the allogeneic group than the autologous group (16.1% of total TIA-1 positive lymphocytes vs 4.5%, P = 0.02; and 37.6/mm 2 vs 3.9/mm 2 , P = 0.005, respectively), although there was some overlap in their frequency. Among individual recipients of allogeneic BMT, the number of epidermal lymphocytes or macrophages correlated with the number of apoptotic keratinocytes. A similar, but weaker, correlation was seen between the number of 'capped' TIA-1-positive lymphocytes and apoptotic keratinocytes. No such relationship was seen in autologous recipients. In allogeneic recipients, TIA-1 expressing CTL were seen in intimate contact with apoptotic keratinocytes, some of which also had detectable cytoplasmic TIA-1. No CTL/keratinocyte interactions were identified in autologous recipients. Our results suggest that apoptotic keratinocytes arise in the skin of BMT patients due to both GVHD and conditioning effects, and that the keratinocyte damage in GVHD is mediated by both CTL-dependent and -independent mechanisms. Increased numbers of apoptotic keratinocytes, in the presence of increased epidermal lymphocytes or 'capped' TIA-1-expressing lymphocytes, support a diagnosis

The Tunel Assay in the Diagnosis of Graft-Versus-Host Disease: Caveats for Interpretation

“…Although apoptotic keratinocytes in skin biopsies can be identified by morphology without TUNEL staining, in our hands the TUNEL assay increased the number of apoptotic cells identified in a given tissue section by two-to three-fold in a blinded study. 1 However, while use of the TUNEL assay may increase the sensitivity of detection of apoptotic keratinocytes, the clinical significance of this remains to be established. By accepting as true-positives only TUNEL-positive cells with apoptotic morphology, it is likely that we are missing cells in the earlier stages of apoptosis.…”

“…We and others have shown that the presence of such bodies, together with other findings such as an activated lymphomonocytic infiltrate, supports the diagnosis of GVHD. [1][2][3] With the recent realisation that these dyskeratotic bodies actually represent apoptotic keratinocytes, 4,5 attention has focused on methods such as the TUNEL reaction (terminal deoxynucleotidyl transferase (TdT)-mediated dUTP-biotin nick end labeling) which in theory might allow more sensitive detection of apoptotic cells in the skin. However, our experience with the TUNEL assay suggests that there are a number of commonly occurring artifacts that can lead to potential misinterpretations in the evaluation of skin biopsies for GVHD.…”

The tunel assay in the diagnosis of graft-versus-host disease: caveats for interpretation

“…[1][2][3] With the recent realisation that these dyskeratotic bodies actually represent apoptotic keratinocytes, 4,5 attention has focused on methods such as the TUNEL reaction (terminal deoxynucleotidyl transferase (TdT)-mediated dUTP-biotin nick end labeling) which in theory might allow more sensitive detection of apoptotic cells in the skin. Unfortunately, the histopathological features of GVHD overlap with other skin pathology frequently seen in BMT recipients, such as drug reactions or alterations secondary to pre-transplant conditioning regimens.…”

“…2C-E). 1 However, while use of the TUNEL assay may increase the sensitivity of detection of apoptotic keratinocytes, the clinical significance of this remains to be established. The high incidence of TUNEL positivity among stromal cells in this system supports the artifactual nature of the staining near borders.…”

The detection of apoptosis in a human in vitro skin explant assay for graft versus host reactions

“…7 In addition to TNF-α, other soluble and membrane bound mediators, such as Fas ligand (FasL), can confer apoptotic cell death. 8 Recent data also strongly suggest a target organ specificity of GVHD related tissue damage, 9 and that the main target cells for GVHD in the skin are keratinocytes. The basis of our investigation was to correlate keratinocyte apoptosis with the severity of graft versus host reactions (GVHR) in skin sections generated from an in vitro skin explant model used as an experimental model to study GVHR, and to predict GVHD in allogeneic stem cell transplants 10 (for a recent review see Sviland and Dickinson 11 ).…”