Mechanism of radiation-induced bystander effects: a unifying model

Abstract

The radiation-induced bystander effect represents a paradigm shift in our understanding of the radiobiological effects of ionizing radiation, in that extranuclear and extracellular events may also contribute to the final biological consequences of exposure to low doses of radiation. Although radiation-induced bystander effects have been well documented in a variety of biological systems, the mechanism is not known. It is likely that multiple pathways are involved in the bystander phenomenon, and different cell types respond differently to bystander signalling. Using cDNA microarrays, a number of cellular signalling genes, including cyclooxygenase-2 (COX-2), have been shown to be causally linked to the bystander phenomenon. The observation that inhibition of the phosphorylation of extracellular signal-related kinase (ERK) suppressed the bystander response further confirmed the important role of the mitogen-activated protein kinase (MAPK) signalling cascade in the bystander process. Furthermore, cells deficient in mitochondrial DNA showed a significantly reduced response to bystander signalling, suggesting a functional role of mitochondria in the signalling process. Inhibitors of nitric oxide (NO) synthase (NOS) and mitochondrial calcium uptake provided evidence that NO and calcium signalling are part of the signalling cascade. The bystander observations imply that the relevant target for various radiobiological endpoints is larger than an individual cell. A better understanding of the cellular and molecular mechanisms of the bystander phenomenon, together with evidence of their occurrence in-vivo, will allow us to formulate a more accurate model for assessing the health effects of low doses of ionizing radiation.

Figure 1

The unequivocal demonstration of the presence of a bystander mutagenic effect in the human hamster hybrid (A_L) cells using a charged particle microbeam in which 20% of the cells were lethally irradiated with 20 α-particles each through the nucleus. Assuming that there was no interaction between the irradiated-but-dead cells and the non-irradiated cells, the resultant mutant fraction from the irradiated population should be similar to the spontaneous or background level. In reality, as a result of the cross-talk between the hit and non-hit cells, the resultant mutant fraction in the irradiated population was three-fold higher than the background yield (refer to Zhou et al 2000 for detail).

Figure 2

The various approaches used to demonstrate bystander endpoints using medium-transfer experiments. The top panel illustrates track segment irradiation at the Radiological Research Accelerator Facilities of Columbia University (www.raraf.org). Cells plated on stainless steel rings with a 6 μm mylar bottom were exposed to a parallel broad-area beam of monoenergetic helium-3 ions. The ions deposited a fraction of their energy in the target cells and the average linear energy transfer over the segment of the track that penetrated the target was around 90 keVμm⁻¹. The middle panel shows a ‘double mylar’ design (Geard et al 2002; Zhou et al 2002) in which irradiated cells plated on the lower mylar bottom secreted soluble mediators into the medium, which modulate the biological behaviour of the non-irradiated cells plated on the upper panel, α-particles having little penetrance. The bottom panel shows the classic medium-transfer experiment as described by Mothersill & Seymour (1997). γH2AX, phosphorylated histone H2AX (a protein often associated with DNA double-strand breaks).

Figure 3

The use of a novel ‘double mylar’ design in which both directly irradiated and bystander cells were cultured on the same vessel. A pathway-specific array was used to identify differentially expressed genes among either non-irradiated or bystander fractions of the double mylar cultures. Two genes were found to be differentially expressed in the bystander cells: a three-fold increase in cyclooxygenase (COX)-2 and a seven-fold decrease in insulin growth factor binding protein-3 (IGFBP-3). To illustrate the functional correlation of COX-2 over-expression with bystander mutagenesis, cultures were concurrently exposed to NS398, a COX-2 inhibitor, which suppressed the bystander mutagenic response (refer to Zhou et al 2005 for detail). HPRT, hypoxanthine guanine phosphoribosyltransferase.

Figure 4

A unifying model of the signalling pathways involved in radiation-induced bystander effects. Expression/secretion of the inflammatory cytokines strongly increased after exposure to ionizing radiation or oxidants. Secreted or membrane-associated forms of cytokines such as tumour necrosis factor (TNF)α activate IκB kinase (IKK)-mediated phosphorylation of IκB, which releases nuclear factor (NF)-κB. NF-κB enters the nucleus and acts as a transcription factor for cyclooxygense-2 (COX-2) and inducible nitric oxide (NO) synthase (iNOS) genes. TNFα also activates mitogen-activated protein kinase (MAPK) pathways (extracellular signal-related kinase (ERK), c-Jun N-terminal kinase (JNK) and p38) that, via the activation protein (AP)-1 transcription factor, additionally up-regulate expression of COX-2 (Zhou et al 2005) and iNOS, which stimulates production of NO. Mitochondrial damage facilities the production of hydrogen peroxide, which migrates freely across plasma membranes and is subjected to antioxidant removal. Activation of COX-2 provides a continuous supply of reactive radicals and cytokines for the propagation of bystander signals through either gap junctions or medium. H₂O₂, hydrogen peroxide; IL, interleukin; OH^•, hydroxyl radicals; ONOO^•, peroxynitrite anions; PTIO, 2-4-(-carboxyphenyl-4,4,5,5,-tetramethylimi-dazoline-1-oxyl-3-oxide (an NO scavenger); -R, receptor; PG-E2, prostaglandin E2; ROS, reactive oxygen species; TGF, transforming growth factor, TNF, tumour necrosis factor; TRAIL, TNF-related apoptosis-inducing ligand.