Merged image of Elk-1 (red) and
MAP-2 (green) immunoreactivity
in rat hippocampal neurons.
Overlapping expression is
indicated in yellow.
Being in the right place at the right time is clearly important. However, traditional gene transfer techniques offer little control of the localization of transfected agents. Barrett et al. describe a method of transfecting discrete regions of the neuron using a laser to permeabilize focally restricted areas, in a recent article in Nature Methods.
The authors proposed that bath-applied mRNA would enter permeabilized regions of neurons by diffusion. They bathed dissociated hippocampal neurons in buffer containing Lucifer yellow dye and permeabilized the cell with several 5-ms pulses from a titanium-sapphire laser. Dye accumulated inside the cell within milliseconds. The authors next bathed hippocampal neurons with buffer containing green fluorescent protein (GFP) mRNA, and they targeted the laser to a 0.16 
m2 region of a single dendrite. Within 30 min, the authors observed GFP fluorescence in the dendrite and cell body. GFP fluorescence increased with time, but was never detected in neuronal processes not specifically targeted by the laser. Transfected neurons remained viable for at least 24 h. The authors named their transfection technique 'phototransfection.'
The authors used this technique to characterize the region-specific effects of the transcription factor Elk-1 within the neuron. In situ hybridization showed Elk-1 mRNA expression in the dendrites as well as the cell bodies of cultured hippocampal neurons. The authors phototransfected Elk-1 mRNA at dendritic sites more than 40 m from the cell body. Most of the phototransfected neurons showed signs of cell death, including somal swelling, within 60 min of phototransfection. Enzymatic assays confirmed cytotoxicity in most cells dendritically phototransfected with Elk-1 mRNA. In contrast, none of the neurons dendritically phototransfected with mRNAs encoding DS-RED, GFP, c-fos or Xenopus elongation factor 1
showed signs of cell death.
In contrast to Elk-1 phototransfection in the dendrites, Elk-1 phototransfection in the cell body did not induce cell death, suggesting that dendritic Elk-1 is functionally distinct from somal Elk-1. The authors preincubated neurons with the translation inhibitor anisomycin before dendritic phototransfection of Elk-1 and found no signs of cell death, suggesting that local translation of Elk-1 in the dendrite is required for this effect. Pre-incubation with the transcription inhibitor actinomycin D and removal of the Elk-1 DNA binding domain also prevented Elk-1-induced cell death, implicating the genes transcriptionally induced by Elk-1 in the dendrite in mediating cell death.
The authors eventually observed GFP fluorescence throughout the cell body, including the nucleus, following dendritic phototransfection of constructs containing GFP mRNA. Therefore, mRNA locally transfected in neural processes is likely retrogradely transported, which may limit the ability to differentiate regionally specific functions of a molecule over an extended period of time. However, phototransfection may be very helpful in determining the role of fast-acting signaling molecules at distal regions of neuronal processes, such as the growth cone.