Response to Letter by Baykal
We appreciate Dr Baykal’s interest in our work and would like to respond to his comments.
We are well aware of the BrdU-labeling paradigm and have published more than 30 research articles using this method and several methodological articles analyzing this technique in detail.1–3 Dr Baykal must have been under the impression that we have labeled with only a single BrdU injection. However, we have used 5 BrdU injections during the first 5 days after ischemia as stated in our publication: “Animals were treated with 20 μg BDNF i.v. or vehicle 1 hour after induction of ischemia. Treatment was repeated on day 2 to 5. Four hours after treatment, dividing cells were labeled with i.p. bromodeoxyuridine (BrdU, 50 mg/kg/d).”
It is our impression that Dr Baykal further draws an incorrect conclusion from the results of our study as indicated by his comment: “Since the doublecortin positive cells in this study are not positively labeled for BrdU, than these cells were present prior to stroke and thus it can be hypothethised that a pool of doublecortin positive cells reside within the brain to migrate and mature as a response to neuronal loss.” We believe this is an incorrect interpretation of our results. First of all, DCX positive cells were not positive for BrdU, because BrdU was given about 6 weeks before perfusion, whereas the phase during which differentiating progenitor cells express DCX is only about 2 weeks.4 Therefore, the time window chosen in this experiment is simply not designed to detect BrdU/DCX double-labeling but rather BrdU colabeling with a mature neuronal marker, such as NeuN.
If we understand correctly, Dr Baykal’s main point is that he is surprised we did not find BrdU-labeled neurons in the neocortex after photothrombotic stroke lesion and suggest that more frequent BrdU-labeling should be required to show cortical neurogenesis. Although it is correct that we did not find BrdU-labeled cortical neurons, we observed new neurons (BrdU/NeuN) in the dentate gyrus, thus proving that we used sufficient amounts of BrdU. Using higher doses of BrdU or more frequent BrdU injections can lead to neuronal toxicity and to possible false-positive–labeling due to incorporation during DNA-repair. We therefore refrained from increased labeling for methodological clarity.
The generation of new neurons in the adult neocortex is a very intensely debated field. We agree with Dr Baykal that the high number of DCX-positive cells in the neocortex without BrdU-NeuN double-labeling is a surprising finding. But it points to a possible transient existence of neural progenitor cells in the lesioned brain area. Additional trophic signals may be required to further differentiate these progenitor cells into functional neurons.
In general, we are not against the possibility that cortical neurogenesis is inducible by lesion and our study was actually designed to detect cortical neurogenesis after stroke. It is possible that, very early after ischemia, reactive cortical neurogenesis is repressed by inflammatory responses but that much later cortical neurogenesis becomes permissive. This scenario, however, needs a completely different study design, with separate groups of animals receiving BrdU at different time points and much longer survival times for the animals. This we have commented on in our discussion: “Among the possible reasons for not detecting cortical neurogenesis in our experiments are … (ii) the possibility that the ventricle wall produces new neurons for the cortex not during the first 5 days after lesion, but rather at a later time interval.”
Cooper-Kuhn CM, Kuhn HG. Is it all DNA repair? Methodological considerations for detecting neurogenesis in the adult brain. Brain Res Dev Brain Re. 2002; 134: 13–21.
Kuhn HG, Peterson DA. Detection and phenotypic characterization of adult neurogenesis. In: Gage FH, Kempermann G, Song H, eds. Adult Neurogenesis. Woodbury, NY: Cold Spring Harbor Laboratory Press: 2007; 25–47.