Edited by: Allan V. Kalueff, ZENEREI Institute (USA), Guangdong Ocean University (China), St Petersburg State University (Russia), USA
Reviewed by: Tom V. Smulders, Newcastle University, UK; Lara LaDage, Penn State University, Altoona Campus, USA
*Correspondence: Hans-Peter Lipp
This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution and reproduction in other forums is permitted, provided the original author(s) or licensor are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.
Many birds are supreme long-distance navigators that develop their navigational ability in the first months after fledgling but update the memorized environmental information needed for navigation also later in life. We studied the extent of juvenile and adult neurogenesis that could provide such age-related plasticity in brain regions known to mediate different mechanisms of pigeon homing: the olfactory bulb (OB), and the triangular area of the hippocampal formation (HP tr). Newly generated neurons (visualized by doublecortin, DCX) and mature neurons were counted stereologically in 35 pigeon brains ranging from 1 to 168 months of age. At the age of 1 month, both areas showed maximal proportions of DCX positive neurons, which rapidly declined during the first year of life. In the OB, the number of DCX-positive periglomerular neurons declined further over time, but the number of mature periglomerular cells appeared unchanged. In the hippocampus, the proportion of DCX-positive neurons showed a similar decline yet to a lesser extent. Remarkably, in the triangular area of the hippocampus, the oldest birds showed nearly twice the number of neurons as compared to young adult pigeons, suggesting that adult born neurons in these regions expanded the local circuitry even in aged birds. This increase might reflect navigational experience and, possibly, expanded spatial memory. On the other hand, the decrease of juvenile neurons in the aging OB without adding new circuitry might be related to the improved attachment to the loft characterizing adult and old pigeons.
Most avian long-distance navigators return to the site of hatching even after long periods (Bonadonna et al.,
The experimental analysis of brain and sensory mechanisms has been predominantly done in homing pigeons (
Lesion or temporary inactivation studies in homing pigeons have focused primarily on the olfactory system. The main techniques included temporary inactivation of the olfactory mucosa or sectioning the olfactory nerves (Wallraff,
Since the age of homing pigeons is usually well documented by means of breeder records and foot rings, it is possible to undertake cross-sectional studies aimed at detecting age-dependent changes in neuronal plasticity in those structures. Postnatal and adult neurogenesis has become a prominent marker for identifying plasticity processes in particular brain regions of mammals and also birds. In mammals, adult hippocampal neurogenesis by progenitor granule cells in the dentate gyrus peaks at juvenile periods and decreases strongly thereafter (Amrein and Lipp,
In birds, adult neurogenesis is more widespread than in mammals, new cells being generated in the ventricular zone appearing not only in the medial (parolfactory lobe) and lateral striatum, the OB and the hippocampal complex but also in a variety of forebrain structures such as the hyperpallium and nidopallium caudolaterale, the dorso-lateral corticoid and various song control nuclei (Vellema et al.,
Age-dependent differences in adult neurogenesis have been observed sporadically in the canary brain (Alvarez-Buylla et al.,
We focused on determining the numbers of newly-born vs. mature periglomerular neurons in the OB, since doublecortin (DCX) immunoreactivity is very sparse and weak in the granule cell layer of the OB in pigeon, and our earlier trials with BrdU injections (unpublished data) labeled much larger numbers of periglomerular cells than granule cells.
This study was carried out under the license 28/2012 of the Veterinary Office of the Canton of Zürich in accordance with the Swiss regulations for use of experimental animals.
The brains of 35 pigeons ranging in age from 1 to 168 months were investigated. Young and young adult pigeons were obtained from the loft of one of the authors in Switzerland (H-PL), while older and aged birds were purchased from breeders in Switzerland, Italy and Germany. The latter were probably not the best racing pigeons and have thus no record of performance, but had returned reliably from races or training releases during several years while other birds from the same loft were lost over time. Sex was determined by inspecting gonads at the end of dissection. In juvenile 1–2 month-old birds, sex was not determined due to immature glands. Therefore, an analysis of potential sex differences did not seem meaningful. To minimize seasonal variations on neurogenesis and anatomical size, all pigeons were sacrificed during a narrow annual time-window from mid-November to mid-December.
Each pigeon was deeply anesthetized with pentobarbital (200 mg/kg) and perfused transcardially with a fixative consisting of 4% paraformaldehyde in 0.1 M phosphate buffer pH 7.4. Brain weight was determined in 25 birds, one age class (80 months) had no information because of lost records. All brains were postfixed for 4 h in the same fixative and sectioned at midline into left and right parts. One hemisphere was dehydrated, embedded into Technovit-resin, and serially sectioned through at 20 μm. The obtained sections were stained with Giemsa and used to quantify the general number of neurons and volumes of areas of interest.
The other hemisphere was equilibrated in 10–30% sucrose and serially cryosectioned in the coronal plane. Forty-micron-thick sections were collected in eight series and stored in cryoprotecting solution at −20°C. Regularly spaced sections (320 μm) were used for immunohistochemistry of DCX. Whenever possible, samples from different birds and age classes were processed simultaneously to avoid batch-dependent differences in intensity of immunostaining. For immunostaining, sections were pretreated for 20 min with 1% hydrogen peroxide. Then they were preincubated in 10% normal rabbit serum in tris-buffered saline, containing 0.2% Triton X-100, and incubated for 36 h in goat polyclonal antibody to DCX. The secondary antibody, biotinylated rabbit anti-goat IgG (1:250; Vector Labs), and Vectastain ABC Elite reagent (1:100; Vector Labs) were applied for 2 and 1 h, respectively. Standard 3,3’-diaminobenzidine reaction was used to visualize DCX positive cells. Sections were mounted onto slides, air dried, and coverslipped without counterstaining. Since the proliferation marker most commonly used in mammals, Ki67, works poorly in birds, we also immunostained corresponding sections for the proliferation marker proliferating cell nuclear antigen (PCNA) and for a second marker for young neurons (PSA-NCAM). However, the results for these two markers were extremely variable, indicating technical problems in the adaptation of the staining procedure. Therefore, we present only the data for Giemsa and DCX stains, characterized by reasonable inter-individual variability of staining.
Within the OB, DCX-positive cells and Giemsa-stained neurons were quantified in the periglomerular cell layer. Volume measurements included the entire OB (Figure
All quantifications were performed using unbiased stereological quantification on Zeiss Imager M2, equipped with a high-resolution video camera CX 900 (mbf Bioscience) and the software package StereoInvestigator 9.0–10 (mbf Bioscience).
The number of adult (Giemsa-stained) neurons in the periglomerular layer of the OB was sampled using an optical fractionator from every 6th section (120 μm interval), which resulted on average in 13 sections per single bulb and coefficients of error (Gundersen)
The numbers of DCX positive cells were estimated using an optical fractionator on every 8th (320 μm) or every 16th (640 μm) section for OB and HP tr, respectively. This scheme resulted in measuring 4–5 sections,
The volumes for the whole OB and HP tr were measured using Cavalieri estimator; every 6th section,
Neuron numbers in the regions of interest were linearly regressed on age (Bates et al.,
The effect of age on number of DCX stained cells and %DCX (number of DCX positive cells divided by total cell number) exhibited a significant quadratic term for both regions (
Effects were considered significant at
Table
Age class | g bodyweight | g brain weight | Neurons HP tr | Neurons OB pgl | DCX HP tr | DCX OB pgl | Vol HP tr mm3 | Vol OB mm3 | |
---|---|---|---|---|---|---|---|---|---|
1 month | 6 | 355 ± 22 | 1.65 ± 0.06 | 739,313 ± 38,768 | 47,642 ± 2263 | 20,513 ± 1359 | 1039 ± 121 | 7.972 ± 0.206 | 2.224 ± 0.396 |
2 months | 5 | 438 ± 12 | 1.89 ± 0.03 | 772,745 ± 27,148 | 59,515 ± 7338 | 19,845 ± 3502 | 862 ± 192 | 8.258 ± 0.235 | 2.566 ± 0.161 |
6 months | 5 | 433 ± 15 | 2.06 ± 0.04 | 798,036 ± 61,863 | 57,801 ± 2896 | 16,650 ± 3330 | 643 ± 97 | 8.470 ± 0.253 | 3.011 ± 0.156 |
12 months | 6 | 475 ± 21 | 2.17 ± 0.07 | 975,437 ± 64,558 | 35,609 ± 2917 | 13,350 ± 2903 | 416 ± 115 | 8.990 ± 0.380 | 2.614 ± 0.138 |
2–3 years | 4 | 469 ± 26 | 2.09 ± 0.07 | 921,604 ± 46,647 | 50,918 ± 6717 | 13,444 ± 3604 | 218 ± 97 | 8.747 ± 0.318 | 2.392 ± 0.178 |
6–7 years | 6 | 1,354,974 ± 86,519 | 34,870 ± 4598 | 10,688 ± 1110 | 151 ± 25 | 9.785 ± 0.379 | 2.737 ± 0.150 | ||
12–14 years | 3 | 482 ± 13 | 2.19 ± 0.07 | 1,682,386 ± 138,693 | 36,879 ± 2844 | 11,050 ± 1537 | 53 ± 26 | 9.920 ± 0.578 | 2.666 ± 0.190 |
In the HP tr formation, we observed a massive increase in cell numbers across age levels. From an average level of about 740,000 neurons in 1-month old birds, the numbers appeared doubled in old adult birds and peaked in the oldest pigeon (14 years) at 1.9 million cells (Table
The number of newly born DCX-positive neurons in the HP tr declined from about 20000 in 1-month old pigeons to 13000 cells in 12-month old birds, leveling off to about 11,000 cells in the oldest birds (Table
In the periglomerular layer, cell numbers are high during the first 6 months but show large variability, however (range 40000–90000 cells, Table
The number of DCX-positive cells in the periglomerular layer of the OB decreased highly significantly with age (
Our data show that the number of neurons in the triangular region of the pigeon hippocampus increased gradually in the young birds but peaked remarkably in the oldest pigeons. Interestingly, the number of DCX-positive neurons decreased only by a factor 2 over the lifespan. Conversely, the number of mature neurons in the periglomerular layer of the OB decreased moderately over the lifespan of the pigeons, while the number of DCX-positive neurons fell strongly after 6 months, and persisted at low yet variable levels through the rest of the life of the investigated pigeons.
Assuming that DCX is a reasonable proxy for postnatal neurogenesis in birds (Balthazart et al.,
A functional interpretation of the findings remains speculative yet, because in such a retrospective study, it is impossible to disentangle the various components of lifetime experience including navigation, and the effects of selection effects on the brains of the birds as compared to those not returning to the loft.
The age-dependent increase of mature hippocampal neurons in the pigeon brain fits explanations associating the volume or cell number of the hippocampus with superior navigational capacities or even expanded spatial memory of migratory birds (Healy et al.,
A positive effect of hippocampal neurogenesis on navigational capacities appears to be indicated by more recent reports about differences in hippocampal neurogenesis in non-migrating and migrating species, usually reporting higher rates in migrants (Barkan et al.,
Given the important role of the olfactory system in pigeon navigation, the observed decrease of mature periglomerular olfactory neurons was somewhat unexpected. In view of the heterogenic sample size, larger samples are probably needed to verify the findings, but an increase such as in the hippocampal formation seems unlikely. Likewise, the rapid decline of olfactory neurogenesis in young birds is most certainly not a chance event. Moreover, this decrease fits the navigational ontogeny of pigeons. It appears that high levels of OB neurogenesis reflect a sensitive period of navigational imprinting to the home loft coordinates (Gagliardo et al.,
The lifetime course of adult neurogenesis in two structures of the pigeon brain follows a general rule observed in mammals: juvenile peaking followed by a logarithmic decline when entering adult age, and subsequent persistence at lower levels that appear, however, much higher than in rodents.
Adult neurogenesis increases permanently the number of mature neurons in the pigeon HP tr, possibly reflecting accumulated experience necessary for reproduction, survival and navigation.
The decay of adult neurogenesis in the OB may correspond to the end of a sensitive period for imprinting positional information of the home loft.
VM initiated and carried out the study, H-PL managed logistics and support. Data analysis was done by SK. The manuscript was written by H-PL with contributions from VM and SK.
The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.
This study has been supported by Swiss National Science Foundation (31-122589) and the Swiss-South African Joint Research Project SSAJRP-09. We thank Prof. David Wolfer for helpful comments and providing infrastructure and resources for stereological data analysis, Anna Gagliardo for providing aged pigeons and for useful comments, Rosemarie Lang and Inger Drescher for help with histology, and Irina Lipp for taking care of the pigeon loft used to raise the younger pigeons.