Transient oxytocin signaling primes the development and function of excitatory hippocampal neurons
“Transient oxytocin signaling primes the development and function of excitatory hippocampal neurons” by Ripamonti et al. presents a very interesting picture of the effects of oxytocin or oxytocin receptor (Oxtr) knockout on hippocampal neurons. The morphological and electrophysiological changes are intriguing, especially with regard to potential translation to human disorders. Although we find the results a bit counterintuitive in light of our work showing that acute application of oxytocin results in increases in synaptic responses in CA2 and CA3 neurons, similar to what we see with activation of vasopressin 1b receptor in CA2 neurons (1), we trust that the results presented accurately reflect the findings of the experiments described, and that the different preparations (cultured neurons vs. acutely prepared hippocampal slices, developmental stages, and drug application times, etc.) could easily account for the differences in observations.
We are primarily concerned, however, about the apparent inaccuracy concerning the identity of the neurons that express the Oxtr in the hippocampus, and importantly, those neurons likely to have had the Oxtr knocked out. Looking at Fig. 4 in the Ripamonti manuscript, we find it quite clear that the “Rostral” and “Medial levels show Venus expression within the CA2 and immediately adjacent CA3 areas, as well as within the dentate gyrus. This distribution is especially evident in the “Medial” row of Fig. 4a. We also see in the Figure supplement 1 no evidence for their reporter expression in the CA1 or in the proximal CA3, with the exception of a single Venus-labeled neuron in the CA1 field. Regarding the “Rostral” row in Fig. 4a, we would like to point out that the pyramidal cell layer in rostral hippocampal sections cut in the coronal plane, is in fact, made of mostly CA2 neurons, with CA1 neurons not evident in these sections. Most mouse and rat brain atlases make this clear, and we show 2 levels below. Incidentally, another study made the same error, attributing the Oxtr staining in rostral CA2 to area CA1 (2). The ‘Caudal’ sections shown in Fig. 4a are difficult to interpret without context, but are not inconsistent with expression primarily in CA2 and CA3. This Venus reporter expression reported in Ripamonti et al. is entirely consistent with the literature and with our in situ hybridization histochemistry in our figure (Oxtr expression in red in panels A and C). As with the Oxtr-Venus reporter expression in Fig. 4, cells in our figure are prominently labeled in area CA2, the immediately adjacent CA3, and in interneurons in the DG hilus, but not in CA1. For confirmation of the CA2 distribution of Oxtr in the “Rostral” level, our Panel B shows two CA2 markers, Amigo 2 (green) and Avpr1b (red), at a comparable level.
__________________
https://cdn.elifesciences.org/annotations-media/3266999698-001-a54e36f79286ce6461664e2991f4417ab064a3cea1944b3c7d3e62d381205104.jpg
Figure. In situ hybridization histochemistry was performed on sections from adult male C57BL/6j using the ViewRNA kit (Life Technologies) according to previously described methods(3). The levels in panels A and C correspond to Ripamonti et al.’s figure 4A “Rostral” and “Medial” levels, respectively, showing the distributions of Oxtr transcripts. The level in Panel B is similar to the “Rostral” level (and our Panel A level) but the section was probed for Avpr1b (red) and Amigo2 (green) transcripts, both markers of CA2. CA2 is indicated between the arrows.
__________________
We would also like to address the issue of what the authors’ cited literature actually reports. Both the cited Hammock and Levitt (4) and our paper (5) show oxytocin binding more heavily in area CA2, with much weaker labeling in the CA1. This binding, of course, does not reveal the location of the cell bodies in which the Oxtr is made or rule out very low levels of the receptor. One of our more recent papers, though, shows that Oxtr expression in mice is limited to CA2 and adjacent CA3 pyramidal cells (ref. 6, suppl. fig. 4a). Mitre et al.’s study (7) is used to support the assertion that all pyramidal cells in the hippocampus express Oxtr. However, a closer examination reveals much greater immunostaining in CA2 and adjacent CA3, with lesser staining in CA1, especially in females. A parsimonious explanation is that both the weaker labeling in the receptor binding as well as immunostaining in CA1 represent axonal processes with presynaptic Oxtr. Either way, though, both the authors’ figure showing the distribution of Venus expression and our in situ hybridization results suggest that Oxtr expression in CA1 neurons is quite low at best.
The bigger problem, then, is with the conclusions of the article. First, because the percentages of the hippocampal neurons in culture expressing the Oxtr is unknown, and indeed likely to be very low in many neurons, it is difficult to know if the effects of the oxytocin were in fact acting directly on the neurons studied, or whether it was acting on a few neurons that then had widespread effects beyond the micro-islands (although we would not venture an explanation for how this would occur). Thus the ‘CA1’ neurons analyzed morphologically were either a) not expressing Oxt receptors, b) expressing them only early in development at the time of exposure, or c) were CA2/CA3 neurons. We recognize that oxytocin receptor levels are not always straightforward to assess, but other CA subfield markers, or Venus expression where appropriate, would suffice to at least identify the neuronal types likely to express the receptors. Indeed, the effect may be even bigger if CA1 neurons were compared separately in the culture study. Alternatively, if the authors are proposing that the effects were due to developmental changes in Oxtr expression, they should have shown that changes supporting this idea actually occur in their culture preparation (perhaps by Western blot or qPCR). Second, although the authors attempted to address this issue by over-expressing Oxtr in striatal neurons, the findings may, or may not be relevant to the hippocampal neurons. Finally, we noticed that in neurons from the knockout mouse (data in Figure 1–figure supplement 3), the lack of effect of oxytocin seems to be due to some rather large differences between the untreated states of the knockouts and the control mice (presented in Figure 1 and Figure 2). Specifically, the untreated knockout neurons more closely resembled the oxytocin-treated controls in many of the measures. Thus, the cause of the effects of oxytocin application and oxytocin receptor knockout on CA1 neurons remains unknown, and several opportunities to clarify the issue of neuron type were missed, despite the care taken in this otherwise rigorous study.
The CA2 region has recently been highlighted for its distinct morphological, molecular and physiological characteristics, as well as for its role contributing to specific behavioral functions (8). Historically, CA2 neurons have been grouped with CA3 neurons, however, it is our belief that moving forward, authors should be careful to distinguish this small but interesting sub-region of the hippocampus.
(1) Pagani JH, Zhao M, Cui Z, Williams Avram SK, Caruana DA, Dudek SM, Young WS (2015): Role of the vasopressin 1b receptor in rodent aggressive behavior and synaptic plasticity in hippocampal area CA2. Mol Psychiat. 20: 490–499.
(2) Tomizawa K, Iga N, Lu Y-F, Moriwaki A, Matsushita M, Li S-T, Miyamoto O, Itano T
Matsui H (2003) Oxytocin improves long-lasting spatial memory during motherhood through MAP kinase cascade, Nat Neurosci. 6, 384 - 390
(3) Young WS, Song J, Mezey E (2016): Hybridization Histochemistry of Neural Transcripts. Curr Protoc Neurosci. 75: 1.3.1–1.3.27.
(4) Hammock EAD, Levitt P (2013): Oxytocin receptor ligand binding in embryonic tissue and postnatal brain development of the C57BL/6J mouse. Front Behav Neurosci. 7: 195.
(5) Pagani JH, Lee H-J, Young WS (2011): Postweaning, forebrain-specific perturbation of the oxytocin system impairs fear conditioning. Genes Brain Behav. 10: 710–719.
(6) Smith AS, Williams Avram SK, Cymerblit-Sabba A, Song J, Young WS (2016): Targeted activation of the hippocampal CA2 area strongly enhances social memory. Mol Psychiat. 21: 1137–1144.
(7) Mitre M, Marlin BJ, Schiavo JK, Morina E, Norden SE, Hackett TA, et al. (2016): A Distributed Network for Social Cognition Enriched for Oxytocin Receptors. J Neurosci. 36: 2517–2535.
(8) Dudek SM, Alexander GM, Farris S (2016) Rediscovering area CA2: unique properties and functions. Nat Rev Neurosci. 17: 89-102.
Authors:
W. Scott Young (1), Sarah K. Williams Avram (1), Shannon Faris (2), Michelle Stackmann (1), Rahul Chaturvedi (1), Serena M. Dudek (2)
(1) Section on Neural Gene Expression, National Institute of Mental Health, National Institutes of Health, Bethesda, MD 20892
(2) Neurobiology Laboratory, National Institute of Environmental Health Sciences, National Institutes of Health, Research Triangle Park, NC 27713
This research was supported by the Intramural Research Program of the National Institutes of Health, National Institute of Mental Health (ZIAMH002498) and National Institute of Environmental Health Sciences (Z01-ES100221).
