Adult Human Hippocampus: No New Neurons in Sight

Adult Human Hippocampus: No New Neurons in Sight. Jon I Arellano, Brian Harding, Jean-Leon Thomas. Cerebral Cortex, bhy106, https://doi.org/10.1093/cercor/bhy106

Abstract: In this issue of Cerebral Cortex, Cipriani et al. are following up on the recent report of Sorrels et al. to add novel immunohistological observations indicating that, unlike rodents, adult and aging humans do not acquire new neurons in the hippocampus. The common finding emerging from these 2 different, but almost simultaneous studies is highly significant because the dentate gyrus of the hippocampus was, until recently, considered as the only structure in the human brain that may continue neurogenesis throughout the full life span.

Keywords: adult neurogenesis, dentate gyrus, human hippocampus

---
During the lifetime of most vertebrate animals, there is continuous neuronal addition and/or turnover, but this seemingly useful capacity decreases drastically during evolution (e.g., Jacobson 1970). The classical neuroanatomists generally believed that, after the developmental period which ends after puberty or sometimes during adolescence, the human neuronal assembly becomes stabilized (e.g., Ramon y Cajal 1913-1914). However, a paper published 2 decades ago in Nature Medicine (Eriksson et al. 1998) reported the detection in brain neural cells of deoxybromouridine (BrdU) initially administered to cancer patients for diagnostic purposes. This finding convinced a great number of scientists and lay people that the hippocampus was not different in humans than in other mammalian species, as it seemed to also generate new neurons during the entire life span. This possibility has been considered by many as a promise for endogenous cell replacement therapies for aging and neurological diseases as well as for CNS injury repair.

Although some studies have suggested caution in the interpretation of BrdU, showing a toxic effect and its incorporation into non-dividing cells damaged by drugs or exposed to hypoxia/ ischemia (e.g., Kuan et al. 2004; Breunig et al. 2007; Spector and Johanson 2007; Duque and Rakic 2015), the report by Eriksson et al. sparkled the field, and was followed by a number of studies that tried to ratify those results using immunohistochemical methods to identify markers of neurogenesis in postmortem human tissue. Those markers were aimed to progenitors (GFAP, Nestin, vimentin, Sox2), proliferating cells (Ki67, MCM2, PCNA), and immature neurons (DCX, PSANCAM, Tuj1). However, those studies have produced heterogeneous, inconclusive, sometimes contradictory results. One important obstacle is the ability to obtain well-preserved human tissues with a short postmortem delay, that may allow to obtain clear, reliable immunostaining.  Another caveat is that many of those reports studied only one marker of neurogenesis, producing inconclusive results. For example, using only Ki67 or MCM2 to identify proliferating cells without further characterization, is not a reliable method to assess neurogenesis, as the labeled cells might be producing oligodendrocytes or microglial cells (Reif et al. 2006; Knoth et al.  2010). The use of PCNA has added a lot of confusion to the field, as it is an inconsistent and unreliable marker of proliferation (Reif et al. 2006; Sanai et al. 2007). Also, morphological analysis of the cells labeled is a must, as for example, the markers of progenitors are shared with reactive astrocytes. Also DCX and PSANCAM expression has been reported in small cells with scant cytoplasm (Knoth et al. 2010; Jin et al. 2004), a morphology that is not expected in immature, migratory neurons.

Spalding et al. (2013) used an alternative technique to assess cell renewal, the neuronal content of C14 in the hippocampus. The increased levels of C14 in hippocampal neurons were interpreted as the consequence of a high and sustained level of neurogenesis in the dentate gyrus along life, in spite of the difficulty to reconcile this data with other studies in human (Knoth et al. 2010) and rodents (Ben Abdallah et al. 2010) showing that hippocampal neurogenesis has an early exponential decline before reaching low, stable adult levels.
At the time of writing this commentary, a new report has been published supporting the model that neurogenesis in the human hippocampus persists throughout adulthood (Boldrini et al. 2018). The authors of this study have, however, based their conclusion on disputable interpretations of immunelabeled cell types: some of the DCX- and PSA-NCAM-positive entities shown belong to the category of small and rounded cells described before, far from the typical elongated morphology of newly generated neurons. Also, cells identified as neuronal progenitors are Nestin- and GFAP-positive cells that do not have the characteristic polarized, radial-like morphology of progenitors and rather look very much like astrocytes.

It is clear that, irrespective of the caveats, there is widespread enthusiasm by the prospect of adult neurogenesis in humans and its therapeutic possibilities. As far as we know, there has been only one report published in 2016 (Dennis et al.  2016) reporting that hippocampal neurogenesis in adult humans is negligible. The authors found only an insignificant number of proliferating progenitors that corresponded to microglial cells and scarce DCX-expressing cells in the adult human hippocampus.

It is therefore quite a coincidence that, almost simultaneously, 2 independent papers coming from different parts of the world have used a similar approach and methodology leading to converging results and the following similar conclusions: hippocampal neurogenesis in humans decays exponentially during childhood and is absent or negligible in the adult. Those 2 papers are Sorrells et al. (2018) from the lab of Alvarez-Buylla in USA published in March in Nature, and the study by Cipriani and coworkers from the Adle-Biassette’s lab in France published in this issue of Cerebral Cortex (2018; 27: 000–000).

Cipriani et al. used a large battery of antibodies to identify progenitors, cell proliferation and differentiating neurons as well as their glial and vascular environment in the human hippocampus from early gestation to aging adults. As expected, they found abundant proliferating progenitors and newly generated neurons in the hippocampus during gestation, but they also observed a sharp decline of all the neurogenic markers after birth. It is worth to note that the analysis of hippocampal tissues at gestational and perinatal stages clearly assessed the presence of numerous proliferating progenitors and immature neurons. But those numbers decreased rapidly in early infancy, and by the age of 7, the authors detected only a few progenitors without significant proliferation and no DCX+ and Tuj1+ colabeled cells. In adults, a single cell co-expressing Nestin and Ki67 was found out of 19 samples, and only a few cells expressing DCX displaying a non-neuronal morphology (small nucleus, scant cytoplasm) were detected.

Interestingly, Cipriani et al. show almost identical results to those of Sorrells, Paredes and coworkers. Both 2 papers provide a high quality analysis of developmental and adult neurogenesis in the human hippocampus, while Sorrells et al. performed an exhaustive analysis, including transcriptome data that is, without a doubt, the most comprehensive to date. They combined the use of well-preserved human postmortem material and surgically resected hippocampi to assess for possible postmortem effects, and of an extensive battery of antibodies completed by electron microscopy analyses. As a key condition to the reliability of their study, they followed stringent criteria to identify differentiating neurons, based on the cellular morphology and the colocalization of DCX and PSA-NCAM. According to both Cipriani et al. and Sorrells, Paredes et al., DCX+ differentiating neurons, were absent from adult hippocampus samples. Both found however, small DCX+ cells with scant cytoplasm in adult samples, that were found by Sorrells, Paredes et al. to express oligodendroglial and microglial markers, expanding previous data of DCX expression in glial cells (Verwer et al. 2007; Zhang et al. 2014).  Additionally, they analyzed the formation of the subgranular zone (SGZ) in humans, but they concluded that actually there is no SGZ compartment in the human dentate gyrus comparable to the SGZ of rodents. The SGZ is an important player, as it is the specific niche where postnatal progenitors coalesce and proliferate in all other mammals exhibiting adult neurogenesis. Thus, the lack of SGZ may explain the lack of adult neurogenesis in the human dentate gyrus.

As pointed out by Cipriani and coworkers in this issue, the reasoning behind their study is to shed light on the neurogenic potential of the human hippocampus as “little information about human adult neurogenesis and neural stem/progenitor cells exists to justify the investment of resources in developing new treatments in humans, and most of the available evidence is inconclusive or contradictory.” Definitely, their study together with Sorrells’ and Dennis’ contributes to bring solid and consistent arguments to inform the field about a real possibility: the human species is once again different, and no significant neurogenesis occurs in the adult human hippocampus. This finding is in tune with the lack of subventricular neurogenesis and migration of new neurons to the olfactory bulb in the adult human, which has already been consistently reported (Sanai et al. 2011; Wang et al. 2011; Bergmann et al. 2012).

Finally, the absence of significant neurogenesis in normal adult humans poses a logical question: why would the human brain loose what seems to be a useful ability: to add, renew and regenerate neurons? As well formulated in the last sentence of the abstract of Sorrells’ paper: “The early decline in hippocampal neurogenesis raises questions about how the function of the dentate gyrus differs between humans and other species in which adult hippocampal neurogenesis is preserved.” However, as pointed out by Rakic (1985), there may be an advantage in keeping your old neurons without adding new ones, when the aim is to acquire and preserve complex knowledge during many decades of life. Stability in the neuronal population over a lifetime may not be a bad thing after all.