Hypoxic-ischaemic brain injury is the most common cause of disability in newborn babies. This happens when the blood supply to the brain is temporarily blocked during birth and cells do not receive the oxygen and nutrients they need to survive. Cooling the babies down after the hypoxic-ischemic attack (via a technique called hypothermic treatment) can to some extent reduce the damage caused by the injury. However, doctors still need new drugs that can protect the brain and improve its recovery after the injury has occurred.

Research in mice suggests that a chemical called lactate might help the brain to recover. Lactate is produced by muscles during hard exercise to provide energy to cells when oxygen levels are low. Recent studies have shown that it can also act as a signalling molecule that binds to a receptor called HCAR1 (short for hydroxycarboxylic acid receptor) on the surface of cells. However, it is poorly understood what role HCAR1 plays in the brain and whether it helps the brain recover from a hypoxic-ischaemic injury.

To investigate, Kennedy et al. compared newborn mice with and without the gene that codes for HCAR1 that had undergone a hypoxic-ischaemic brain injury. While HCAR1 did not protect the mice from the disease, it did help their brains to heal. Mice with the gene for HCAR1 partly recovered some of their damaged brain tissue six weeks after the injury. Their cells switched on thousands of genes involved in the immune system and cell cycle, resulting in new brain cells being formed that could repopulate the injured areas. In contrast, the brain tissue of mice lacking HCAR1 barely produced any new cells.

These findings suggest that HCAR1 may help with brain recovery after hypoxia-ischemia in newborn mice. This could lead to the development of drugs that might reduce or repair brain damage in newborn babies. However, further studies are needed to investigate whether HCAR1 has the same effect in humans.

Cerebral hypoxia-ischemia (HI) affects around 1.5 per 1000 live born births in the developed countries (Kurinczuk et al., 2010). It is characterised by an insufficient supply of blood and oxygen to the brain, leading to cell death and brain tissue damage. Hypothermia is the mainstay of today’s treatment (Shankaran et al., 2012), but a high percentage of survivors still experience long-term neurological effects, including cerebral palsy, epilepsy and cognitive disabilities (Douglas-Escobar and Weiss, 2015). Following a hypoxic-ischemic event, the neonatal brain has the ability to partly regenerate (Donega et al., 2013). This process of brain tissue regeneration requires a coordinated increase in microglia-induced inflammation, cell proliferation and angiogenesis. After the acute phase of cerebral HI, debris from dead cells activates microglia, the resident immune cells of the brain. These cells have the ability to remove cell debris by phagocytosis and release factors that stimulate tissue repair (Jin and Yamashita, 2016). At the same time, the hypoxic-ischemic event leads to the release of various growth factors, which stimulate proliferation of neural progenitor cells (Donega et al., 2013) as well as angiogenesis (Donega et al., 2013; Shweiki et al., 1992). In mammalian postnatal brain, the majority of progenitor cells are located in proliferating areas of the brain, of which the most well described are the subventricular zone, located adjacent to the lateral ventricles, and the dentate gyrus of the hippocampus. Proliferating cells migrate from these areas to repopulate the damaged tissue. Targeting these areas is therefore a potential strategy to stimulate repair processes after an ischemic insult.

Recent studies in mice have shown improved recovery after neonatal HI by administration of lactate (Roumes et al., 2021; Tassinari et al., 2020), but the underlying mechanisms for this beneficial effect are unclear. It is unknown whether lactate improves recovery by giving metabolic support to the cells or by signalling via the Hydroxycarboxylic acid receptor 1 (HCAR1), or both.

HCAR1 is a G_i-protein-coupled receptor. It was first described in adipose tissue where it inhibits lipolysis through lowering of cyclic adenine monophosphate (cAMP; Ahmed et al., 2010). In the brain, HCAR1 activation can modulate neuronal firing rates in vitro (de Castro Abrantes et al., 2019) and stimulate brain angiogenesis in vivo (Morland et al., 2017). Until now, a role of HCAR1 in tissue protection or repair after ischemia has not been demonstrated, although studies from different cell lines suggest that it may regulate cell proliferation and differentiation (Stäubert et al., 2015; Wu et al., 2018b).

Here, we investigate the role of HCAR1 in neonatal HI in mice. We show that HCAR1 knockout (KO) mice have a reduced ability to regenerate brain tissue after HI. By examination of neurosphere cultures in vitro and immunohistochemical staining of brain tissue after HI, we find that HCAR1 KO mice display impaired proliferation of neural progenitor cells. Furthermore, HI-reactive proliferation of microglia, astrocytes and oligodendrocyte progenitors is perturbed in the HCAR1 KO. In addition, we find that the microglia are less activated post HI. Transcriptome analysis revealed that subventricular zones from HCAR1 KO mice have an almost complete lack of transcriptional response to HI. This was specific to the subventricular zone as hippocampal samples from HCAR1 KO mice responded similar to that of wildtype (WT) mice. Thus, HCAR1 is a crucial transcriptional regulator of tissue response to ischemia in the subventricular zone. HCAR1 could therefore be targeted to promote tissue repair after HI.

We report that HCAR1 KO mice have a substantial deficit in the restoration of brain tissue after neonatal HI, indicating that lactate receptor HCAR1 plays a crucial role in the processes that lead to tissue repair. Since no exogenous lactate was administered in our experiments, the observed effect must be due to endogenous lactate or possible baseline receptor activity. Indeed, a similar HI mouse model showed a lactate rise in the ipsilateral hemisphere (Mikrogeorgiou et al., 2020). Moreover, the lactate level rises in neonatal humans and piglets after a hypoxic-ischemic episode (Roelants-Van Rijn et al., 2001; Wu et al., 2018a; Zheng and Wang, 2017). It is likely that administration of lactate could further leverage the effect of HCAR1: two recent studies showed that mouse pups injected with lactate before, or in the hours or days following HI had improved recovery (Roumes et al., 2021; Tassinari et al., 2020). The authors suggested that the protective effect of lactate was mainly due to lactate being used as a metabolite to make ATP (Roumes et al., 2021; Tassinari et al., 2020). However, our data suggest that recovery after lactate injection is partly mediated by HCAR1, which promotes a stronger transcriptional response to HI and thereby facilitates neurogenesis and tissue regeneration after injury. On the other hand, these studies also showed an effect of lactate injections on acute infarct volume. Therefore, a putative explanation is that lactate injected before or immediately after HI reduces lesion size by mainly working as a metabolite, whereas lactate injected at a later time point in large works via HCAR1. Lactate injections can also be protective after ischemic stroke in adult mice (Berthet et al., 2009; Berthet et al., 2012; Buscemi et al., 2020; Castillo et al., 2015). Here, it also seems to be a combination of metabolic and HCAR1-dependent effects since replacing lactate with either the HCAR1 receptor agonist 3,5-dihydroxybenzoic acid (3, 5 DHBA) or the metabolic substrate pyruvate offered partial protection (Castillo et al., 2015). In the current study, we induced a permanent occlusion of the left carotid artery. In human HI, reperfusion often occurs, which can lead to reperfusion injury. Although lactate administrations has been shown to reduce reperfusion injury in adult mice and humans (Annoni et al., 2021), a recent study from adult rats suggested that the effect was HCAR1 independent (Buscemi et al., 2021). We are not aware of any studies on the effect of lactate and HCAR1 on reperfusion injury in neonatal HI.

An ischemic event will induce a significant transcriptional response in the neonatal as well as the adult brain (Androvic et al., 2020; Hedtjärn et al., 2004). By RNA sequencing of tissue samples from the subventricular zone, we found that HCAR1 KO mice displayed a weak transcriptional response to HI, with a 90% reduction in DEGs compared with WT mice. Hence, HCAR1 appears to be important for the induction of the transcriptional response to HI. This HCAR1 dependence is specific for the subventricular zone, as hippocampal samples showed little difference between WT and HCAR1 KO mice in the transcriptional profile after HI. In line with our findings, a recent study shows an effect of HCAR1 activation on neurogenesis in the subventricular zone and not in the hippocampal subgranular zone (Lambertus et al., 2020). Furthermore, the structure and cellular composition of the two neurogenic niches are in themselves unique and react differently to HI in the neonatal and adult brain (Semple et al., 2013). A possible role of HCAR1 in other, more recently discovered neurogenic areas (Li et al., 2009) remains to be investigated.

An essential part of the repair process after a neonatal brain injury is the generation of new cells. This occurs by increased proliferation and differentiation of progenitor cells and involves upregulation of genes involved in the cell cycle pathway (Prasad et al., 2012; Wen et al., 2005). By use of neurosphere assays, we showed that neural progenitor cells from HCAR1 KO mice have reduced proliferation ability. Moreover, while WT mice doubled the ratio of proliferating cells after HI in the striatal SVZ-IZ, HCAR1 KO mice were unable to increase cell proliferation (Figure 2P). In the V-SVZ, there was no difference in cell proliferation post ischemia in the WT or HCAR1 KO mice. This lacking proliferative response at the basal part of the SVZ has been reported earlier in a study using the same HI-protocol as we present here (Sejersted et al., 2011). The same study showed similar reactive neurogenesis in the deeper part of the SVZ and IZ in WT mice. Importantly, the SVZ is in a transitional phase at the time of our analyses: during the first weeks of development, the SVZ shrinks from being up to 300 μm thick, to under 100 μm, and shifts closer to the VZ (Fiorelli et al., 2015). In the same period, there is a gradual decrease in the embryonic radial neuronal migration and a shift to the adult rostral neuronal migration towards the olfactory bulb (Inta et al., 2008). Upon a hypoxic ischemic insult, the radial pathway reopens and provides striatum and cortex with cells from the SVZ in adult (Goings et al., 2004) and postnatal/young mice (Covey et al., 2010). Therefore, the increased number of progenitors we observed in the SVZ-IZ, presumably represent either radial embryonic or reactive neuronal migration/neurogenesis. Given that we did not see any differences between the genotypes in the number of progenitors in the contralateral SVZ-IZ (Figure 2), we conclude that the increased progenitors at the ipsilateral hemisphere represent HI-reactive neurogenesis.

Despite the lack of differences in cell proliferation between the contralateral WT and HCAR1 KO in vivo, the in vitro cultured neurospheres showed reduced proliferation in HCAR1 KO compared with WT. This discrepancy could be due to compensatory mechanisms in vivo: as neurospheres lack vasculature and the extracellular milieu that is present in vivo, they may not be able to compensate a lack of HCAR1-driven proliferation. It is also possible that the experimental steps needed to produce neurospheres cause a stress response in the cells that does not occur in the contralateral hemispheres in vivo.

In addition to the effect of HCAR1 on neurogenesis, our analyses of astrocytes, microglia and oligodendrocyte suggest that HCAR1 can affect proliferation post HI in glial cells. Our analysis was done as density of cells per area (in Supplements to Figures 2—4) and as ratio per total number of cells (in Figures 2—4). While the cell densities per area give information about the overall change in each cell population, the cell ratios say how much a cell population has changed in comparison to the total density of cells. A significant increase in cell ratio for a cell type would require that this cell type increased more than the total cell increase. In WT mice, this was the case for proliferating microglia and astrocytes, but not for oligodendrocytes, indicating that the density of proliferating microglia, astrocytes and neuronal progenitors had increased more than that of oligodendrocytes. Still, the density of proliferating oligodendrocytes per se was increased after HI in WT mice, but not in KO mice. Therefore, we argue that HCAR1 KO affects proliferation after HI in all the analysed cell lineages.

Transcriptome analysis of the subventricular niche revealed that the cell cycle genes were strongly upregulated after HI in WT, but not in HCAR1 KO mice. Hence, it seems that HCAR1 can act as a transcriptional regulator of cell cycle genes, thereby controlling cell proliferation. It is important to note that transcriptome data do not always correspond with proteome data, and therefore a proteome analysis would be useful to confirm the transcriptome results in this study. Nevertheless, the combination of transcriptome data with western blots partly confirming down-regulation of some cell cycle genes, and the immunostainings showing reduced cell proliferation after HI, support our hypothesis. A role of HCAR1 in cell proliferation was previously shown in cancer and osteoblast cell lines (Stäubert et al., 2015; Wu et al., 2018a). Altogether, these findings may provide new insights of importance for treatments of hypoxic ischemic brain injury in the perinatal period and perhaps in the adult brain. However, induced reactive proliferation is increased in the postnatal mice (P9) compared with adult mice, and the time window of this robust induced neurogenesis decreases substantially within the next two weeks of the rodents life (Covey et al., 2010).

Microglia carry out several vital functions in response to brain injury. These include clearance of damaged tissue, resistance to infections and restoration of tissue homeostasis (Jin and Yamashita, 2016). We detected an increase in the proliferation and activation of microglia in the peri-infarct zone after HI in WT, but not in HCAR1 KO mice. The lack of microglia proliferation could be due to the reduced induction of cell cycle response, as discussed above. The absence of microglia activation was confirmed on a transcriptional level, as genes associated with activated microglia were upregulated in WT but not in KO. This effect may be explained by the reduced complement system response in HCAR1 KO. Altogether, the differences in microglial data suggest that the damaging effects of HCAR1 KO in HI are due to interplay of both the immune response and proliferation and regeneration of the damaged brain cells.

In addition to the effects on cell cycle and microglia activation discussed above, the transcriptome analysis revealed a large number of differentially expressed genes and pathways, including genes involved in DNA repair and glutamate signalling. These processes will likely also influence the ability of the brain tissue to repair after injury.

Apoptotic pathways are upregulated after HI and are a major contributor to cell loss. But our transcriptome data did not reveal any differences between the genotypes in expression of apoptotic genes. There was also no significant upregulation of apoptotic pathways after HI within each genotype. Our RNA seq was performed on SVZ tissue, which lies a distance away from the infarcted area and may therefore experience less HI-induced apoptosis. However, apoptosis was also not upregulated in the hippocampus after HI (data not shown). Another possible explanation for the lack of apoptosis in our RNA seq data could be the timing: apoptotic activity has been shown to peak around 24 hr after HI, at least in rats (Wang et al., 2001). Still, we cannot exclude the possibility that a transcriptome analysis from an area closer to the infarct core (e.g. the striatal peri-infarct zone), and potentially taken at a different time point after HI, would have revealed differences in apoptotic pathways between HCAR1 KO and WT. Thus, a potential difference in the number of dead cells could also contribute to the different outcomes in the two genotypes.

Overall, our data show that HCAR1 is a key transcriptional regulator of brain tissue response to an ischemic insult. We therefore propose a model in which activation of HCAR1 by elevated lactate during and after HI stimulates a transcriptional response involving pathways responsible for tissue repair (Figure 6). HCAR1 could be a target of future treatment for neonatal HI and possibly other forms of brain injury.

Figure 6

Download asset Open asset

The RNA sequence data are available at Dryad, https://doi.org/10.5061/dryad.8w9ghx3kw. Script for analysis of immunostaining and algorithms used for WEKA segmentation for Figures 3 and 4 are provided in Figure 3—source code 1 - BrdU WEKA, Figure 3—source code 2 - DAPI WEKA, Figure 3—source code 3 - IBA1 WEKA, Figure 3—source code 4 - Script Microglia analysis, Figure 4—source code 1 - Script GFAP analysis. Raw images of western blots in Figure 5 are provided in Figure 5—source data 1.

1. 1. Rinholm JE
   2. Kennedy L
   3. Glesaaen E
   4. Palibrk V
   5. Pannone M
   6. Wang W
   7. Al-Jabri A
   8. Suganthan R
   9. Meyer N
   10. Lin X
   11. Bergersen L
   12. Bjørås M

   (2021) Dryad Digital Repository

   HCAR1 KO and WT hypoxia-ischemia RNA sequencing data.

   https://doi.org/10.5061/dryad.8w9ghx3kw

1. 1. Ahmed K
   2. Tunaru S
   3. Tang C
   4. Müller M
   5. Gille A
   6. Sassmann A
   7. Hanson J
   8. Offermanns S

   (2010) An autocrine lactate loop mediates insulin-dependent inhibition of lipolysis through GPR81

   Cell Metabolism 11:311–319.

   https://doi.org/10.1016/j.cmet.2010.02.012

   • PubMed
   • Google Scholar
2. 1. Amat JA
   2. Ishiguro H
   3. Nakamura K
   4. Norton WT

   (1996) Phenotypic diversity and kinetics of proliferating microglia and astrocytes following cortical stab wounds

   GLIA 16:368–382.

   https://doi.org/10.1002/(SICI)1098-1136(199604)16:4<368::AID-GLIA9>3.0.CO;2-W

   • PubMed
   • Google Scholar
3. Software

   1. Andrews S

   (2022) FastQC: a quality control tool for high throughput sequence data

   FastQC.

   https://www.bioinformatics.babraham.ac.uk/projects/fastqc/
4. 1. Androvic P
   2. Kirdajova D
   3. Tureckova J
   4. Zucha D
   5. Rohlova E
   6. Abaffy P
   7. Kriska J
   8. Valny M
   9. Anderova M
   10. Kubista M
   11. Valihrach L

   (2020) Decoding the transcriptional response to ischemic stroke in young and aged mouse brain

   Cell Reports 31:107777.

   https://doi.org/10.1016/j.celrep.2020.107777

   • PubMed
   • Google Scholar
5. 1. Annoni F
   2. Peluso L
   3. Gouvêa Bogossian E
   4. Creteur J
   5. Zanier ER
   6. Taccone FS

   (2021) Brain protection after anoxic brain injury: Is lactate supplementation helpful?

   Cells 10:1714.

   https://doi.org/10.3390/cells10071714

   • PubMed
   • Google Scholar
6. 1. Arganda-Carreras I
   2. Kaynig V
   3. Rueden C
   4. Eliceiri KW
   5. Schindelin J
   6. Cardona A
   7. Sebastian Seung H

   (2017) Trainable weka segmentation: A machine learning tool for microscopy pixel classification

   Bioinformatics 33:2424–2426.

   https://doi.org/10.1093/bioinformatics/btx180

   • PubMed
   • Google Scholar
7. 1. Berthet C
   2. Lei H
   3. Thevenet J
   4. Gruetter R
   5. Magistretti PJ
   6. Hirt L

   (2009) Neuroprotective role of lactate after cerebral ischemia

   Journal of Cerebral Blood Flow and Metabolism 29:1780–1789.

   https://doi.org/10.1038/jcbfm.2009.97

   • PubMed
   • Google Scholar
8. 1. Berthet C
   2. Castillo X
   3. Magistretti PJ
   4. Hirt L

   (2012) New evidence of neuroprotection by lactate after transient focal cerebral ischaemia: extended benefit after intracerebroventricular injection and efficacy of intravenous administration

   Cerebrovascular Diseases 34:329–335.

   https://doi.org/10.1159/000343657

   • PubMed
   • Google Scholar
9. 1. Buscemi L
   2. Blochet C
   3. Price M
   4. Magistretti PJ
   5. Lei H
   6. Hirt L

   (2020) Extended preclinical investigation of lactate for neuroprotection after ischemic stroke

   Clinical and Translational Neuroscience 4:2514183X2090457.

   https://doi.org/10.1177/2514183X20904571

   • Google Scholar
10. 1. Buscemi L
    2. Blochet C
    3. Magistretti PJ
    4. Hirt L

    (2021) Hydroxycarboxylic acid receptor 1 and neuroprotection in a mouse model of cerebral ischemia-reperfusion

    Frontiers in Physiology 12:689239.

    https://doi.org/10.3389/fphys.2021.689239

    • PubMed
    • Google Scholar
11. 1. Castillo X
    2. Rosafio K
    3. Wyss MT
    4. Drandarov K
    5. Buck A
    6. Pellerin L
    7. Weber B
    8. Hirt L

    (2015) A probable dual mode of action for both L- and D-lactate neuroprotection in cerebral ischemia

    Journal of Cerebral Blood Flow and Metabolism 35:1561–1569.

    https://doi.org/10.1038/jcbfm.2015.115

    • PubMed
    • Google Scholar
12. 1. Covey MV
    2. Jiang Y
    3. Alli VV
    4. Yang Z
    5. Levison SW

    (2010) Defining the critical period for neocortical neurogenesis after pediatric brain injury

    Developmental Neuroscience 32:488–498.

    https://doi.org/10.1159/000321607

    • PubMed
    • Google Scholar
13. 1. de Castro Abrantes H
    2. Briquet M
    3. Schmuziger C
    4. Restivo L
    5. Puyal J
    6. Rosenberg N
    7. Rocher AB
    8. Offermanns S
    9. Chatton JY

    (2019) The lactate receptor HCAR1 modulates neuronal network activity through the activation of Gα and Gβγ subunits

    The Journal of Neuroscience 39:4422–4433.

    https://doi.org/10.1523/JNEUROSCI.2092-18.2019

    • Google Scholar
14. 1. Deng Y
    2. Xie D
    3. Fang M
    4. Zhu G
    5. Chen C
    6. Zeng H
    7. Lu J
    8. Charanjit K

    (2014) Astrocyte-derived proinflammatory cytokines induce hypomyelination in the periventricular white matter in the hypoxic neonatal brain

    PLOS ONE 9:e87420.

    https://doi.org/10.1371/journal.pone.0087420

    • PubMed
    • Google Scholar
15. 1. DePaula-Silva AB
    2. Gorbea C
    3. Doty DJ
    4. Libbey JE
    5. Sanchez JMS
    6. Hanak TJ
    7. Cazalla D
    8. Fujinami RS

    (2019) Differential transcriptional profiles identify microglial- and macrophage-specific gene markers expressed during virus-induced neuroinflammation

    Journal of Neuroinflammation 16:152.

    https://doi.org/10.1186/s12974-019-1545-x

    • Google Scholar
16. 1. Donega V
    2. Van Velthoven CTJ
    3. Nijboer CH
    4. Kavelaars A
    5. Heijnen CJ

    (2013) The endogenous regenerative capacity of the damaged newborn brain: Boosting neurogenesis with mesenchymal stem cell treatment

    Journal of Cerebral Blood Flow and Metabolism 33:625–634.

    https://doi.org/10.1038/jcbfm.2013.3

    • Google Scholar
17. 1. Douglas-Escobar M
    2. Weiss MD

    (2015) Hypoxic-ischemic encephalopathy: A review for the clinician

    JAMA Pediatrics 169:397–403.

    https://doi.org/10.1001/jamapediatrics.2014.3269

    • PubMed
    • Google Scholar
18. 1. Falahati S
    2. Breu M
    3. Waickman AT
    4. Phillips AW
    5. Arauz EJ
    6. Snyder S
    7. Porambo M
    8. Goeral K
    9. Comi AM
    10. Wilson MA
    11. Johnston MV
    12. Fatemi A

    (2013) Ischemia-induced neuroinflammation is associated with disrupted development of oligodendrocyte progenitors in a model of periventricular leukomalacia

    Developmental Neuroscience 35:182–196.

    https://doi.org/10.1159/000346682

    • PubMed
    • Google Scholar
19. 1. Fiorelli R
    2. Azim K
    3. Fischer B
    4. Raineteau O

    (2015) Adding a spatial dimension to postnatal ventricular-subventricular zone neurogenesis

    Development 142:2109–2120.

    https://doi.org/10.1242/DEV.119966

    • Google Scholar
20. 1. Fumagalli S
    2. Perego C
    3. Pischiutta F
    4. Zanier ER
    5. De Simoni MG

    (2015) The ischemic environment drives microglia and macrophage function

    Frontiers in Neurology 6:81.

    https://doi.org/10.3389/fneur.2015.00081

    • PubMed
    • Google Scholar
21. 1. Goings GE
    2. Sahni V
    3. Szele FG

    (2004) Migration patterns of subventricular zone cells in adult mice change after cerebral cortex injury

    Brain Research 996:213–226.

    https://doi.org/10.1016/j.brainres.2003.10.034

    • PubMed
    • Google Scholar
22. 1. Hagberg H
    2. Ichord R
    3. Palmer C
    4. Yager JY
    5. Vannucci SJ

    (2002) Animal models of developmental brain injury: relevance to human disease A summary of the panel discussion from the Third Hershey Conference on Developmental Cerebral Blood Flow and Metabolism

    Developmental Neuroscience 24:364–366.

    https://doi.org/10.1159/000069040

    • PubMed
    • Google Scholar
23. 1. Hammad A
    2. Westacott L
    3. Zaben M

    (2018) The role of the complement system in traumatic brain injury: A review

    Journal of Neuroinflammation 15:1–15.

    https://doi.org/10.1186/s12974-018-1066-z

    • PubMed
    • Google Scholar
24. 1. Hayashi T
    2. Iwai M
    3. Ikeda T
    4. Jin G
    5. Deguchi K
    6. Nagotani S
    7. Zhang H
    8. Sehara Y
    9. Nagano I
    10. Shoji M
    11. Ikenoue T
    12. Abe K

    (2005) Neural precursor cells division and migration in neonatal rat brain after ischemic/hypoxic injury

    Brain Research 1038:41–49.

    https://doi.org/10.1016/j.brainres.2004.12.048

    • PubMed
    • Google Scholar
25. 1. Hedtjärn M
    2. Mallard C
    3. Eklind S
    4. Gustafson-Brywe K
    5. Hagberg H

    (2004) Global gene expression in the immature brain after hypoxia-ischemia

    Journal of Cerebral Blood Flow and Metabolism 24:1317–1332.

    https://doi.org/10.1097/01.WCB.0000141558.40491.75

    • PubMed
    • Google Scholar
26. 1. Inta D
    2. Alfonso J
    3. Von Engelhardt J
    4. Kreuzberg MM
    5. Meyer AH
    6. Van Hooft JA
    7. Monyer H

    (2008) Neurogenesis and widespread forebrain migration of distinct GABAergic neurons from the postnatal subventricular zone

    PNAS 105:20994–20999.

    https://doi.org/10.1073/PNAS.0807059105

    • Google Scholar
27. 1. Jin X
    2. Yamashita T

    (2016) Microglia in central nervous system repair after injury

    Journal of Biochemistry 159:491–496.

    https://doi.org/10.1093/jb/mvw009

    • PubMed
    • Google Scholar
28. 1. Kim D
    2. Langmead B
    3. Salzberg SL

    (2015) HISAT: A fast spliced aligner with low memory requirements

    Nature Methods 12:357–360.

    https://doi.org/10.1038/nmeth.3317

    • PubMed
    • Google Scholar
29. 1. Kurinczuk JJ
    2. White-Koning M
    3. Badawi N

    (2010) Epidemiology of neonatal encephalopathy and hypoxic-ischaemic encephalopathy

    Early Human Development 86:329–338.

    https://doi.org/10.1016/j.earlhumdev.2010.05.010

    • PubMed
    • Google Scholar
30. 1. Lambertus M
    2. Thøring Øverberg L
    3. Andersson KA
    4. Hjelden MS
    5. Hadzic A
    6. Haugen ØP
    7. Storm‐Mathisen J
    8. Hildegard Bergersen L
    9. Geiseler S
    10. Morland C

    (2020) L‐lactate induces neurogenesis in the mouse ventricular‐subventricular zone via the lactate receptor HCA1

    Acta Physiologica 231:e13587.

    https://doi.org/10.1111/apha.13587

    • Google Scholar
31. 1. Li G
    2. Kataoka H
    3. Coughlin SR
    4. Pleasure SJ

    (2009) Identification of a transient subpial neurogenic zone in the developing dentate gyrus and its regulation by cxcl12 and reelin signaling

    Development 136:327–335.

    https://doi.org/10.1242/dev.025742

    • PubMed
    • Google Scholar
32. 1. Liao Y
    2. Smyth GK
    3. Shi W

    (2014) FeatureCounts: an efficient general purpose program for assigning sequence reads to genomic features

    Bioinformatics 30:923–930.

    https://doi.org/10.1093/bioinformatics/btt656

    • PubMed
    • Google Scholar
33. 1. Liao Y
    2. Wang J
    3. Jaehnig EJ
    4. Shi Z
    5. Zhang B

    (2019) WebGestalt 2019: gene set analysis toolkit with revamped UIs and APIs

    Nucleic Acids Research 47:W199–W205.

    https://doi.org/10.1093/nar/gkz401

    • PubMed
    • Google Scholar
34. 1. Lindvall O
    2. Kokaia Z

    (2015) Neurogenesis following stroke affecting the adult brain

    Cold Spring Harbor Perspectives in Biology 7:a019034.

    https://doi.org/10.1101/cshperspect.a019034

    • PubMed
    • Google Scholar
35. 1. Love MI
    2. Huber W
    3. Anders S

    (2014) Moderated estimation of fold change and dispersion for RNA-seq data with DESeq2

    Genome Biology 15:550.

    https://doi.org/10.1186/s13059-014-0550-8

    • PubMed
    • Google Scholar
36. 1. Mattiesen WRC
    2. Tauber SC
    3. Gerber J
    4. Bunkowski S
    5. Brück W
    6. Nau R

    (2009) Increased neurogenesis after hypoxic-ischemic encephalopathy in humans is age related

    Acta Neuropathologica 117:525–534.

    https://doi.org/10.1007/s00401-009-0509-0

    • PubMed
    • Google Scholar
37. 1. Mikrogeorgiou A
    2. Chen Y
    3. Lee BS
    4. Bok R
    5. Sheldon RA
    6. Barkovich AJ
    7. Xu D
    8. Ferriero DM

    (2020) A metabolomics study of hypoxia ischemia during mouse brain development using hyperpolarized 13C

    Developmental Neuroscience 42:49–58.

    https://doi.org/10.1159/000506982

    • PubMed
    • Google Scholar
38. 1. Morland C
    2. Andersson KA
    3. Haugen ØP
    4. Hadzic A
    5. Kleppa L
    6. Gille A
    7. Rinholm JE
    8. Palibrk V
    9. Diget EH
    10. Kennedy LH
    11. Stølen T
    12. Hennestad E
    13. Moldestad O
    14. Cai Y
    15. Puchades M
    16. Offermanns S
    17. Vervaeke K
    18. Bjørås M
    19. Wisløff U
    20. Storm-Mathisen J
    21. Bergersen LH

    (2017) Exercise induces cerebral VEGF and angiogenesis via the lactate receptor HCAR1

    Nature Communications 8:15557.

    https://doi.org/10.1038/ncomms15557

    • PubMed
    • Google Scholar
39. 1. Morrison HW
    2. Filosa JA

    (2013) A quantitative spatiotemporal analysis of microglia morphology during ischemic stroke and reperfusion

    Journal of Neuroinflammation 10:4.

    https://doi.org/10.1186/1742-2094-10-4

    • PubMed
    • Google Scholar
40. 1. Neumann H
    2. Kotter MR
    3. Franklin RJM

    (2009) Debris clearance by microglia: an essential link between degeneration and regeneration

    Brain 132:288–295.

    https://doi.org/10.1093/brain/awn109

    • PubMed
    • Google Scholar
41. 1. Palibrk V
    2. Suganthan R
    3. Scheffler K
    4. Wang W
    5. Bjørås M
    6. Bøe SO

    (2016) PML regulates neuroprotective innate immunity and neuroblast commitment in a hypoxic-ischemic encephalopathy model

    Cell Death & Disease 7:e2320.

    https://doi.org/10.1038/cddis.2016.223

    • PubMed
    • Google Scholar
42. 1. Plane JM
    2. Liu R
    3. Wang TW
    4. Silverstein FS
    5. Parent JM

    (2004) Neonatal hypoxic-ischemic injury increases forebrain subventricular zone neurogenesis in the mouse

    Neurobiology of Disease 16:585–595.

    https://doi.org/10.1016/j.nbd.2004.04.003

    • PubMed
    • Google Scholar
43. 1. Prasad SS
    2. Russell M
    3. Nowakowska M
    4. Williams A
    5. Yauk C

    (2012) Gene expression analysis to identify molecular correlates of pre- and post-conditioning derived neuroprotection

    Journal of Molecular Neuroscience 47:322–339.

    https://doi.org/10.1007/s12031-012-9751-3

    • PubMed
    • Google Scholar
44. 1. Roelants-Van Rijn AM
    2. van der Grond J
    3. de Vries LS
    4. Groenendaal F

    (2001) Value of 1H-MRS using different echo times in neonates with cerebral hypoxia-ischemia

    Pediatric Research 49:356–362.

    https://doi.org/10.1203/00006450-200103000-00009

    • PubMed
    • Google Scholar
45. 1. Roumes H
    2. Dumont U
    3. Sanchez S
    4. Mazuel L
    5. Blanc J
    6. Raffard G
    7. Chateil JF
    8. Pellerin L
    9. Bouzier-Sore AK

    (2021) Neuroprotective role of lactate in rat neonatal hypoxia-ischemia

    Journal of Cerebral Blood Flow and Metabolism 41:342–358.

    https://doi.org/10.1177/0271678X20908355

    • PubMed
    • Google Scholar
46. 1. Sejersted Y
    2. Hildrestrand GA
    3. Kunke D
    4. Rolseth V
    5. Krokeide SZ
    6. Neurauter CG
    7. Suganthan R
    8. Atneosen-Åsegg M
    9. Fleming AM
    10. Saugstad OD
    11. Burrows CJ
    12. Luna L
    13. Bjørås M

    (2011) Endonuclease VIII-like 3 (Neil3) DNA glycosylase promotes neurogenesis induced by hypoxia-ischemia

    PNAS 108:18802–18807.

    https://doi.org/10.1073/pnas.1106880108

    • PubMed
    • Google Scholar
47. 1. Semple BD
    2. Blomgren K
    3. Gimlin K
    4. Ferriero DM
    5. Noble-Haeusslein LJ

    (2013) Brain development in rodents and humans: Identifying benchmarks of maturation and vulnerability to injury across species

    Progress in Neurobiology 106–107:1–16.

    https://doi.org/10.1016/j.pneurobio.2013.04.001

    • PubMed
    • Google Scholar
48. 1. Shankaran S
    2. Pappas A
    3. McDonald SA
    4. Vohr BR
    5. Hintz SR
    6. Yolton K
    7. Gustafson KE
    8. Leach TM
    9. Green C
    10. Bara R
    11. Petrie Huitema CM
    12. Ehrenkranz RA
    13. Tyson JE
    14. Das A
    15. Hammond J
    16. Peralta-Carcelen M
    17. Evans PW
    18. Heyne RJ
    19. Wilson-Costello DE
    20. Vaucher YE
    21. Bauer CR
    22. Dusick AM
    23. Adams-Chapman I
    24. Goldstein RF
    25. Guillet R
    26. Papile L-A
    27. Higgins RD
    28. Eunice Kennedy Shriver NICHD Neonatal Research Network

    (2012) Childhood outcomes after hypothermia for neonatal encephalopathy

    The New England Journal of Medicine 366:2085–2092.

    https://doi.org/10.1056/NEJMoa1112066

    • PubMed
    • Google Scholar
49. 1. Shweiki D
    2. Itin A
    3. Soffer D
    4. Keshet E

    (1992) Vascular endothelial growth factor induced by hypoxia may mediate hypoxia-initiated angiogenesis

    Nature 359:843–845.

    https://doi.org/10.1038/359843a0

    • PubMed
    • Google Scholar
50. 1. Sidak Z

    (1967) Rectangular confidence regions for the means of multivariate normal distributions

    Journal of the American Statistical Association 62:626–633.

    https://doi.org/10.1080/01621459.1967.10482935

    • Google Scholar
51. 1. Stäubert C
    2. Broom OJ
    3. Nordström A

    (2015) Hydroxycarboxylic acid receptors are essential for breast cancer cells to control their lipid/fatty acid metabolism

    Oncotarget 6:19706–19720.

    https://doi.org/10.18632/oncotarget.3565

    • PubMed
    • Google Scholar
52. 1. Tassinari ID
    2. Andrade MKG
    3. da Rosa LA
    4. Hoff MLM
    5. Nunes RR
    6. Vogt EL
    7. Fabres RB
    8. Sanches EF
    9. Netto CA
    10. Paz AH
    11. de Fraga LS

    (2020) Lactate administration reduces brain injury and ameliorates behavioral outcomes following neonatal hypoxia-ischemia

    Neuroscience 448:191–205.

    https://doi.org/10.1016/j.neuroscience.2020.09.006

    • PubMed
    • Google Scholar
53. 1. Vinod KR
    2. Jones D
    3. Udupa V

    (2016) A simple and effective heat induced antigen retrieval method

    MethodsX 3:315–319.

    https://doi.org/10.1016/j.mex.2016.04.001

    • PubMed
    • Google Scholar
54. 1. Wang X
    2. Karlsson JO
    3. Zhu C
    4. Bahr BA
    5. Hagberg H
    6. Blomgren K

    (2001) Caspase-3 activation after neonatal rat cerebral hypoxia-ischemia

    Biology of the Neonate 79:172–179.

    https://doi.org/10.1159/000047087

    • PubMed
    • Google Scholar
55. 1. Wang W
    2. Osenbroch P
    3. Skinnes R
    4. Esbensen Y
    5. Bjørås M
    6. Eide L

    (2010) Mitochondrial DNA integrity is essential for mitochondrial maturation during differentiation of neural stem cells

    Stem Cells 28:2195–2204.

    https://doi.org/10.1002/stem.542

    • PubMed
    • Google Scholar
56. 1. Wen Y
    2. Yang S
    3. Liu R
    4. Simpkins JW

    (2005) Cell-cycle regulators are involved in transient cerebral ischemia induced neuronal apoptosis in female rats

    FEBS Letters 579:4591–4599.

    https://doi.org/10.1016/j.febslet.2005.07.028

    • PubMed
    • Google Scholar
57. 1. Wu TW
    2. Tamrazi B
    3. Hsu KH
    4. Ho E
    5. Reitman AJ
    6. Borzage M
    7. Blüml S
    8. Wisnowski JL

    (2018a) Cerebral lactate concentration in neonatal hypoxic-ischemic encephalopathy relation to time, characteristic of injury, and serum lactate concentration

    Frontiers in Neurology 9:11.

    https://doi.org/10.3389/FNEUR.2018.00293

    • Google Scholar
58. 1. Wu Y
    2. Wang M
    3. Zhang K
    4. Li Y
    5. Xu M
    6. Tang S
    7. Qu X
    8. Li C

    (2018b) Lactate enhanced the effect of parathyroid hormone on osteoblast differentiation via GPR81-PKC-Akt signaling

    Biochemical and Biophysical Research Communications 503:737–743.

    https://doi.org/10.1016/j.bbrc.2018.06.069

    • PubMed
    • Google Scholar
59. 1. Zhao S
    2. Guo Y
    3. Sheng Q
    4. Shyr Y

    (2014) Advanced heat map and clustering analysis using heatmap3

    BioMed Research International 2014:986048.

    https://doi.org/10.1155/2014/986048

    • PubMed
    • Google Scholar
60. 1. Zheng Y
    2. Wang XM

    (2017) Measurement of lactate content and amide proton transfer values in the basal ganglia of a neonatal piglet hypoxic-ischemic brain injury model using mri

    AJNR. American Journal of Neuroradiology 38:827–834.

    https://doi.org/10.3174/ajnr.A5066

    • PubMed
    • Google Scholar

Decision letter after peer review:

[Editors’ note: the authors submitted for reconsideration following the decision after peer review. What follows is the decision letter after the first round of review.]

Thank you for submitting your work entitled "Lactate receptor HCAR1 regulates neurogenesis and microglia activation after neonatal hypoxia-ischemia" for consideration by eLife. Your article has been reviewed by 3 peer reviewers, one of whom is a member of our Board of Reviewing Editors, and the evaluation has been overseen by a Senior Editor. The following individual involved in review of your submission has agreed to reveal their identity: Achira Roy (Reviewer #2).

We are sorry to say that, after consultation with the reviewers, we have decided that your work will not be considered further for publication by eLife, given the number of items that need to be addressed. Given the general interest and compelling initial findings from this study, the reviewers provide constructive feedback on data interpretation, methods, and specific experiments (specifically rescue experiments) to further support the role of HCAR1 and bring this paper to a level for publication in eLife.

Reviewer #1 (Recommendations for the authors):

This is a very interesting study implicating HCAR1, a G-protein coupled lactate receptor, in the pathophysiology of hypoxia-ischemia (HI)-induced impairments in brain tissue regeneration, cellular proliferation, neurogenesis and microglial activation. The authors use a HCAR1 KO mouse and provide convincing in vitro and in vivo evidence that newly proliferating cells in the intermediate zone (adjacent to SVZ) ipsilateral to the infarct/HI injury, as well as ipsilateral neurogenesis, is impaired after HI when HCAR1 is not present. Furthermore, microglial proliferation as well as activation is also reduced in the setting of HCAR1 KO. And finally, the authors provide RNA-sequencing data from HI-injured WT and HCAR1 KO mice, demonstrating that the significant differential gene expression occurring after HI in WT mice is not seen in HCAR1 KO mice. Overall, this study would be of major interest to scientists studying mechanisms of HI injury and intervention, possibly informing design of molecular targets to preserve cellular proliferation and tissue regeneration after ischemic injury.

Strengths:

The data presented here are very convincing that HCAR1 is required for tissue regeneration in the chronic period after HI injury (figure1). Figure 2 further shows evidence from both neurospheres and immunostaining that at the level of the intermediate zone, cellular proliferation and differentiation are directly impacted in the absence of HCAR1. Contralateral uninfarcted hemisphere is used as an intra-individual control, and very clearly demonstrates impact on proliferation and neurogenesis is specific to the hemisphere ipsilateral to the damage. It is interesting to note that HCAR1 absence alone does not impact proliferation and differentiation. Figure 3 shows that in the peri-infarct area, microglial activation and proliferation is repressed in HCAR1 KO mice. Figure 4 is descriptive in that it shows cell cycle pathway and complement pathway downregulation in HCAR1 KO SVZ from infarcted side of brain.

Weaknesses:

There is not significance found in the proliferative assessment of the SVZ, in that both WT and HCAR1 mice have reduced proliferation (did not reach significance, however, the figure Supp 1F suggests a strong trend). It is also interesting that there was a transcriptional response to HI in hippocampus in HCAR1 KO but this was not explored further. It would be a logical next step, and one that would enhance the paper, to disclose what the downstream genes discovered from the RNA-seq data set that are involved in neurogenesis and innate immunity that are not activated in HCAR1KO, as this would implicate a more specific mechanisms as to how HCAR1 regulates these processes.

The authors discuss lactate as a key metabolic signal in the mechanism of HI injury, and in their discussion and graphical illustration, propose that endogenous lactate is indeed increased. Demonstrating lactate increase in their model would provide more support for this hypothesis (if this is possible). Also, Figure 2o-r data are from the intermediate zone only; when SVZ is analyzed (Supp Figure 1) there is a reduction of the neural progenitors in the HCAR1 KO, but not a reduction in the proliferative cells. I would expect the proliferation from the subventricular zone to be reduced in HCAR1 KO mice, and this could be discussed further in the manuscript. Also- are the data comparing contralateral proliferation and differentiation in WT and KO mice statistically compared, to ensure there is no baseline change to proliferation and neurogenesis on that side of the brain? Additionally, this study is very descriptive as it stands now. A rescue experiment in the HI-induced HCAR1 KO, potentially with one of the differentially expressed genes discovered in their RNA seq data set, would make a much stronger mechanistic case for lactate-HCAR1 dependent expression of genes directly involved in proliferative or differentiation processes. I think this type of rescue experiment is necessary to move this paper toward publication. Finally, There is also an opportunity to look at hippocampal DG proliferation and differentiation here, and if it is impaired in HCAR1 mice, there can be again an attempt for rescue. Justification for this experiment is further supported by the data in Figure 4 and Supp Figure 6 showing that HCAR1 KO mice show a transcriptional response to HI in the hippocampus, but not in the subventricular zone.

Reviewer #2 (Recommendations for the authors):

Kennedy et al. investigated the functions of lactate receptor HCAR1 towards tissue repair and associated neurogenesis in a neonatal mouse model of cerebral hypoxia-ischemia (HI). The authors successfully showed that they can mimic the neonatal HI pathology in mouse. They clearly demonstrated that compared to controls HCAR1 knockout mouse brains fail to recover post-insult, even after days. Using RNA sequencing, the authors also identified differentially modulated molecular pathways, predominantly affecting cell cycle regulation and tissue repair, indicating a role of HCAR1 as a transcriptional regulator. Finally, differences in the repair mechanisms of neurogenic niches in hippocampal and rostral subventricular zones were also demonstrated.

Although this study has significant translational impact, following concerns in experimental design and data interpretation were observed:

1. Normal expression pattern of HCAR1 mRNA and protein and their lack in the mutant brain has not been shown. Without this evidence, comparison across regions of interest will be incomplete.

2. Since the cell density changes significantly across some groups, the remaining proportion calculations (Figure 2P-R) should be done as a fraction of total number of cells, rather than area. Otherwise, the very relevance of differential cell densities is lost.

3. Figure 2K-N and Supp Figure 1A-D are identical images; authors should remove one set or assemble all the data to Figure 2. Further, if both IZ and SVZ have proliferating cells at postnatal day (P)9, other markers specifically demarcating IZ and SVZ should be used to check any kind of zonal expansion/reduction.

4. In postnatal and adult neurogenesis, the cell cycle period is short ranging from 17-22hours. Especially in injury models, S-phase and in turn total cycle in pathological conditions like injury/stroke is further shortened to ~3hrs and 12hrs respectively (PMID: 29765834). So, the scientific basis of performing multi-day BrdU injections with 24h gap to determine proliferation rate seems not convincing. Unless the authors can provide evidence/reasoning that the used paradigm is essential, it would be critical to repeat all BrdU proliferation experiments with a shorter 1-2hr BrdU pulse, and then reinterpreted for proliferation and differentiation (cell cycle exit).

5. It is unclear what is meant by "DCX+ and Ki67+ or BrdU⁺ cells". Ki67 marks all cell cycle phases while BrdU is specific to S-phase. These cannot be counted and analyzed together as they mark different cell populations. Also, no image depicting Ki67-staining was provided in the paper.

6. Regular neurogenesis and astrogliogenesis in hippocampus continue perinatally in mouse. As the mouse experiments are done at P9, how do the authors distinguish the normal hippocampal processes from insult-driven processes? Can any co-labeling differentiate these two? Also, Figure 1A demonstrates that large part of ipsilateral midline takes the TTC stain, hence is viable. Can that be a reason for less effect of ischemia in the hippocampus of this model?

7. Neurospheres were generated from P3 whole forebrains of control and mutant mice and analyzed ~10 days in vitro, as per Methods. And, in vivo proliferation/differentiation assays were done in P9 mice post-HI injury. However, the trend of significantly different proliferation and differentiation rates between control and mutant seen in vitro is not reproduced in vivo. What can be the scientific explanation? How will that impact the overall data interpretation?

8. Emphasis should be given in the Abstract, Introduction and Discussion that this work is on postnatal neurogenesis to avoid confusion. Some statements in Discussion are only true for adult neurogenesis, and not for neonatal mouse brain as used in this study. Portions thus should be reviewed and modified accordingly.

9. For cell counting, co-labeling with nuclear markers is preferable to filament markers (b-tubulin, GFAP).

10. Increasing the size of Figure 2 J may help in clarity.

11. Embryonically, neural stem cells are also called primary progenitors or radial glial cells. Later in development (neonatal/adults), there are Type B progenitors/precursors (mostly dedifferentiated astrocytes), intermediate Type C cells and Type A stages for both neuron and glia. Which cell type did the authors in this study refer to by the term "neural stem-progenitor cells"?

12. Lettering of figures (capital letters) and figure legends do not match.

13. Sentences in many Figure legends are incomplete.

14. It may help to show representative images of sham-treated control and mutant brains for better comparison.

15. As this is an injury model, inspecting apoptosis rate in the proliferative zones will be important.

16. Supplementary Figures 4,5: If FDR{less than or equal to}0.05 is the cut-off, the reason of showing all the non-significant GSEA groups is unclear.

Reviewer #3 (Recommendations for the authors):

Kennedy et al. examine the role of the specific lactate receptor HCAR1 in the response to neonatal HI, both in the short- (24h) and long- (42 day) term. Use of KO helps to clarify the importance of this receptor in tissue recovery, and use of microglial and neuronal staining, as well as in vitro and in vivo analyses, approach this analysis with different methods which strengthen the study. In addition the use of RNAseq suggest differences in the transcriptome of a particular region of the brain (SVZ) theorized to be heavily involved in the response to HI.

The authors have an interesting hypothesis and make a case for conclusive findings, although I would not say that they succeed. They do support the findings that more work needs to be done to understand the role of this receptor, and that it may play a critical role in response to injury, but there needs to be a more comprehensive examination of cell-type specific effects (including astrocytes and oligos), in other regions of the brain, and also the downstream signaling pathways need to be further evaluated to get at the mechanism of repair. For example, qPCR and protein quantification of specific factors and pathways. This is a good first step, but more detailed analyses are necessary to confirm.

If true, this would be an interesting finding in understanding injury and repair in the developing brain.

I will comment based on line/section of the manuscript.

47: I think your use of "stem cell" here and later in the manuscript is somewhat vague. You should probably stick to precursor or progenitor cell, or more clearly define the term and how you use it.

53: True that SVZ and SGZ are most well-described, but more recent evidence suggest role for local response to injury, also sub-pial and other regions.

82: Question – this model includes permanent occlusion of L CCA. What do you think is the role of reperfusion injury (as occurs in humans) on this particular receptor and response?

93: I'm not sure I understand the use of showing damage in individual sub-sections. You should consider including specific regions of the brain rather than a particular section, or remove altogether.

Section starting at 112: The use of in vitro neurospheres are interesting way to approach neurogenesis, where you show decreased proliferation at baseline. But the overall reduction in neurogenesis suggested by this manuscript don't conclusively show to me, if it is a decrease at baseline or a decrease in the response to injury, or both. Did you do OGD in vitro to see if there was also a decrease in the proportional response to injury?

131: How did you decide on BrdU on days 4-7? Why not start earlier? Also, a more thorough analysis of cell fate including cell types other than neurons and microglia would be useful, to see specific effects on immature/mature oligos and astrocytes after injury.

Section 179: Did you consider scRNAseq? I wonder if there are cell-type specific differences in transcription. Also, it is a fairly big leap to make any conclusions on the pathways based on bulk RNAseq alone. You need qPCR or protein quantification to draw more conclusions. Related, your wording in discussion needs to be softened in regards to mechanism without more conclusive evidence.

[Editors’ note: further revisions were suggested prior to acceptance, as described below.]

Thank you for resubmitting your work entitled "Lactate receptor HCAR1 regulates neurogenesis and microglia activation after neonatal hypoxia-ischemia" for further consideration by eLife. Your revised article has been evaluated by Marianne Bronner (Senior Editor) and a Reviewing Editor.

The manuscript has been improved but there are some remaining issues that need to be addressed, as outlined below:

- Please address each Reviewer's comments, as these are overall addressable issues. Most of these comments request clearer explanations in the manuscript text, or clarifications. Addressing these comments will greatly improve this manuscript, thank you.

Reviewer #2 (Recommendations for the authors):

This is a revised version of the manuscript that was previously submitted to eLife. Overall, compared to the original version, the authors largely addressed one major experimental suggestion and improved textual explanation of certain parts of the manuscript. However, majority of the remaining critical concerns remain inadequately addressed. There is also a mild unwanted tendency to fit the data interpretation as per some pre-fixed hypotheses/ideas. Hence some of the concerns that were pointed out in last review are reiterated here along with some new ones.

1. Figures 2,3,4, S1: BrdU marks cells undergoing DNA synthesis. As the authors mentioned, it is true that there are many BrdU protocols; however, each one has a logical basis for separate usage. The protocol that was used in this manuscript spans over 3-4 days, this will only provide information about the total number of cells that have undergone DNA synthesis during this period. This does NOT provide any information about the proliferation rate, unless other proliferation markers are also co-studied appropriately (like Hayashi et al. 2016, cited by the authors themselves). BrdU⁺ cells will exit cell-cycle within 3-12hrs (becoming non-proliferating); hence it is incorrect to state BrdU⁺ cells as "all proliferating cells" under the current experimental paradigm. This was already brought up in the last review, but was not satisfactorily addressed. Further, there is lack of clarity on the experimental motive – is this done to study proliferation rate or just to have a bigger BrdU⁺ cell pool?

2. Proper differentiation/cell exit assays that is required in order to comment on regulation of cell differentiation are lacking. Just calculating mature cells in a cohort (Figure 2I) is not a measure of differentiation. Additionally, GFAP is also a marker of early neural progenitors and glial precursors, not just a marker of mature astrocytes.

3. The authors proposed in the address to reviewers' comments that neurosphere assay may have several constraints that are not faced in vivo and hence they are giving different results. It is then unclear what biological relevance the authors want to show with the neurosphere assay data.

4. Cell density is a very important measurement, on which calculation of other cell proportions depend. Proportion of different cell types (Figure 2P-R) cannot be measured as absolute numbers when cell density is altered in a comparative group. This was also requested for reanalysis previously. Since the authors failed/were reluctant to perform a simple data reanalysis, here is a small illustrative example in the hope of getting the correct interpretation out:

Similar to the data scenario, suppose

WT brain: 6 total cells/unit area; 3 cell-type A/unit area

Mut brain: 6 total cells/unit area; 3 cell-type A/unit area

After HI,

WT(HI): become 10 total cells/unit area; 5 cell-type A/unit area

Mut(HI): remain at 6 total cells/unit area; 3 cell-type A/unit area

a. Analysis for cell-type A done as absolute number normalizing across area (like the manuscript): cell-type A in WT(HI) (5) is greater than that in all other groups (3).

b. Analysis for cell-type A done as proportion of total number of cells: All groups have same proportion of cell-type A (3/6=2 or 5/10=2).

The second type of analysis does not mask the effect of the data, as claimed by the authors; rather it is the only way to bring out the actual picture in view. This data analysis is required to be modified accordingly.

5. What was the landmark/marker used to precisely dissect striatal subventricular zone tissue for RNA sequencing? It is not clear why the RNA sequencing was done 3 days after HI while all the proliferation/neurogenesis experiments were performed 7days after HI despite hippocampus being damaged – any biological reason for this choice of ages?

6. There are multiple instances throughout the manuscript where data is interpreted without giving enough value to the related statistics, leading to misleading statements. Some examples are as follows, which need to be rectified:

Lines 122-123: The average size of HCAR1 KO spheres also tended to be smaller, although this was not statistically significant.

Lines 204-206: On the other hand, the contralateral hemisphere in HCAR1 KO mice was on average higher than in WT mice (albeit not statistically significant).

Lines 269-270: The average B1 levels were 78% lower in the HCAR1 KO compared with WT, but this result was not statistically significant (p=0.057)

Reviewer #3 (Recommendations for the authors):

This is a very interesting and important study testing the role of the lactate receptor HCAR1 in recovery after neonatal hypoxia-ischemia using a HCAR1 knockout. The authors have been thoughtful and thorough in their response to the previous reviews. They show significantly increased infarct size in the KO compared with wild-type animals, and then proceed to examine and demonstrate remarkable effects of the KO on processes likely to be involved in repair following injury: proliferation of neural progenitor cells, as well as astrocytes, oligodendrocytes, microglia; activation of microglia; and transcriptional activation in the subventricular zone. The authors are really to be commended for their wide-ranging analysis of the impact of HCAR1 KO. The study is largely descriptive, but this is not a liability as this is the foundation required for future studies. For example, given what they show, it would be of great interest to determine whether conditional knockout of HCAR1 in a particular cell type recapitulates the phenotype of the constitutive KO. One important issue is that the authors assume that deficits in repair are largely responsible for the differences in outcome between the WT and KO animals. However, in adult stroke, it is well-known that there is a penumbra of injured cells around the central infarcted area, and a great deal of work has been focused on the issue of saving the cells in the penumbra from being irreversibly committed to death. There is no mention of this issue in the manuscript, but it is an important problem, because it is conceivable that the effects of HCAR1 are entirely due to the protection of cells surviving the initial insult rather than the generation of new cells. At a minimum, the authors should address this issue in the discussion.

[Editors’ note: the authors resubmitted a revised version of the paper for consideration. What follows is the authors’ response to the first round of review.]

Reviewer #1 (Recommendations for the authors):

This is a very interesting study implicating HCAR1, a G-protein coupled lactate receptor, in the pathophysiology of hypoxia-ischemia (HI)-induced impairments in brain tissue regeneration, cellular proliferation, neurogenesis and microglial activation. The authors use a HCAR1 KO mouse and provide convincing in vitro and in vivo evidence that newly proliferating cells in the intermediate zone (adjacent to SVZ) ipsilateral to the infarct/HI injury, as well as ipsilateral neurogenesis, is impaired after HI when HCAR1 is not present. Furthermore, microglial proliferation as well as activation is also reduced in the setting of HCAR1 KO. And finally, the authors provide RNA-sequencing data from HI-injured WT and HCAR1 KO mice, demonstrating that the significant differential gene expression occurring after HI in WT mice is not seen in HCAR1 KO mice. Overall, this study would be of major interest to scientists studying mechanisms of HI injury and intervention, possibly informing design of molecular targets to preserve cellular proliferation and tissue regeneration after ischemic injury.

Strengths:

The data presented here are very convincing that HCAR1 is required for tissue regeneration in the chronic period after HI injury (figure1). Figure 2 further shows evidence from both neurospheres and immunostaining that at the level of the intermediate zone, cellular proliferation and differentiation are directly impacted in the absence of HCAR1. Contralateral uninfarcted hemisphere is used as an intra-individual control, and very clearly demonstrates impact on proliferation and neurogenesis is specific to the hemisphere ipsilateral to the damage. It is interesting to note that HCAR1 absence alone does not impact proliferation and differentiation. Figure 3 shows that in the peri-infarct area, microglial activation and proliferation is repressed in HCAR1 KO mice. Figure 4 is descriptive in that it shows cell cycle pathway and complement pathway downregulation in HCAR1 KO SVZ from infarcted side of brain.

We thank the Reviewer for acknowledging that the study will be of major interest to the field and for pointing out that the data are very convincing in showing an effect of HCAR1 on tissue regeneration and cell proliferation after HI injury.

Weaknesses:

There is not significance found in the proliferative assessment of the SVZ, in that both WT and HCAR1 mice have reduced proliferation (did not reach significance, however, the figure Supp 1F suggests a strong trend). It is also interesting that there was a transcriptional response to HI in hippocampus in HCAR1 KO but this was not explored further. It would be a logical next step, and one that would enhance the paper, to disclose what the downstream genes discovered from the RNA-seq data set that are involved in neurogenesis and innate immunity that are not activated in HCAR1KO, as this would implicate a more specific mechanisms as to how HCAR1 regulates these processes.

The authors discuss lactate as a key metabolic signal in the mechanism of HI injury, and in their discussion and graphical illustration, propose that endogenous lactate is indeed increased. Demonstrating lactate increase in their model would provide more support for this hypothesis (if this is possible).

Elevated lactate in the ipsilateral hemisphere after HI has been demonstrated previously in a similar mouse HI model (Mikrogeorgiou et al., Dev Neurosci 2020, https://doi.org/10.1159/000506982). We have now added this point to the beginning of the discussion (page 11).

Also, Figure 2o-r data are from the intermediate zone only; when SVZ is analyzed (Supp Figure 1) there is a reduction of the neural progenitors in the HCAR1 KO, but not a reduction in the proliferative cells. I would expect the proliferation from the subventricular zone to be reduced in HCAR1 KO mice, and this could be discussed further in the manuscript.

We agree that an effect of HCAR1 was expected to be seen in the more basal VZ/SVZ immunolabeling. In the revised manuscript, we have increased the number of animals that were included in the analysis (Supp Figure 1). However, this did not lead to a statistically significant effect. There remains some controversy and disagreement in naming the SVZ, especially in the developing brain where the VZ is in transition from embryonic to the adult state. We believe both areas represent part of the sub-ventricular niche and have therefore renamed the areas to VZ/SVZ and SVZ/IZ, as has been done by several other investigators. Furthermore, Sejersted et al. 2011 (https://doi.org/10.1073/pnas.1106880108) reported the same findings using the exact same HImodel. As requested, we have now discussed this in the manuscript (page 12).

Also- are the data comparing contralateral proliferation and differentiation in WT and KO mice statistically compared, to ensure there is no baseline change to proliferation and neurogenesis on that side of the brain?

Yes, contralateral sides in KO versus WT were compared (described under Materials and methods in the section “Statistical Analysis”). There were no statistically significant differences between the contralateral hemispheres.

Additionally, this study is very descriptive as it stands now. A rescue experiment in the HI-induced HCAR1 KO, potentially with one of the differentially expressed genes discovered in their RNA seq data set, would make a much stronger mechanistic case for lactate-HCAR1 dependent expression of genes directly involved in proliferative or differentiation processes. I think this type of rescue experiment is necessary to move this paper toward publication.

We think that this experiment would be unlikely to succeed, given the high number of differentially expressed genes: we detected 6440 differentially expressed genes in the subventricular zone, of which 3182 were down regulated. Only related to the cell cycle, we found 53 down regulated genes. Therefore, we believe that upregulation of a single gene within this system would be highly unlikely to have a measurable effect. We have been in contact with the editor, who allowed us to resubmit without addressing this point.

Finally, there is also an opportunity to look at hippocampal DG proliferation and differentiation here, and if it is impaired in HCAR1 mice, there can be again an attempt for rescue. Justification for this experiment is further supported by the data in Figure 4 and Supp Figure 6 showing that HCAR1 KO mice show a transcriptional response to HI in the hippocampus, but not in the subventricular zone.

We agree that this would be interesting. Unfortunately, the hippocampus was completely degenerated on day 7 after HI (the time of fixation for immunostaining) in nearly all the mice that had undergone HI. Therefore, it was not possible to obtain comparable immunostaining data from the hippocampus. We have now mentioned this observation in the Results section (page 6). The reason we could still obtain RNA sequencing data from the hippocampus, was that tissue for RNAseq was isolated on day 3 after HI. We chose this difference in timing between RNAseq and immunostaining to be able to detect pathways that had been activated before most of the cell proliferation had taken place.

Reviewer #2 (Recommendations for the authors):

Kennedy et al. investigated the functions of lactate receptor HCAR1 towards tissue repair and associated neurogenesis in a neonatal mouse model of cerebral hypoxia-ischemia (HI). The authors successfully showed that they can mimic the neonatal HI pathology in mouse. They clearly demonstrated that compared to controls HCAR1 knockout mouse brains fail to recover post-insult, even after days. Using RNA sequencing, the authors also identified differentially modulated molecular pathways, predominantly affecting cell cycle regulation and tissue repair, indicating a role of HCAR1 as a transcriptional regulator. Finally, differences in the repair mechanisms of neurogenic niches in hippocampal and rostral subventricular zones were also demonstrated.

Although this study has significant translational impact, following concerns in experimental design and data interpretation were observed:

We appreciate that the reviewer acknowledges the significant translational impact of the study.

1. Normal expression pattern of HCAR1 mRNA and protein and their lack in the mutant brain has not been shown. Without this evidence, comparison across regions of interest will be incomplete.

mRNA expression of HCAR1, and its lack in the mutant brain, has been previously confirmed by us (Supplementary Table 1 in Morland et al., Nat Comm 2017, https://doi.org/10.1038/ncomms15557). Moreover, HCAR1 RFP reporter mice have been used to show HCAR1 expression in the brain, including the lining of the subventricular zone (Hadzic et al., Int J Mol Sci 2020, https://doi.org/10.3390/IJMS21186457 ).

Although several antibodies to HCAR1 are available commercially, none have proven to be specific (de Castro Abrantes, J Neurosci 2019, https://doi.org/10.1523/JNEUROSCI.2092-18.2019). Therefore, a further confirmation of HCAR1 protein expression in the different brain areas by immunostaining or western blot is not feasible.

2. Since the cell density changes significantly across some groups, the remaining proportion calculations (Figure 2P-R) should be done as a fraction of total number of cells, rather than area. Otherwise, the very relevance of differential cell densities is lost.

We disagree with this. Since we see that the total cell density increases after HI in WT, but not in KO (Figure 2O), we believe that showing the different cell types as a proportion of total cells instead of area would mask part of the effect.

3. Figure 2K-N and Supp Figure 1A-D are identical images; authors should remove one set or assemble all the data to Figure 2. Further, if both IZ and SVZ have proliferating cells at postnatal day (P)9, other markers specifically demarcating IZ and SVZ should be used to check any kind of zonal expansion/reduction.

We have removed the images in Supplementary Figure 1A-D, as requested.

There are, as far as we know, no clear consensus or markers to clearly define the layers of the SVZ and IZ. We agree that a more comprehensive analysis of the different cell populations and their density and possible zonal expansion would be interesting, but is in our opinion beyond the scope of this paper. We chose to count type A Neuroblasts as they are determined to a neuronal fate and thus tell us whether there is an effect on neurogenesis from HCAR1 before and after HI.

4. In postnatal and adult neurogenesis, the cell cycle period is short ranging from 17-22hours. Especially in injury models, S-phase and in turn total cycle in pathological conditions like injury/stroke is further shortened to ~3hrs and 12hrs respectively (PMID: 29765834). So, the scientific basis of performing multi-day BrdU injections with 24h gap to determine proliferation rate seems not convincing. Unless the authors can provide evidence/reasoning that the used paradigm is essential, it would be critical to repeat all BrdU proliferation experiments with a shorter 1-2hr BrdU pulse, and then reinterpreted for proliferation and differentiation (cell cycle exit).

Several different protocols exist for BrdU analyses. We are unsure of the exact experimental setup suggested by the reviewer and the reasoning for it.

1) If the reviewer here means that we should have performed a single 1-2hr BrdU pulse, we think this would have yielded a lower total density of countable proliferated cells compared with our setup with BrdU injections over several days. Therefore, our paradigm increases the strength of our BrdU analysis due to of the higher number of total counted cells.

2) If the reviewer means that we should have performed injections with 1-2hr intervals over several days, we agree that this would have increased our total number of quantified cells. Thus, this might have led us to detect even larger differences between KO and WT. So it is possible that our setup leads to some underestimation of the differences between KO and WT. However, such a setup would have put much more strain on the animals and could have led to unwanted stress responses.

Thus, we do not see any reason that our setup would lead to erroneous results. In support of our experimental setup, such (less frequent) multi day BrdU injections have been commonly used for similar purposes (Plane et al., Neurobiol Dis 2004, https://doi.org/10.1016/j.nbd.2004.04.003; Hayashi et al., Brain Res 2005 https://doi.org/10.1016/j.brainres.2004.12.048; Palibrk et al., Cell Death Dis 2016, https://doi.org/10.1038/cddis.2016.223).

5. It is unclear what is meant by “DCX+ and Ki67+ or BrdU⁺ cells”. Ki67 marks all cell cycle phases while BrdU is specific to S-phase. These cannot be counted and analyzed together as they mark different cell populations. Also, no image depicting Ki67-staining was provided in the paper.

We agree. The analysis was done in the previous manuscript due to shortage of BrdU treated samples, and the two groups were pooled together to reflect all proliferation both asymmetrically and symmetrical of the neuroblasts. We have now increased sample size, excluded the Ki67 samples and show only BrdU in this figure.

6. Regular neurogenesis and astrogliogenesis in hippocampus continue perinatally in mouse. As the mouse experiments are done at P9, how do the authors distinguish the normal hippocampal processes from insult-driven processes? Can any co-labeling differentiate these two?

We understand this question to be about the subventricular niche. The “basal” rate of peri/postnatal neurogenesis is determined by comparing the contralateral (undamaged) side between the genotypes. There were no statistically significant differences.

Also, Figure 1A demonstrates that large part of ipsilateral midline takes the TTC stain, hence is viable. Can that be a reason for less effect of ischemia in the hippocampus of this model?

Although the midline is viable, the hippocampus appeared severely injured at later stages and was often completely lost in our section on day 7 after HI (not shown, but now included in main text, page 6). Therefore, we do not believe that the hippocampus was less affected by ischemia compared with the SVZ. This is supported by the RNA sequencing data (from day 3 after HI), which shows a strong transcriptional response to HI in the hippocampus (see PCA plot in Suppl. Figure X and number of DEGs in Supplementary Table 1). Rather, the transcriptional response to HI in the hippocampus is similar between WT and HCAR1 KO mice, whereas in the SVZ there is a reduced response to HI in HCAR1 KO mice. Thus, the response to HI appears to be HCAR1 dependent in SVZ, but not in the hippocampus. These findings are supported by a recent publication showing an effect of HCAR1 on neurogenesis in the subventricular zone and not in the hippocampal subgranular zone (Lambertus et al. 2020, Acta Physiologica. https://doi.org/10.1111/apha.13587).

7. Neurospheres were generated from P3 whole forebrains of control and mutant mice and analyzed ~10 days in vitro, as per Methods. And, in vivo proliferation/differentiation assays were done in P9 mice post-HI injury. However, the trend of significantly different proliferation and differentiation rates between control and mutant seen in vitro is not reproduced in vivo. What can be the scientific explanation? How will that impact the overall data interpretation?

Neurospheres are isolated clusters of neural stem cells that lack the other components of the brain (including a vascular system). One reason for the different results between neurospheres and in vivo (contralateral hemispheres) could be compensatory mechanisms in vivo that are not present in vitro. Also, the process of creating the neurosphere assay from dissection, cell dissociation and culturing might cause a stress response in the cells that does not occur in the contralateral hemispheres in vivo. This has now been discussed in more detail in the manuscript (page 13).

8. Emphasis should be given in the Abstract, Introduction and Discussion that this work is on postnatal neurogenesis to avoid confusion. Some statements in Discussion are only true for adult neurogenesis, and not for neonatal mouse brain as used in this study. Portions thus should be reviewed and modified accordingly.

We agree. Changes have been made in the manuscript to clarify.

9. For cell counting, co-labeling with nuclear markers is preferable to filament markers (b-tubulin, GFAP).

We agree. However, we have chosen our antibodies based on our experience and their reliability. Except for neuroblasts, which often lie in clusters, we have used automatic counting using advanced image segmentation with WEKA-trainable segmentation tool and automatic thresholding algorithms combined with our own scripts, which exclude small objects/not whole cells and which are able to do overlap-analysis with DAPI and BrdU. Thus, our image analysis tools gives us the same possibility to count cells as one would with a nuclear marker. (If needed examples of the steps in image analysis can be provided).

10. Increasing the size of Figure 2 J may help in clarity.

This has been fixed.

11. Embryonically, neural stem cells are also called primary progenitors or radial glial cells. Later in development (neonatal/adults), there are Type B progenitors/precursors (mostly dedifferentiated astrocytes), intermediate Type C cells and Type A stages for both neuron and glia. Which cell type did the authors in this study refer to by the term “neural stem-progenitor cells”?

The term was written to include the in vivo and in vitro data we present. Our in vivo data present SVZ derived type A Neuroblasts (progenitors). Before differentiation, neurospheres consist mainly of neural stem and progenitor cells (NSPCs). We have now explained this point in a separate section under Materials and methods (page 20). For simplicity in the main text, we have termed the cells “progenitor cells” (as requested by Reviewer 3).

12. Lettering of figures (capital letters) and figure legends do not match.

13. Sentences in many Figure legends are incomplete.

Fixed.

14. It may help to show representative images of sham-treated control and mutant brains for better comparison.

As the contralateral hemisphere of our hypoxic-ischemia treated animals has been presented and validated as a control previously (Sejerstad et al., PNAS 2011, https://doi.org/10.1073/pnas.1106880108), we have not performed sham treatment for our injury analyses (Figure 1) or immunohistochemistry experiments (Figure 2,3,4).

15. As this is an injury model, inspecting apoptosis rate in the proliferative zones will be important.

Our data suggest that the major difference between the genotypes lies in the regeneration and innate immune response rather than in initial cell death and apoptosis. This is supported by the measurements of acute injury (Figure 1A) as well as the RNA sequencing from SVZ three days post injury, in which apoptosis was not among the altered pathways (Figure 5B). Although it could still be interesting to inspect apoptosis rates directly, we feel that this is not necessary for the conclusions in this manuscript.

16. Supplementary Figures 4,5: If FDR{less than or equal to}0.05 is the cut-off, the reason of showing all the non-significant GSEA groups is unclear.

Although the data are not statistically significant with our cut-off, they show tendencies that could have biological significance. We have therefore chosen to keep them in the supplement.

Reviewer #3 (Recommendations for the authors):

Kennedy et al. examine the role of the specific lactate receptor HCAR1 in the response to neonatal HI, both in the short- (24h) and long- (42 day) term. Use of KO helps to clarify the importance of this receptor in tissue recovery, and use of microglial and neuronal staining, as well as in vitro and in vivo analyses, approach this analysis with different methods which strengthen the study. In addition the use of RNAseq suggest differences in the transcriptome of a particular region of the brain (SVZ) theorized to be heavily involved in the response to HI.

The authors have an interesting hypothesis and make a case for conclusive findings, although I would not say that they succeed. They do support the findings that more work needs to be done to understand the role of this receptor, and that it may play a critical role in response to injury, but there needs to be a more comprehensive examination of cell-type specific effects (including astrocytes and oligos), in other regions of the brain, and also the downstream signaling pathways need to be further evaluated to get at the mechanism of repair. For example, qPCR and protein quantification of specific factors and pathways. This is a good first step, but more detailed analyses are necessary to confirm.

If true, this would be an interesting finding in understanding injury and repair in the developing brain.

I will comment based on line/section of the manuscript.

47: I think your use of "stem cell" here and later in the manuscript is somewhat vague. You should probably stick to precursor or progenitor cell, or more clearly define the term and how you use it.

Agreed. We have now changed the text to say “progenitor cell” throughout the manuscript. In addition, we have added a small section under Materials and methods (page 20) where we specify the types of progenitors in vivo and in vitro.

53: True that SVZ and SGZ are most well-described, but more recent evidence suggest role for local response to injury, also sub-pial and other regions.

We have now changed the wording in the introduction as to not exclude other neurogenic areas and we have mentioned these in the discussion (page 12).

82: Question – this model includes permanent occlusion of L CCA. What do you think is the role of reperfusion injury (as occurs in humans) on this particular receptor and response?

Recent data from adult rats suggest that HCAR1 is not involved in the effects of lactate on reperfusion injury (Buscemi et al., Front Physiol 2021, https://doi.org/10.3389/fphys.2021.689239), but this has not been tested in neonatal HI as far as we know. We have now discussed this topic in the manuscript (page 12).

93: I'm not sure I understand the use of showing damage in individual sub-sections. You should consider including specific regions of the brain rather than a particular section, or remove altogether.

We agree that the sub-sections figure is not so informative. Unfortunately the large damage on the ipsilateral side makes it near-impossible to measure specific regions on this side of the brain. Some regions, especially the hippocampus is completely missing 42 days after HI in many sections. Therefore, we have chosen to remove the measurements of sub-sections from Figure 1. We mention the loss of hippocampus in the text, but have not quantified this, for the reason explained here.

Section starting at 112: The use of in vitro neurospheres are interesting way to approach neurogenesis, where you show decreased proliferation at baseline. But the overall reduction in neurogenesis suggested by this manuscript don't conclusively show to me, if it is a decrease at baseline or a decrease in the response to injury, or both. Did you do OGD in vitro to see if there was also a decrease in the proportional response to injury?

We have not done OGD in vitro. The discrepancy between the neurosphere data and the in vivo control data (contralateral hemispheres) could be due to compensatory mechanisms in vivo or possibly (unintended) stress to the cells caused by the neurosphere protocol. We have now discussed this in the manuscript (page 13) and in the response to reviewer #2 above.

131: How did you decide on BrdU on days 4-7? Why not start earlier? Also, a more thorough analysis of cell fate including cell types other than neurons and microglia would be useful, to see specific effects on immature/mature oligos and astrocytes after injury.

Various protocols exist for BrdU injections after ischemic injury, with the starting point mostly varying between one and five days after the induced injury. One argument for starting around day 35 post injury is to avoid BrdU incorporation into cells that are taking up BrdU due to DNA repair (rather than proliferation, Ong et al., Pediatr Res 2005, https://doi.org/10.1203/01.PDR.0000179381.86809.02; Cooper-Kuhn and Kuhn, Dev Brain Res 2002, https://doi.org/10.1016/S0165- 3806(01)00243-7), which presumably occurs more frequently in the first days after the injury. Nevertheless, we agree that starting earlier with injections might have shown larger differences between WT and HCAR1 KO mice since more cells would have been included in the analyses. Thus, our data may underestimate the effect of HCAR1 on ischemia induced proliferation.

We thank the reviewer for suggesting to include oligodendrocytes and astrocytes in the analyses. We have now performed these experiments (Figure 4). We find a significant increase in the proliferation of astrocytes and oligodendrocytes in WT, but not in HCAR1 KO mice. We believe these data further strengthen the current study by underlining the role of HCAR1 in promoting overall cell proliferation after HI.

Section 179: Did you consider scRNAseq? I wonder if there are cell-type specific differences in transcription. Also, it is a fairly big leap to make any conclusions on the pathways based on bulk RNAseq alone. You need qPCR or protein quantification to draw more conclusions. Related, your wording in discussion needs to be softened in regards to mechanism without more conclusive evidence.

Although single cell RNA sequencing would be very interesting to investigate the cell-specific effects of HCAR1, this is beyond the scope of the current study. Given that we have now analysed proliferation of different glial cells by immunolabelling (previous point), we feel that we have partly addressed the cell-specific effects of HCAR1. We have been in contact with the editor, who allowed us to resubmit without performing scRNAseq.

As suggested, we have now confirmed the changes in some cell cycle genes on protein level, by western blotting (Figure 5F-H). In addition, we have softened the claims in the discussion by mentioning the need for further confirmation on the protein level (bottom of page 13).

[Editors’ note: what follows is the authors’ response to the second round of review.]

The manuscript has been improved but there are some remaining issues that need to be addressed, as outlined below:

Reviewer #2 (Recommendations for the authors):

This is a revised version of the manuscript that was previously submitted to eLife. Overall, compared to the original version, the authors largely addressed one major experimental suggestion and improved textual explanation of certain parts of the manuscript. However, majority of the remaining critical concerns remain inadequately addressed. There is also a mild unwanted tendency to fit the data interpretation as per some pre-fixed hypotheses/ideas. Hence some of the concerns that were pointed out in last review are reiterated here along with some new ones.

1. Figures 2,3,4, S1: BrdU marks cells undergoing DNA synthesis. As the authors mentioned, it is true that there are many BrdU protocols; however, each one has a logical basis for separate usage. The protocol that was used in this manuscript spans over 3-4 days, this will only provide information about the total number of cells that have undergone DNA synthesis during this period. This does NOT provide any information about the proliferation rate, unless other proliferation markers are also co-studied appropriately (like Hayashi et al. 2016, cited by the authors themselves). BrdU⁺ cells will exit cell-cycle within 3-12hrs (becoming non-proliferating); hence it is incorrect to state BrdU⁺ cells as "all proliferating cells" under the current experimental paradigm. This was already brought up in the last review, but was not satisfactorily addressed. Further, there is lack of clarity on the experimental motive – is this done to study proliferation rate or just to have a bigger BrdU⁺ cell pool?

Our motive to use the current BrdU protocol was to compare density of proliferated cells after HI between WT and HCAR1 KO mice. If we understand the reviewer correctly, the issue is not with our BrdU protocol, but with our wording, where we inaccurately had used the terms “proliferation rate” and “all proliferating cells” at two places in the manuscript. We have now replaced “Reduced proliferation rate” with “less proliferation” (line 123). “All proliferating cells” is replaced with “proliferated cells” (Figure legend for Figure 2).

2. Proper differentiation/cell exit assays that is required in order to comment on regulation of cell differentiation are lacking. Just calculating mature cells in a cohort (Figure 2I) is not a measure of differentiation. Additionally, GFAP is also a marker of early neural progenitors and glial precursors, not just a marker of mature astrocytes.

We have softened our wording throughout the manuscript such that we no longer claim an effect of HCAR1 on differentiation (e.g. line 131). We have also changed the wording to say GFAP+ cells rather than astrocytes and have mentioned the fact that GFAP also labels early neural progenitors (Figure legend for Figure 2 and line 127)

3. The authors proposed in the address to reviewers' comments that neurosphere assay may have several constraints that are not faced in vivo and hence they are giving different results. It is then unclear what biological relevance the authors want to show with the neurosphere assay data.

The neurospheres offer a simplified and isolated in vitro system where proliferation, self-renewal and differentiation can be tested in a controlled environment. This is now mentioned in the manuscript (line 119).

4. Cell density is a very important measurement, on which calculation of other cell proportions depend. Proportion of different cell types (Figure 2P-R) cannot be measured as absolute numbers when cell density is altered in a comparative group. This was also requested for reanalysis previously. Since the authors failed/were reluctant to perform a simple data reanalysis, here is a small illustrative example in the hope of getting the correct interpretation out:

Similar to the data scenario, suppose

WT brain: 6 total cells/unit area; 3 cell-type A/unit area

Mut brain: 6 total cells/unit area; 3 cell-type A/unit area

After HI,

WT(HI): become 10 total cells/unit area; 5 cell-type A/unit area

Mut(HI): remain at 6 total cells/unit area; 3 cell-type A/unit area

a. Analysis for cell-type A done as absolute number normalizing across area (like the manuscript): cell-type A in WT(HI) (5) is greater than that in all other groups (3).

b. Analysis for cell-type A done as proportion of total number of cells: All groups have same proportion of cell-type A (3/6=2 or 5/10=2).

The second type of analysis does not mask the effect of the data, as claimed by the authors; rather it is the only way to bring out the actual picture in view. This data analysis is required to be modified accordingly.

We have now performed the requested analyses (Figure 2P-R, Figure 3F-G and Figure 4E-Q, and results are described in the main text).

The results are mostly the same as previously, except for in oligodendrocytes, where there is no longer a significant increase in Olig2 cells after HI and no difference between WT and HCAR1 KO. Thus, the main conclusion of the study remains the same.

However, we do not agree with the reviewer that this is the only way to bring out the actual picture. Cells/DAPI (as requested by the reviewer) will say how the labelled cell type (e.g. Olig2 cells) changed in comparison with all cells (DAPI), whereas cells/unit area (our “old” analyses) will say how this cell type changed overall (independently of all the other cells). Thus, for Olig2 cells, we see that these cells increased in overall density (cells/unit area) after HI, but since all cells (i.e. DAPI+ cells) increased equally much, olig2/DAPI is not increased. We still think that the overall density of a cell type physiologically relevant. We have therefore chosen to include our old analysis as supplement to Figures2-4 and have included a discussion of the differences between the two analyses (paragraph starting on line 359 of the discussion). Of note, our “old” analysis, now in the supplement, is commonly used to study neurogenesis (see for instance Hoshi et al., Science Advances 2021 https://doi.org/10.1126/sciadv.abj8080 or Lambertus et al., Acta Physiologica 2021 https://doi.org/10.1111/apha.13587).

5. What was the landmark/marker used to precisely dissect striatal subventricular zone tissue for RNA sequencing? It is not clear why the RNA sequencing was done 3 days after HI while all the proliferation/neurogenesis experiments were performed 7days after HI despite hippocampus being damaged – any biological reason for this choice of ages?

The Corpus callosum and the shape of the lateral ventricles were used as landmarks for SVZ dissection. A detailed description of the SVZ dissection has now been included under “RNA sequencing and analysis” in the Methods section.

We chose to perform immunostaining 7 days after HI because we wanted to detect proliferating cells between days 4 and 7 after HI by BrdU injections. The rationale for this timing was partly to avoid BrdU incorporation into cells that are taking up BrdU due to DNA repair (rather than proliferation, Ong et al., 2005 DOI: 10.1203/01.PDR.0000179381.86809.02; Cooper-Kuhn and Kuhn 2002 https://doi.org/10.1016/S0165-3806(01)00243-7), which presumably occurs more frequently in the first days after the injury. Moreover, cell proliferation after HI is high 3-7 days after HI (Plane et al., Neurobiol of Disease, 2004 https://doi.org/10.1016/j.nbd.2004.04.003; Velthoven et al., Brain, Behavior, and Immunity 2010 https://doi.org/10.1016/j.bbi.2009.10.017). Thus, since BrdU injections started already on Day 4 after HI, the immunostaining would be able to capture the cell proliferations occurring already at that stage.

The rationale for performing RNAseq at day 3, was to try and capture the cellular pathways that were initiated at the start of the brain tissue regeneration process, in the sub-acute phase of HI.

6. There are multiple instances throughout the manuscript where data is interpreted without giving enough value to the related statistics, leading to misleading statements. Some examples are as follows, which need to be rectified:

Lines 122-123: The average size of HCAR1 KO spheres also tended to be smaller, although this was not statistically significant.

Lines 204-206: On the other hand, the contralateral hemisphere in HCAR1 KO mice was on average higher than in WT mice (albeit not statistically significant).

Lines 269-270: The average B1 levels were 78% lower in the HCAR1 KO compared with WT, but this result was not statistically significant (p=0.057)

The above mentioned statements have been removed.

Reviewer #3 (Recommendations for the authors):

This is a very interesting and important study testing the role of the lactate receptor HCAR1 in recovery after neonatal hypoxia-ischemia using a HCAR1 knockout. The authors have been thoughtful and thorough in their response to the previous reviews. They show significantly increased infarct size in the KO compared with wild-type animals, and then proceed to examine and demonstrate remarkable effects of the KO on processes likely to be involved in repair following injury: proliferation of neural progenitor cells, as well as astrocytes, oligodendrocytes, microglia; activation of microglia; and transcriptional activation in the subventricular zone. The authors are really to be commended for their wide-ranging analysis of the impact of HCAR1 KO. The study is largely descriptive, but this is not a liability as this is the foundation required for future studies. For example, given what they show, it would be of great interest to determine whether conditional knockout of HCAR1 in a particular cell type recapitulates the phenotype of the constitutive KO. One important issue is that the authors assume that deficits in repair are largely responsible for the differences in outcome between the WT and KO animals. However, in adult stroke, it is well-known that there is a penumbra of injured cells around the central infarcted area, and a great deal of work has been focused on the issue of saving the cells in the penumbra from being irreversibly committed to death. There is no mention of this issue in the manuscript, but it is an important problem, because it is conceivable that the effects of HCAR1 are entirely due to the protection of cells surviving the initial insult rather than the generation of new cells. At a minimum, the authors should address this issue in the discussion.

We thank the reviewer for the positive comments on our manuscript and for good suggestions for improvements. We have now included a paragraph about apoptosis in the discussion (starting at line 397).

Author details

1. Lauritz Kennedy

   1. Department of Microbiology, Oslo University Hospital and University of Oslo, Oslo, Norway
   2. Division of Physiology, Institute of Basic Medical Sciences, University of Oslo, Oslo, Norway

   Contribution

   Conceptualization, Data curation, Formal analysis, Investigation, Resources, Software, Validation, Visualization, Writing – original draft, Writing – review and editing

   Contributed equally with

   Emilie R Glesaaen

   Competing interests

   No competing interests declared

2. Emilie R Glesaaen

   1. Department of Microbiology, Oslo University Hospital and University of Oslo, Oslo, Norway
   2. Division of Physiology, Institute of Basic Medical Sciences, University of Oslo, Oslo, Norway

   Contribution

   Data curation, Formal analysis, Investigation, Validation, Visualization, Writing – original draft, Writing – review and editing

   Contributed equally with

   Lauritz Kennedy

   Competing interests

   No competing interests declared

3. Vuk Palibrk

   Department of Clinical and Molecular Medicine, Norwegian University of Science and Technology, Trondheim, Norway

   Contribution

   Formal analysis, Investigation, Visualization, Writing – review and editing

   Competing interests

   No competing interests declared

4. Marco Pannone

   1. Department of Microbiology, Oslo University Hospital and University of Oslo, Oslo, Norway
   2. Department of Clinical and Molecular Medicine, Norwegian University of Science and Technology, Trondheim, Norway

   Contribution

   Formal analysis, Investigation, Software, Visualization, Writing – review and editing

   Competing interests

   No competing interests declared

5. Wei Wang

   Department of Clinical and Molecular Medicine, Norwegian University of Science and Technology, Trondheim, Norway

   Contribution

   Formal analysis, Investigation, Validation, Visualization, Writing – review and editing

   Competing interests

   No competing interests declared

6. Ali Al-Jabri

   1. Department of Microbiology, Oslo University Hospital and University of Oslo, Oslo, Norway
   2. Division of Physiology, Institute of Basic Medical Sciences, University of Oslo, Oslo, Norway

   Contribution

   Formal analysis, Investigation, Writing – review and editing

   Competing interests

   No competing interests declared

7. Rajikala Suganthan

   Department of Microbiology, Oslo University Hospital and University of Oslo, Oslo, Norway

   Contribution

   Formal analysis, Investigation, Methodology, Writing – review and editing

   Competing interests

   No competing interests declared

8. Niklas Meyer

   1. Department of Microbiology, Oslo University Hospital and University of Oslo, Oslo, Norway
   2. Division of Physiology, Institute of Basic Medical Sciences, University of Oslo, Oslo, Norway

   Contribution

   Formal analysis, Investigation, Writing – review and editing

   Competing interests

   No competing interests declared

   "This ORCID iD identifies the author of this article:" 0000-0002-8377-097X

9. Marie Landa Austbø

   Department of Microbiology, Oslo University Hospital and University of Oslo, Oslo, Norway

   Contribution

   Formal analysis, Investigation, Writing – review and editing

   Competing interests

   No competing interests declared

10. Xiaolin Lin

    1. Department of Microbiology, Oslo University Hospital and University of Oslo, Oslo, Norway
    2. Department of Clinical and Molecular Medicine, Norwegian University of Science and Technology, Trondheim, Norway

    Contribution

    Formal analysis, Investigation, Supervision, Writing – review and editing

    Competing interests

    No competing interests declared

11. Linda H Bergersen

    1. The Brain and Muscle Energy Group, Institute of Oral Biology, Faculty of Dentistry, University of Oslo, Oslo, Norway
    2. Center for Healthy Aging, Department of Neuroscience and Pharmacology, Faculty of Health Sciences, University of Copenhagen, Copenhagen, Denmark

    Contribution

    Conceptualization, Resources, Validation, Writing – review and editing

    Competing interests

    LHB is a stock owner and head of the board at Lactate AS

12. Magnar Bjørås

    1. Department of Microbiology, Oslo University Hospital and University of Oslo, Oslo, Norway
    2. Department of Clinical and Molecular Medicine, Norwegian University of Science and Technology, Trondheim, Norway

    Contribution

    Conceptualization, Funding acquisition, Methodology, Supervision, Writing – original draft, Writing – review and editing

    Contributed equally with

    Johanne E Rinholm

    Competing interests

    No competing interests declared

13. Johanne E Rinholm

    1. Department of Microbiology, Oslo University Hospital and University of Oslo, Oslo, Norway
    2. Division of Physiology, Institute of Basic Medical Sciences, University of Oslo, Oslo, Norway

    Contribution

    Conceptualization, Funding acquisition, Investigation, Project administration, Supervision, Validation, Writing – original draft, Writing – review and editing

    Contributed equally with

    Magnar Bjørås

    For correspondence

    j.e.rinholm@medisin.uio.no

    Competing interests

    No competing interests declared

    "This ORCID iD identifies the author of this article:" 0000-0003-3741-850X

• 1,823

  Page views

• 321

  Downloads

• 12

  Citations

Article citation count generated by polling the highest count across the following sources: Crossref, PubMed Central, Scopus.