Apolipoprotein E regulates the maturation of injury-induced adult-born hippocampal neurons following traumatic brain injury.

Various brain injuries lead to the activation of adult neural stem/progenitor cells in the mammalian hippocampus. Subsequent injury-induced neurogenesis appears to be essential for at least some aspects of the innate recovery in cognitive function observed following traumatic brain injury (TBI). It has previously been established that Apolipoprotein E (ApoE) plays a regulatory role in adult hippocampal neurogenesis, which is of particular interest as the presence of the human ApoE isoform ApoE4 leads to significant risk for the development of late-onset Alzheimer's disease, where impaired neurogenesis has been linked with disease progression. Moreover, genetically modified mice lacking ApoE or expressing the ApoE4 human isoform have been shown to impair adult hippocampal neurogenesis under normal conditions. Here, we investigate how controlled cortical impact (CCI) injury affects dentate gyrus development using hippocampal stereotactic injections of GFP-expressing retroviruses in wild-type (WT), ApoE-deficient and humanized (ApoE3 and ApoE4) mice. Infected adult-born hippocampal neurons were morphologically analyzed once fully mature, revealing significant attenuation of dendritic complexity and spine density in mice lacking ApoE or expressing the human ApoE4 allele, which may help inform how ApoE influences neurological diseases where neurogenesis is defective.

Introduction

Among all trauma-related insults, traumatic brain injury (TBI) represents the highest contributor to death and disability globally, remaining a major public health concern worldwide [1, 2]. Individuals suffering from traumatic brain injuries can spontaneously recover to some degree, suggesting the existence of innate repair mechanisms. One such potential mechanism is hippocampal injury-induced neurogenesis [3]. Mouse modeling of TBI leads to neuronal cell death in the dentate gyrus of the hippocampus with loss of both mature and immature doublecortin-expressing neurons [4, 5]. It has been shown that injury-induced activation of resident neural stem/progenitor cells (NSPCs) from the subgranular zone of the dentate leads to neurogenesis that compensates for the loss of these vulnerable cell populations [5, 6]. These injury-induced neurons functionally integrate into the pre-existing circuitry [7] and appear to be necessary for at least some aspects cognitive recovery in mice as ablation of injury-induced hippocampal adult-born neurons impairs cognitive recovery after TBI in mice [3]. It has also been reported that these new neurons have morphologic aberrancies that may increase seizure susceptibility and contribute to other aspects of brain injury pathophysiology [8]. Thus, it remains unclear what the overall effect of injury-induced neurogenesis following TBI might be, though its relevance to the overall adaptation for the hippocampus to traumatic brain injury is clear [7, 9–13]

Alterations in adult hippocampal neurogenesis is a hallmark of hippocampus-associated neurological and neurodegenerative diseases, inducing cognitive deficits (e.g. difficulty in learning new information, memory loss) as observed in major depressive disorder, schizophrenia, Alzheimer’s disease, epilepsy, and TBI [14-19]. Understanding how adult neurogenesis is regulated is therefore essential before manipulation of dentate gyrus progenitor cells can be a viable therapeutic strategy in promoting neuronal regeneration. However, despite ever-expanding interest around the incorporation of adult-born granule cells to the hippocampal circuitry, the mechanisms directing this dentate gyrus neurogenic response remain poorly understood. One key regulatory gene for adult neurogenesis is Apolipoprotein E (ApoE), which has been shown to negatively regulate postnatal proliferation of dentate gyrus NSPCs [20, 21], while ablation of ApoE expression shifts NSPC differentiation towards astrogenesis instead of neurogenesis [22].

ApoE is primarily of astrocytic origin though it is also expressed by Type I neural stem cells before they develop into more mature neurons [17, 21]. ApoE is the main lipid carrier in the brain and regulates its transport and homeostasis, which in turn is critical for supporting neuronal development, beta-amyloid metabolism, and blood-brain barrier integrity maintenance [23-26]. In humans, ApoE is found in three commonly occurring protein alleles with ApoE3 considered the “wildtype” form as it is most predominant in humans and is Alzheimer’s disease risk-neutral [27]. ApoE2 is the rarest allele in the population and is thought to have neuroprotective effects [28], while ApoE4 occurs in approximately twenty percent of the population and is the greatest known genetic risk factor associated with the development of late-onset Alzheimer’s disease [29]. Moreover, ApoE4 has also been associated with poorer outcomes after TBI, including emergence of dementia and cognitive decline [30-33], and is also associated with small vessel disease and cognitive impairment on a vascular basis i.e. hypoxic-ischemic white matter damage and resulting dementia [34]. ApoE4 is also linked to a greater incidence of moderate or severe contusions [35] as well as concussions [36]. ApoE4 genotype combined with TBI is thought to increase the risk of developing Tauopathy and Alzheimer’s Disease [37] as well as post-traumatic epilepsy [38] and impaired spontaneous blood brain barrier repair [24]. Neurogenesis is affected in an ApoE isoform-dependent manner after both ischemic stroke [39] and controlled cortical impact [17] in mouse models but not after concussion [40]. Recently a large meta-analysis in humans concluded that outcome following TBI occurs in an ApoE isoform-dependent manner thus validating its importance in overall TBI, but the underlying mechanism remains entirely unknown [41].

ApoE regulates proliferation of NSPCs in both the healthy and injured mouse brain, and is influenced by human ApoE isoforms [17]. Moreover, ApoE deficiency and ApoE4 are both detrimental to dendritogenesis of adult-born granule neurons in the intact mouse brain [42], leading to significantly decreased dendritic complexity and spine density. In this study, we explored the effects of ApoE-deficiency and ApoE4 on the neuronal development of injury-induced adult-born cells.

Material and methods

Animals

All experimental procedures were in accordance with the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health and approved by the Institutional Animal Care and Use Committee at Columbia University (Protocol Number: AC-AAAT5462). Experimental animals were humanely housed and cared for under the supervision of the Institute of Comparative Medicine at Columbia University. All survival and non-survival surgeries were carried out under general isoflurane anesthesia, and animal suffering was minimized as further described. C57BL/6J (RRID:IMSR_JAX:000664; WT) and ApoE Knock-Out mice (B6.129P2-Apoetm1Unc/J; RRID:IMSR_JAX:002052) were both purchased from The Jackson Laboratory. ApoE3 (B6.129P2-Apoetm2(APOE*3)Mae N8; RRID:IMSR_TAC:1548) and ApoE4 (B6.129P2-Apoetm3(APOE*4)Mae N8; RRID:IMSR_TAC:1549) humanized mice were purchased from Taconic.

Controlled cortical impact injury and retroviral injections

Sixteen mixed-sex WT, ApoE-deficient, ApoE3, and ApoE4 mice (2 males and 2 females in each condition) underwent CCI at six-weeks of age as previously described [3]. Twenty minutes following intraperitoneal injection of analgesics (5 mg/kg Carprofen–Rimadyl), mice were placed in an induction chamber and subjected to general anesthesia using 4% isoflurane supplemented with 1L flow of oxygen. Head fur was clipped and animal heads were fixed on a stereotactic frame where anesthesia was maintained throughout surgery using a nose cone linked to the Isoflurane Evaporator (Summit Anesthesia Solutions). The isoflurane dose was progressively reduced from 4% to 2%, and the level of anesthesia was assessed using toe pinch while internal temperature was kept at 37°C via a heating pad (Adroit, Loudon, TN). After the scalp was disinfected using three separate antiseptic swabs (B10800 –Prevantics), local analgesics (2mg/kg Bupivicaine–Hospira) were subcutaneously injected and ophthalmic ointment was applied (one drop/eye). Following a scalp midline incision, the soft tissues were reflected and a 5 mm by 5 mm craniotomy was performed between bregma and lambda (to the left of the sagittal suture) to expose the underlying dura. Single moderate cortical contusion injuries (3 mm stainless steel impounder tip; speed: 4.4 m/sec; deformation: 0.7 mm; dwell time: 0.3 sec) were delivered with the Leica One Stereotaxic Impactor device (Leica, Houston, TX). Immediately after injury, a retrovirus carrying enhanced green fluorescent protein (eGFP) was stereotactically infused into the mouse dentate gyrus (Moloney Murine Leukemia Viral vectors were generated by the GT3 Core Facility of the Salk Institute). Retroviruses only infect mitotic (actively dividing) cells in a stochastic manner. As a result, only neural stem and progenitor cells from the neuronal lineage express GFP in the infected dentate gyrus. Other potentially dividing cells such as reactive astrocytes and microglia can morphologically be differentiated without ambiguity. The constitutive expression of eGFP in infected NSPCs allowed for permanent labelling and further dendritic arborization tracing. As previously described [42], to capture NSPCs from both the dorsal and ventral parts of the dentate gyrus and from both hemispheres, one microliter of packaged retrovirus [1*109 transducing units (TU)/mL)] was injected at the rate of 0.1μl per minute as: Antero/Posterior = -2.0mm & -2.5mm, Medio/Lateral = ±1.55mm & ±2mm, Dorso/Ventral = -2.0mm & -2.25mm all coordinates from bregma, using a micro infusion pump (KD scientific) linked to a 10μl Hamilton syringe (Model #801), for a total of 4 injected sites per animal. After surgery, the scalp was closed with sutures, topical antibiotic ointment was applied (one drop of Neosporin) and animals were placed in their cages and allowed to recover from anesthesia.

Tissue processing and immunohistochemistry

Since it takes approximately 4 weeks for adult born granule cells to mature and integrate into the trisynaptic circuitry [43], we chose this post-surgery time point to perform our histological analysis. Transcardiac perfusion was performed with 150 mL of 4% paraformaldehyde (PFA) in 1x PBS delivered via a peristaltic pump at a rate of 10ml/min (MasterFlex L/S, Cole Parmer); animals were subjected to general anesthesia during perfusions (Isothesia Isoflurane, Henry Schein Animal Health). Post-fixation was achieved by incubating whole brains overnight in in 4% PFA/1x PBS before sectioning of serial 50μm coronal brain slices (vibratome VT1000S, Leica). Samples that encompassed the hippocampus were all sequentially collected in six Eppendorf tubes with either 1x PBS or antifreeze solution (30% Glycerol/30% ethylene glycol in 1x PBS) and free-floating sections were used for further immunohistochemistry while the remaining samples were preserved in the dark at -20°C. Samples were first rinsed with 1x PBS to eliminate antifreeze solution when applicable (3x5min). Then a standard immunostaining procedure was performed at room temperature (RT): permeabilization step [0.3% Triton X-100 in 1x PBS (PBST; 3x10mn)] followed by blocking step [PBST with 5% Normal Donkey Serum (NDS, Jackson ImmunoResearch Labs, Cat# 017-000-001; 1 hour)]. Brain sections were then incubated overnight at 4°C with primary antibodies (1:5000 Goat polyclonal α-ApoE, EMD Millipore, AB947; 1:1000 Rabbit polyclonal α-GFP, Invitrogen, Cat# A-11122; 1:1000 Polyclonal Guinea Pig α-GFAP, Mybiosource, MBS834682) in PBST/5% NDS. The next day, tissues were washed with PBST (3x5mn) and incubated with biotinylated or alexa-conjugated secondary antibodies (1:200 Biotin Donkey α-Goat, Jackson ImmunoResearch Labs, Cat# 705-006-147; 1:200 Biotin Donkey α-Rabbit, Jackson ImmunoResearch Labs, Cat# 711-545-152; 1:200 Biotin Donkey α-Guinea Pig, Jackson ImmunoResearch Labs, Cat# 706-605-148) for 3 hours at RT. Sections were then washed with 1x PBS (3x5mn) and incubated with alexa594-conjugated Streptavidin antibodies (1:200, Jackson ImmunoResearch Labs, Cat# 016-580-084) for 2 hours at RT. Finally, tissue samples were rinsed with 1x PBS (3x5min) before they were slide-mounted and coverslip-sealed (H-1500, Vector Laboratories).

Neuronal morphological analysis

As previously described [42], upon completion of immunostaining, slides were visualized and imaged with a Zeiss microscope (Axio Imager M2, Zeiss) equipped with a Hamamatsu camera (Orca-R2, Hamamatsu). Interval z-stack images at 1μm were obtained under a 20x objective using an optical fractionator (Zeiss Apotome.2) and stereological image analysis software (MBF Bioscience, RRID:SCR_002526). Acquired z-stacks were then opened with the Neurolucida360 software (RRID:SCR_001775) where 3D neural tracing was performed by a blinded experimenter. Neuronal traces were analyzed using the same software with several morphological parameters of the dendritic arborization chosen for further comparisons (proximal dendritic length before the first division, total cumulated dendritic length, dendritic complexity, dendritic spawn and Sholl analysis) while axons were not studied.

For each animal, a set of brain samples enclosing every sixth section was mounted on slides and endogenous eGFP-expressing dendritic spines were visualized using a Laser Scan confocal microscope (TCS SP8, Leica). Z-stack images at 0.1μm intervals were acquired in the molecular layer of the dentate gyrus under a 63x oil objective with a five-time digital zoom (pixel size = 57.21nm, NA = 1.44, resolution = 512x512, frame average = 4) and were then deconvolved using Autoquant software (RRID:SCR_002465). Z-stacks of dendritic fragments were then visualized in three dimensions using Neurolucida 360 (RRID:SCR_001775) [44], 10μm dendritic fragments were randomly selected for spine counting (no more than one fragment per dendrite) and analyzed by a blinded experimenter. To avoid underestimation of the spine density that could be due to spherical aberrations over the Z axis, spine counts were expressed relative to fragment lengths [45].

Statistical analysis

Graphpad Prism (RRID:SCR_015807) was used to perform all statistical analyses. The Shapiro-Wilk was used to assess the normality of data. All results are shown as the mean ± Standard Error of the Mean (SEM) and statistical details are presented in Table 1. Because CCI induces outward migration of adult-born granule neurons [7], we further divided reconstructed neurons from the ipsilateral side into the inner granule cell layer (iGCL, soma located in the inner one third of the granule cell layer) and the outer granule cell layer (oGCL, soma located in the outer two thirds of the granule cell layer), however, this analysis was performed on the ipsilateral side only because outward migration was rarely observed on the contralateral side. Our group previously published the dendritic morphology of adult-born neurons in the uninjured dentate gyrus of WT and various ApoE conditions [42]. We compared these to the contralateral neurons in injured animals analyzed in the present manuscript and found no significant differences. Thus, neurons residing in the contralateral side of injured brains were used as controls in the current study. A Kruskal-Wallis test was used to study non-parametric distributions followed by Dunn’s post hoc tests and one-way ANOVA followed by Tukey’s honestly significant difference (HSD) post hoc test was used for parametric distributions. Two-way ANOVA was used to study the effect of different conditions on the Sholl Analysis followed by an uncorrected least significant difference (LSD) Fisher’s post hoc tests where the Type I error increases with the number of multiple comparisons [46]. Statistical analyses for each individual experiment are summarized in Table 1.

Table 1

Summary of statistics.

Table Analyzed	Condition	Statistical Test	p value	DF	R2	Statistic
1st Branch	WT	Kruskal-Wallis	<0.0001			H (3) = 18.44
	WT: iGCL vs oGCL	Dunn's test	<0.0001
	WT: iGCL vs Contralateral	Dunn's test	0.5977
	WT: oGCL vs Contralateral	Dunn's test	0.0073
	WT vs ApoE KO	Kruskal-Wallis	<0.0001			H (6) = 43.62
	WT vs ApoE KO: iGCL	Dunn's test	0.5543
	WT vs ApoE KO: oGCL	Dunn's test	0.9172
	WT vs ApoE KO: Contralateral	Dunn's test	0.2871
	ApoE3 vs ApoE4	Kruskal-Wallis	<0.0001			H (6) = 46.92
	ApoE3 vs ApoE4: iGCL	Dunn's test	0.0061
	ApoE3 vs ApoE4: oGCL	Dunn's test	0.2892
	ApoE3 vs ApoE4: Contralateral	Dunn's test	0.0227
Nodes	WT	Kruskal-Wallis	0.5029			H (3) = 1.375
	WT: iGCL vs oGCL	Dunn's test	0.8527
	WT: iGCL vs Contralateral	Dunn's test	>0.9999
	WT: oGCL vs Contralateral	Dunn's test	0.9441
	WT vs ApoE KO	Kruskal-Wallis	<0.0001			H (6) = 32.02
	WT vs ApoE KO: iGCL	Dunn's test	0.0003
	WT vs ApoE KO: oGCL	Dunn's test	0.006
	WT vs ApoE KO: Contralateral	Dunn's test	0.0062
	ApoE3 vs ApoE4	Kruskal-Wallis	<0.0001			H (6) = 46.08
	ApoE3 vs ApoE4: iGCL	Dunn's test	0.0001
	ApoE3 vs ApoE4: oGCL	Dunn's test	0.0001
	ApoE3 vs ApoE4: Contralateral	Dunn's test	0.0003
Total Length	WT	Kruskal-Wallis	0.1369			H (3) = 0.1369
	WT: iGCL vs oGCL	Dunn's test	0.187
	WT: iGCL vs Contralateral	Dunn's test	>0.9999
	WT: oGCL vs Contralateral	Dunn's test	0.2969
	WT vs ApoE KO	Kruskal-Wallis	<0.0001			H (6) = 40.64
	WT vs ApoE KO: iGCL	Dunn's test	0.0398
	WT vs ApoE KO: oGCL	Dunn's test	0.002
	WT vs ApoE KO: Contralateral	Dunn's test	0.0008
	ApoE3 vs ApoE4	Kruskal-Wallis	<0.0001	5	0.0862	F (5, 474) = 8.945
	ApoE3 vs ApoE4: iGCL	Tukey's HSD	<0.0001
	ApoE3 vs ApoE4: oGCL	Tukey's HSD	<0.0001
	ApoE3 vs ApoE4: Contralateral	Tukey's HSD	0.4031
Angle	WT	Kruskal-Wallis	0.0157			H (3) = 0.0157
	WT: iGCL vs oGCL	Dunn's test	0.0134
	WT: iGCL vs Contralateral	Dunn's test	0.2533
	WT: oGCL vs Contralateral	Dunn's test	0.6427
	WT vs ApoE KO	Kruskal-Wallis	0.0005			H (6) = 22.25
	WT vs ApoE KO: iGCL	Dunn's test	0.2197
	WT vs ApoE KO: oGCL	Dunn's test	0.0927
	WT vs ApoE KO: Contralateral	Dunn's test	0.0154
	ApoE3 vs ApoE4	Kruskal-Wallis	<0.0001			H (6) = 48.8
	ApoE3 vs ApoE4: iGCL	Dunn's test	0.0006
	ApoE3 vs ApoE4: oGCL	Dunn's test	0.0907
	ApoE3 vs ApoE4: Contralateral	Dunn's test	0.0004
Spine density	WT vs ApoE KO	One-way ANOVA	<0.0001	3	0.4861	F (3, 258) = 81.33
	Ipsilateral: WT vs ApoE KO	Tukey's HSD	<0.0001
	Contralateral: WT vs ApoE KO	Tukey's HSD	<0.0001
	WT Ipsilateral vs Contralateral	Tukey's HSD	0.5789
	ApoE KO Ipsilateral vs Contralateral	Tukey's HSD	0.6437
	ApoE3 vs ApoE4	Kruskal-Wallis	<0.0001			H (4) = 196.4
	Ipsilateral: ApoE3 vs ApoE4	Dunn's test	<0.0001
	Contralateral: ApoE3 vs ApoE4	Dunn's test	<0.0001
	ApoE3 Ipsilateral vs Contralateral	Dunn's test	>0.9999
	ApoE4 Ipsilateral vs Contralateral	Dunn's test	>0.9999
Sholl Analysis	WT: iGCL vs oGCL vs Contralateral	Two-way ANOVA
		Interaction	<0.0001	56		F (56, 4727) = 2.651
		Row Factor	<0.0001	28		F (28, 4727) = 220.5
		Column Factor	<0.0001	2		F (2, 4727) = 12.95
	WT: iGCL vs Contralateral	Fisher's LSD	p < 0.05, 0.05, 0.01 for each 10μm increment between 30 and 50μm from the soma, p < 0.05 at 110μm from the soma, p < 0.05, 0.01, 0.001 for each 10μm increment between 130 and 150μm from the soma
	WT: oGCL vs Contralateral	Fisher's LSD	p < 0.05, 0.05 at 110 and 120μm from the soma, p < 0.05, 0.05 at 190 and 200μm from the soma
	WT: iGCL vs oGCL	Fisher's LSD	p < 0.01, 0.01, 0. 01, 0.05 for each 10μm increment between 30μm and 60μm from the soma, p < 0.01, 0.001, 0.001, 0.001, 0.01, 0.001, 0.01, 0.001, 0.01, 0.01 for each 10μm increment between 100 and 190μm from the soma
	WT vs ApoE KO: iGCL	Two-way ANOVA
		Interaction	<0.0001	28		F (28, 3944) = 4.607
		Row Factor	<0.0001	28		F (28, 3944) = 169.7
		Column Factor	<0.0001	1		F (1, 3944) = 62.34
		Fisher's LSD	p < 0.0001; p < 0.05, 0.001, 0.01, 0.001, 0.001, 0.001, 0.001, 0.001, 0.001, 0.001 and 0.05 for each 10μm increment between 70μm and 170μm from the soma
	WT vs ApoE KO: oGCL	Two-way ANOVA
		Interaction	0.2707	28		F (28, 3335) = 1.147
		Row Factor	<0.0001	28		F (28, 3335) = 122.3
		Column Factor	<0.0001	1		F (1, 3335) = 71.14
		Fisher's LSD	p < 0.01, 0.01, 0.001, 0.05, 0.01, 0.01, 0.05, 0.05, 0.05, non-significant, 0.05, 0.05, 0.05 for each 10μm increment between 80μm and 200μm from the soma
	WT vs ApoE KO: Contralateral	Two-way ANOVA
		Interaction	<0.0001	28		F (28, 4263) = 3.014
		Row Factor	<0.0001	28		F (28, 4263) = 203.8
		Column Factor	<0.0001	1		F (1, 4263) = 74.25
		Fisher's LSD	p < 0.001; p < 0.05, non-significant, 0.05, 0.05, 0.05, 0.01, 0.001, 0.001, 0.001, 0.001, 0.001, 0.001 and 0.01 for each 10μm increment between 40μm and 160μm from the soma
	ApoE3 vs ApoE4: iGCL	Two-way ANOVA
		Interaction	<0.0001	27		F (27, 5068) = 6.679
		Row Factor	<0.0001	27		F (27, 5068) = 208.4
		Column Factor	<0.0001	1		F (1, 5068) = 143.9
		Fisher's LSD	p < 0.01, 0.001, 0.001, 0.001, 0.001, 0.001, 0.001, 0.001, 0.001, 0.001, 0.001, 0.001, 0.001 and 0.05 for each 10μm increment between 50μm and 180μm from the soma
	ApoE3 vs ApoE4: oGCL	Two-way ANOVA
		Interaction	<0.0001	27		F (27, 3500) = 3.279
		Row Factor	<0.0001	27		F (27, 3500) = 162.6
		Column Factor	<0.0001	1		F (1, 3500) = 151.8
		Fisher's LSD	p < 0.05, 0.001, 0.001, 0.001, 0.001, 0.001, 0.01, 0.01, 0.001, 0.001, 0.001, 0.001, 0.001, 0.01 and 0.001 for each 10μm increment between 50μm and 190μm from the soma
	ApoE3 vs ApoE4: Contralateral	Two-way ANOVA
		Interaction	<0.0001	27		F (27, 4704) = 4.133
		Row Factor	<0.0001	27		F (27, 4704) = 217.8
		Column Factor	0.0419	1		F (1, 4704) = 4.143
		Fisher's LSD	p < 0.01, 0.001, 0.001 and 0 0.001 for each 10μm increment between 40μm and 70μm from the soma, p < 0.05 at 100μm from the soma, p < 0.01, 0.01, 0.01 and 0.01 for each 10μm increment between 180μm and 210μm from the soma

DF = Degrees of Freedom; iGCL = Inner Granule Cell Layer, oGCL = Outer Granule Cell Layer, in the ipsilateral dentate gyrus; WT = Wild Type. KO = Knockout. The Sholl Analysis represents the average number of intersections in each condition (Column factor) function of the distance to the soma (Row factor).

Results

GFP-expressing retroviruses track adult-born neuronal dendrite development which are found in close proximity to ApoE-expressing astrocytes

During normal adult hippocampal neurogenesis, NSPCs migrate short distances from the subgranular zone to the inner one-third of the granule cell layer [7, 10, 47]. Typically, adult-generated dentate gyrus neurons migrate to the inner one-third of the granule cell layer (iGCL) and project their dendrites toward the GCL and the Molecular layer, with the first dendritic division occurring around 35μm from the cell body [42]. However, this migratory pattern is not observed after experimental brain injury as CCI causes aberrant migration of newborn immature neurons in the hippocampus with proliferation peaking in the first week and eventually leading to around 50% misplacement of mature injury-induced adult-born granule neurons, many of which become permanently incorporated into the existing dentate gyrus [7, 10, 11]. Because the functional implications of this aberrancy remain unclear, we separately analyzed neurons born in the inner (iGCL) and outer GCL (oGCL) in the ipsilateral injury side, whereas on the contralateral side, we only very rarely observed misplaced adult-born neurons.

Mixed genotype and sex mice underwent moderate CCI at 6 weeks of age immediately followed by stereotactic brain injections of a GFP-expressing retrovirus. Mice were perfused 4 weeks later and brains were removed to assess both quality of the injury and retroviral labelling of mature granule cells before further histological analysis was performed (Fig 1A). Next, sections were stained for ApoE and the astrocyte-specific marker GFAP to determine its expression in relationship to GFP-expressing dendrites where astrocytic expression of ApoE is seen in close proximity with GFP-expressing mature granule cell dendrites by (Fig 1B–1F). We next confirmed an absence of ApoE staining in the ApoE null animal (Fig 1G–1K). Finally, we confirmed ApoE astrocytic staining in the targeted replacement ApoE3 (Fig 1L–1P) and ApoE4 (Fig 1Q–1U) mouse dentate gyrus, indicating close physical interaction of astrocytes with maturing injury-induced neurons (Fig 1P–1U, white arrowheads).

Fig 1

ApoE-expressing astrocytes physically interact with injury-induced GFP-expressing dendrites.

(A) Representative section of the contralateral and ipsilateral cortex and hippocampus 4 weeks after Controlled Cortical Impact and stereotaxic injection of GFP-expressing retroviruses in the dentate gyrus of a wildtype mouse reveals the typical morphology of the injured hippocampus along with efficient long-term labelling of neural progenitor cells. Scale bar 500μm. (B-U) Representative pictures of ApoE (first column, antigen is recombinant human ApoE and can react with human e2, e3 and e4, as well as non-human primate and mouse ApoE), GFP (second column) and GFAP (third column) immunostaining in the dentate gyrus of WT (B-F), ApoE KO (G-K), ApoE3 (L-P) and ApoE4 mice (Q-U). (E, J, O, T) Merged pictures (scale bar 100μm), white squares indicate enlarged area shown in last column (F, K, P, U), white arrowheads designate ApoE-expressing GFAP-positive astrocytes wrapping around GFP-expressing dendrites, red arrowhead designate ApoE-deficient GFAP-positive astrocyte, scale bar 10μm.

Injury-induced adult-born neurons lacking ApoE show impaired dendritic development after CCI

Since it is known that ApoE deficiency leads to less sophisticated granule neurons during postnatal neurogenesis [42] and that CCI induces misplacement of adult-born granule neurons in the dentate gyrus, we investigated the effect of ApoE deficiency on the dendritic maturation of CCI-induced newborn neurons.

We observed no differences in the distance to the first branch when comparing WT to ApoE-deficient injury-induced neurons whether they were in the inner or outer portions of the ipsilateral dentate gyrus or on the contralateral side (Fig 2A). However, when we compared the number of nodes and the total dendritic length, we observed highly significant attenuation in the ApoE-deficient dentate gyrus (Fig 2B–2D). We also observed significantly narrower dendritic trees in the ApoE-deficient neurons found in the contralateral dentate gyrus, when compared to WT matching conditions (Fig 2E), representative traces are shown in Fig 2C.

Fig 2

ApoE deficiency leads to impaired dendritic development of injury-induced adult-born granule neurons.

(A) ApoE deficiency did not affect the distance to the first dendritic branching but did significantly affect dendritic complexity (B) in the iGCL (p < 0.001), oGCL (p < 0.01) and contralateral (p < 0.01) dentate gyrus when compared to WT. (C) Representative tracings of GFP-expressing cells found in the iGCL, oGCL and contralateral dentate gyrus from WT and ApoE-deficient injured dentate gyrus. (D) The cumulative dendritic length is significantly attenuated in mouse lacking ApoE when compared to matching Wild Type cells (iGCL: p < 0.05, oGCL: p < 0.01, Contra: p < 0.001). (E) Injury-induced cells found in the contralateral dentate gyrus in the absence of ApoE also showed decreased dendritic span when compared to WT (p < 0.05). Sholl analysis of dendritic arborizations of CCI-induced adult-born granule neurons exposed differences in the dendritic branching number observed in both proximal and distal regions when comparing WT with ApoE-deficient GFP-expressing cells found in the iGCL (F), in the oGCL (G) and in the contralateral dentate gyrus (H). (I-K) In each condition, all traced neurons have been categorized depending on their complexity (≤4…≥11 nodes), highlighting higher proportions of less complex neurons (<4 nodes) in ApoE-deficient, particularly in the oGCL. 4 mice/condition and at least 10 neurons/mouse were analyzed; WT iGCL: 61 cells; WT oGCL: 47 cells; WT Contra: 60 cells; ApoE KO iGCL: 77 cells; ApoE KO oGCL: 73 cells; ApoE KO Contra: 89 cells. Contra = Contralateral dentate gyrus; iGCL = Inner Granule Cell Layer, oGCL = Outer Granule Cell Layer, in the ipsilateral dentate gyrus; WT = Wild Type, KO = ApoE Knock-out. ***p < 0.001. (L) ApoE deficiency leads to significantly reduced spine density in injury-induced adult-born granule neurons when compared to WT cells from both the Ipsilateral and Contralateral sides (p < 0.001). High power representative pictures of dendritic fragments from mature adult-born granule neurons in wildtype and ApoE-deficient (M). 4 mice/condition; Number of dendritic fragments analyzed: WT Ipsilateral = 60, WT Contralateral = 62, ApoE KO Ipsilateral = 87, ApoE KO Contralateral = 53. Scale bar = 5μm.

We next studied the distribution of the dendritic branches with a Sholl Analysis. A two-way ANOVA was conducted to quantify the effect of the different conditions on the dendritic complexity relative to the distance to the soma. We found a significant interaction in the iGCL and significant column (number of intersections) and row (distance to the soma) effects in both the iGCL and oGCL. Fisher’s LSD post-hoc test exposed an attenuation of branching number in ApoE KO adult-born granule neurons found in the iGCL when compared to WT (Fig 2F), as well as in the oGCL (Fig 2G) and in the contralateral GCL (Fig 2H).

As previously described [42], imaging of adult-born granule cells revealed that their distribution and number of dendritic branches were not uniform independent of genotype or soma localization. To quantify this observation, for each condition, we categorized neuronal traces as a function of their dendritic complexity 4 weeks following GFP-expressing retroviral infection (Fig 2I–2K). Consistent with previous findings [42], we observed that adult-born granule cells lacking ApoE demonstrated a greater number of cells with up to 4 nodes, particularly in the oGCL (23.3%) when compared with WT (6.4%; Fig 2J), while we detected that more than 40% of WT injury-induced newborn neurons dendritic arborizations had 8 nodes or more compared to ~23% for ApoE-deficient mice in the iGCL (Fig 2I) and contralateral side (Fig 2J), and ~25% in the oGCL (Fig 2K). This result highlights that the attenuation in dendritic complexity in ApoE-deficient adult-born neurons following CCI is due to the presence of a higher proportion of less complex neurons, particularly in the oGCL. Finally, we examined the spine density of dentate gyrus adult-born neurons lacking ApoE, by reconstructing randomly selected 10μm dendritic fragments from GFP-expressing cells. The reduced spine density observed in ApoE-deficient adult-born granule neurons [42] is also present after CCI (Fig 2L and 2M).

Together, these results demonstrate morphological dendritic aberrancies of injury-induced neurons in the ApoE deficient mouse dentate gyrus compared to WT mice in all three studied areas.

ApoE4 leads to less complex adult-born granule neurons after CCI when compared to ApoE3

We next compared the development of adult-born granule neurons in human ApoE4 and ApoE3 mice following controlled cortical impact injury. We examined the first branch length and found a significant difference based on genotype and further analysis revealed differences for neurons located in the iGCL and the contralateral side but not the oGCL (Fig 3A). We then compared the complexity of injury-induced neurons in the different conditions as a function of genotype. This analysis revealed a significant effect on complexity and post-hoc tests indicated that in the iGCL, oGCL, and contralateral side to the injury, ApoE4 adult-born neurons have fewer nodes when compared to ApoE3 (Fig 3B), as well as having reduced cumulative dendritic length in both the iGCL and oGCL (Fig 3D). Representative traces are shown in Fig 3C.

Fig 3

Impaired dendritic development of adult-born granule neurons after CCI in ApoE4 mouse.

(A) GFP-cells found in the iGCL (p < 0.01) and contralateral dentate gyrus (p < 0.05) of ApoE4 mice have the first dendritic branching occurring further from their cell body when compared to ApoE3. (B) The number of dendritic intersections of adult-born granule cells located in the iGCL, oGCL and contralateral dentate gyrus of ApoE4 mice is significantly reduced when compared to ApoE3 (p < 0.001). (C) Representative tracings of GFP-expressing cells found in the iGCL, oGCL and contralateral dentate gyrus from ApoE3 and ApoE4 adult mouse injured dentate gyrus. (D) The cumulative length of ApoE4-expressing injury-induced neurons’ dendrites is significantly reduced in both the iGCL and oGCL (p < 0.001). (E) ApoE4 significantly impairs the dendritic span of adult-born granule cells from both the iGCL and the contralateral dentate gyrus (p < 0.001). Sholl analysis revealed significant differences in the number of dendritic intersections observed in both proximal and distal regions when comparing ApoE3 with ApoE4-expressing cells found in the iGCL (F), in the oGCL (G) and in the contralateral dentate gyrus (H). 4 mice/condition and at least 10 neurons/mouse were analyzed; ApoE3 iGCL: 81 cells; ApoE3 oGCL: 62 cells; ApoE3 Contra: 81 cells; ApoE4 iGCL: 102 cells; ApoE4 oGCL: 65 cells; ApoE4 Contra: 89 cells. Contra = Contralateral dentate gyrus; iGCL = Inner Granule Cell Layer, oGCL = Outer Granule Cell Layer, in the ipsilateral dentate gyrus. *p < 0.05, **p < 0.01, ***p < 0.001. (I-K) In each condition, all traced neurons have been categorized depending on their complexity (≤4…≥11 nodes), highlighting higher proportions of less complex neurons (<4 nodes) in ApoE4-expressing mice, particularly in the oGCL. 4 mice/condition and at least 10 neurons/mouse were analyzed; ApoE3 iGCL: 81 cells; ApoE3 oGCL: 62 cells; ApoE3 Contra: 84 cells; ApoE4 iGCL: 102 cells; ApoE4 oGCL: 65 cells; ApoE4 Contra: 82 cells. Contra = Contralateral dentate gyrus; iGCL = Inner Granule Cell Layer, oGCL = Outer Granule Cell Layer, in the ipsilateral dentate gyrus; E3 = ApoE3, E4 = ApoE4. ***p < 0.001. (L) ApoE4-expressing adult-born granule neurons show significantly attenuated spine density when compared to ApoE3 matching cells from both dentate gyrus (p < 0.001). High power representative pictures of dendritic fragments from mature adult-born granule neurons in ApoE3 and ApoE4 (M) ipsilateral and contralateral dentate gyrus. 4 mice/condition; Number of dendritic fragments analyzed: ApoE3 Ipsilateral = 99, ApoE3 Contralateral = 91, ApoE4 Ipsilateral = 71, ApoE4 Contralateral = 58. Scale bar = 5μm.

We next examined the dendritic span of adult-born GC dendritic arborizations by measuring the angle formed by the two most distal branches when projected onto two dimensions. CCI had a significant effect on ApoE4 dendritic span as shown by Kruskal-Wallis test and Dunn’s post-hoc tests for cells located in the iGCL and contralateral dentate gyrus but not in the oGCL (Fig 3E). Finally, we studied the distribution of the dendritic branches with a Sholl Analysis. A two-way ANOVA was conducted that examined the effect of the different conditions on the number of dendritic intersections relative to the distance from the soma. We found a significant interaction of the parameters in the Inner GCL (Fig 3F), as well as in the oGCL (Fig 3G) and in the contralateral side GCL (Fig 3H).

We then categorized neuronal traces as a function of their dendritic complexity and demonstrated that 4 weeks following eGFP-infusions. Consistent with previous findings [42], we observed that adult-born granule cells expressing the E4 human isoform demonstrated a greater number of cells with up to 4 nodes, particularly in the oGCL (24.6%) compared with ApoE3 (6.5%; Fig 3J), while we detected that ~50% of ApoE3 injury-induced newborn neurons dendritic arborizations had 8 nodes or more compared to ~30% for ApoE4 mice in the iGCL (Fig 3I) and contralateral side (Fig 3J), and ~25% in the oGCL (Fig 3K). This result highlights that similarly to what we observed with ApoE deficient mice, the attenuation in dendritic complexity in ApoE4 adult-born neurons following CCI is due to the presence of a higher proportion of less complex neurons, particularly in the oGCL. Finally we studied the spine density of dentate gyrus adult-born neurons expressing the ApoE4 human isoform by reconstructing randomly selected 10μm dendritic fragments from GFP-expressing cells. The reduced spine density observed in ApoE4 adult-born granule neurons (29) is also present after CCI (Fig 3L and 3M).

Together, these results demonstrate morphological dendritic aberrancies of injury-induced neurons in the ApoE4 mouse dentate gyrus compared to ApoE3 mice in all three studied areas.

Finally, we describe the intra-condition comparisons of dendritic morphology to highlight the changes driven by the injury in each genotype. Both the dendritic complexity and the spine density of injury-induced adult-born granule neurons were similar in the 3 studied areas (iGCL/oGCL and Contralateral hemispheres) for all 4 genotypes, revealing no injury-driven effect for these parameters. Thus, the differences observed in dendritic morphology and spine density when comparing WT with KO or E3 with E4 were strictly driven by genotype (S1 and S2 Figs).

Discussion

Previous studies revealed that ApoE deficiency in mice leads to a depletion of the neural stem and progenitor pool over time with a subsequent decrease in the activation of Type I cells. This observation is accompanied by further impaired dendritic development in mice lacking ApoE or expressing the human isoform ApoE4 [42]. It is also known that ApoE is necessary for normal injury-induced neurogenesis following experimental moderate TBI, and is associated with ApoE human isoform-dependent effects [17]. In the present study, we establish that ApoE deficiency and human ApoE genotype both influence the dendritic development of newborn granule neurons in the injured adult mouse hippocampus, which includes impaired dendritic arborization and diminished spine density in mice either lacking ApoE or expressing human ApoE4. We also observe that CCI induces outward migration of granule neurons into the oGCL, which is why we further divided adult-born granule neurons from the ipsilateral side into the inner one-third or outer two-thirds of the GCL. Importantly, and contrary to what has been previously shown [7], CCI did not affect the complexity, total dendritic length, or spine density of mature adult-born granule neurons, 4 weeks after brain injury (when compared to contralateral matching neurons). These findings are consistent with a recent study in which experimental ischemic stroke was induced in adult mice [48]. Finally, by grouping adult-born granule neurons in the different conditions as a function of their complexity, we further highlighted that the observed attenuation in dendritic branching of injury-induced granule cells lacking ApoE or expressing the ApoE4 human isoform was the result of a higher proportion of less complex neurons when compared to WT and ApoE3.

TBI triggers selective secondary cell death in the dentate gyrus and immature granule neurons localized in the inner GCL constitute the most vulnerable population while both nestin-expressing NSPCs and NeuN-expressing mature granule cells have been shown to be less vulnerable [4, 5]. TBI commonly causes cognitive functional deficits though some degree of spontaneous recovery occurs, which appears to be mediated at least in part by injury-induced hippocampal neurogenesis [3]. It has also been shown that injury-induced adult-born granule neurons mature more rapidly than those observed in usual adult neurogenesis [7, 48]. These observations suggest that injury-induced neurogenesis may in part compensate for the loss of young adult-born granule neurons by generating new adult-born granule neurons and accelerating their dendritic development.

A typical hallmark of TBI is the outward migration of injury-induced neurons in the ipsilateral dentate gyrus to the outer two-thirds of the granule cell layer at one week and one month after CCI [10]. Here, we observe that, after CCI, most new neurons were found in the oGCL, where their first dendritic branching was found closer to the cell body than what is observed in the iGCL. In addition, these newborn neurons demonstrated a greater lateral spread of their dendrites, suggesting that their morphology is not aberrant, but adaptive. This is supported by the fact that new neurons found in the oGCL are not less sophisticated, though they do show some degree of dendritic reorganization. Moreover, injury-induced neurons do not develop aberrant electrophysiological properties as they demonstrate preserved early electrophysiological maturation and functionally integrate to the pre-existing try-synaptic circuitry four weeks following CCI [7], supporting the idea that they can function normally.

The key regulators of neuronal migration (i.e. onset, speed, directionality, and arrest) remain poorly understood. It has been shown that experimental stroke [49], neonatal hypoxia [50], entorhinal cortical lesions [51], and seizures [8, 15, 52, 53] all induce ectopic migration of newborn granule cells and/or dendritic arborization aberrancies. Several mechanisms have been implicated in the outward migration of injury-induced adult-born granule neurons including cell-intrinsic impairments, regionally localized abnormalities, granule cell layer dispersion, and somatic translocation [8]. It is unclear why approximately half of the adult-born granule neurons migrate to the oGCL after CCI, while the NSPCs may be functionally normal, the injury-induced environment may disrupt migratory cues, which would lead to ectopic migration. This is supported by recent work ex vivo which suggests that the regionalized presence of the extracellular matrix protein reelin in the molecular layer controls the directionality of granule cell migration, but not the actual migratory process or speed [54].

ApoE genotype influences dendritogenesis

We demonstrate here that the previously observed morphologic impairments of adult-born granule neurons lacking ApoE or expressing the human ApoE4 isoform also occur after experimental TBI, but in a much more exaggerated manner. Indeed, after CCI, adult-born granule cells from mice lacking ApoE or expressing ApoE4 show impaired dendritic arborization (i.e. complexity, cumulative length, Sholl analysis, and spine density). Moreover, both the dendritic complexity and the spine density of injury-induced adult-born granule neurons were similar in the 3 studied areas (iGCL/oGCL and Contralateral hemispheres) for all 4 genotypes, revealing no injury-driven effect for these parameters (S1 & S2 Figs). Thus, the differences observed in dendritic morphology and spine density when comparing WT with KO or E3 with E4 were strictly driven by genotype. Interestingly, we observe that ApoE deficiency nearly phenocopies the substitution of mouse ApoE with human ApoE4, thereby strengthening the hypothesis that ApoE4 works more as negative regulator of hippocampal neurogenesis and development. This is consistent with our previous published observations regarding ApoE and neurogenesis but the underlying mechanisms remain unclear [17, 21, 42].

ApoE is the major lipid carrier of the brain and participates primarily in cholesterol transportation from astrocytes to neurons, where it affects both dendritic development and synaptogenesis in an LDL receptor-dependent manner [55]. The role of cholesterol in neurite development and synapse formation were established first in vitro [56, 57] where it has been shown that while human ApoE3 promotes neurite outgrowth, ApoE4 inhibits it [58]. Since granule neurons do not express ApoE once they mature [20], the observed dendritic impairments are likely from astrocytic-ApoE, which is consistent with several studies that demonstrated an astrocytic requirement for normal dendritic development [59-61]. More recently, ApoE isoform-dependent roles in regulating synaptic pruning by astrocytes in the developing mouse brain have been established [62].

Together, these previous findings may help to explain the cognitive impairments exhibited by mice lacking ApoE or expressing ApoE4 particularly in spatial learning/memory and in olfactory memory, two neurogenic-dependent behaviors [63-71]. Although it remains unclear whether recovery from injuries such as TBI is dependent on neurogenesis in general and ApoE state in particular, the present study adds needed insight into a potential mechanism linking the two.

Conclusions

In the present study, we demonstrate that the dendritic complexity, dendritic length, and spine density of injury-induced mature adult-born granule neurons are not impaired 4 weeks after CCI in WT adult mice. We have, however, observed dendritic reorganization in injury-induced adult-born granule neurons, suggesting that they adapt their dendritic tree in response to the locally injured environment. We have also further uncovered the central role for ApoE in injury-induced neurogenesis in the adult hippocampus, where it influences dendritogenesis and synaptogenesis. Because injury-induced neurogenesis is necessary for at least some aspects of cognitive recovery, the dendritic impairments observed in mice lacking ApoE or expressing the ApoE4 human isoform may help inform the greater risk for poor outcomes in individuals possessing an ApoE4 genotype and subjected to TBI, and strengthen the ApoE4 allele association with the emergence of hippocampal-related cognitive decline.

WT and KO intra-condition comparisons of dendritic morphology and Sholl analysis.

WT injury-induced adult-born granule neurons found in the oGCL branched closer to their cell body when compared to adult-born granule neurons found in the iGCL (p < 0.001) or contralateral (p < 0.01) dentate gyrus (A) while the number of dendritic divisions is similar between adult-born granule neurons independently of their soma localization (B). (C) Sholl analysis exposed significant differences in dendritic branching observed in both proximal and distal regions when comparing the oGCL with the Contralateral side (red stars) and in more distal regions when comparing iGCL with Contralateral side (green stars). (D) The cumulative dendritic length is similar in the three conditions. (E) Injury-induced adult-born granule neurons found in the oGCL have a wider dendritic span when compared to matching cells from the contralateral side (p < 0.05) but not the iGCL. (F) Sholl analysis also revealed differences in the dendritic patterns of cells found in the iGCL with the oGCL in both proximal and distal regions. 4 mice/condition and at least 10 neurons/mouse were analyzed; iGCL: 61 cells; oGCL: 45 cells; Contra: 60 cells. ApoE KO injury-induced adult-born granule neurons found in the oGCL branched closer to their cell body when compared to adult-born granule neurons found in the iGCL (p < 0.001) or contralateral (p < 0.001) dentate gyrus (G) while the number of dendritic divisions is similar between adult-born granule neurons independently of their soma localization (H). (I) Sholl analysis exposed significant differences in dendritic branching observed in both proximal and distal regions when comparing the oGCL with the Contralateral side (red stars) and in more distal regions when comparing iGCL with Contralateral side (green stars). (J) The cumulative dendritic length was decreased in neurons found in the oGCL when compared to iGCL or Contralateral side (p < 0.001). (K) Injury-induced adult-born granule neurons found in the oGCL have a wider dendritic span when compared to matching cells from the iGCL (p < 0.05) but not the contralateral side. (L) Sholl analysis also revealed differences in the dendritic patterns of cells found in the iGCL with the oGCL in both proximal and distal regions. 4 mice/condition and at least 10 neurons/mouse were analyzed; iGCL: 77 cells; oGCL: 73 cells; Contra: 89 cells; iGCL = Inner Granule Cell Layer, oGCL = Outer Granule Cell Layer, in the ipsilateral dentate gyrus. Contra = Contralateral dentate gyrus. *p < 0.05, **p < 0.01, ***p < 0.001. Detailed statistics can be found in S1 Table.

(TIF)

Click here for additional data file.

ApoE3 and ApoE4 intra-condition comparisons of dendritic morphology and Sholl analysis.

ApoE3 injury-induced adult-born granule neurons found in the oGCL branched closer to their cell body when compared to adult-born granule neurons found in the iGCL (p < 0.01) or contralateral (p < 0.05) dentate gyrus (A) while the number of dendritic divisions is similar between adult-born granule neurons independently of their soma localization (B). (C) Sholl analysis exposed significant differences in dendritic branching observed in both proximal and distal regions when comparing the iGCL with the contralateral side (blue stars) and minor differences in proximal regions when comparing oGCL with contralateral side (purple stars). (D) The cumulative dendritic length in the contralateral side was reduced when compared to iGCL (p < 0.05). (E) Injury-induced adult-born granule neurons found in the oGCL have a wider dendritic span when compared to matching cells from the iGCL (p < 0.05) but not the contralateral side. (F) Sholl analysis also revealed differences in the dendritic patterns of cells found in the iGCL with the oGCL in both proximal and distal regions. 4 mice/condition and at least 10 neurons/mouse were analyzed; iGCL: 81 cells; oGCL: 62 cells; Contra: 81 cells. ApoE4 injury-induced adult-born granule neurons found in the oGCL branched closer to their cell body when compared to adult-born granule neurons found in the iGCL (p < 0.001) or contralateral (p < 0.01) dentate gyrus (G) while the number of dendritic divisions is similar between adult-born granule neurons independently of their soma localization (H). (I) Sholl analysis exposed significant differences in dendritic branching observed in both proximal and distal regions when comparing the oGCL with the Contralateral side (purple stars) while minor differences have been found when comparing iGCL with Contralateral side (blue stars). (J) The cumulative dendritic length was decreased in neurons found in the oGCL when compared to iGCL (p < 0.05) or contralateral side (p < 0.01). (K) Injury-induced adult-born granule neurons found in the oGCL have a wider dendritic span when compared to matching cells from the iGCL (p < 0.001) and the contralateral side (p < 0.05). (L) Sholl analysis also revealed differences in the dendritic patterns of cells found in the iGCL with the oGCL in both proximal and distal regions. 4 mice/condition and at least 10 neurons/mouse were analyzed; iGCL: 102 cells; oGCL: 65 cells; Contra: 89 cells; iGCL = Inner Granule Cell Layer, oGCL = Outer Granule Cell Layer, in the ipsilateral dentate gyrus. Contra = Contralateral dentate gyrus. *p < 0.05, **p < 0.01, ***p < 0.001. Detailed statistics can be found in S1 Table.

(TIF)

Click here for additional data file.

Summary of statistics for intra-genotype comparisons.

DF = Degrees of Freedom; iGCL = Inner Granule Cell Layer, oGCL = Outer Granule Cell Layer, in the ipsilateral dentate gyrus; WT = Wild Type. KO = Knockout. The Sholl Analysis represents the average number of intersections in each condition (Column) function of the distance to the soma (Row).

(PDF)

Click here for additional data file.

29 Oct 2019

PONE-D-19-27231

Apolipoprotein E regulates the maturation of injury-induced adult-born hippocampal neurons following traumatic brain injury

PLOS ONE

Dear Dr. Kernie,

Thank you for submitting your manuscript to PLOS ONE. After careful consideration by 2 Reviewers and an Academic Editor, all of the critiques of both Reviewers must be addressed in detail in a revision to determine publication status. If you are prepared to undertake the work required, I would be pleased to reconsider my decision, but revision of the original submission without directly addressing the critiques of the two Reviewers does not guarantee acceptance for publication in PLOS ONE. If the authors do not feel that the queries can be addressed, please consider submitting to another publication medium. A revised submission will be sent out for re-review. The authors are urged to have the manuscript given a hard copyedit for syntax and grammar.

==============================

Comments to the Author

1. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Yes

Reviewer #2: Partly

**********

2. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

Reviewer #2: Yes

**********

3. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

Reviewer #2: Yes

**********

4. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: Yes

**********

5. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: This manuscript describes the influence of APOE genotype on morphological characteristics of adult born granule cells. They have used stereotactic injections of GFP expressing retroviruses which labels adult born hippocampal neurons. They have also used the stereoinvestigator and neurolucida platforms to perform their analyses of morphological changes to dendrites and spines of adult born granule cells.

The findings from this manuscript supports previous studies in the field and provide some new perspectives in the morphological changes in the brains of injured APOE-KO and APOE4 mice. However the reviewer feels that separate analyses and a reanalyses of their datasets are needed, including a reconsideration of their conclusions.

Introduction:

46-52: If TBI activates NSPCs to give rise to adult born granule neurons, and ablation of injury induced hippocampal adult born neurons impairs cognitive recovery after TBI, does this imply that the injury effect is neuroprotective? Would be helpful if the "controversy" or different ideas are explained a little further to lay the groundwork as it regards to the general consensus in the field.

Types of TBI are not addressed (in relation to APOE/TBI outcomes and TBI/neurogenesis effects).

Effect of APOE4 in the context of TBI involves different mechanisms, which have been well explored molecularly in several studies. How much neurogenesis plays in the APOE4 dependent neurodegenerative or behavioral effects remains unknown?

Significant papers downplaying the role of neurogenesis in TBI are not referenced.

Design/Methods/results:

No sham mice have been included, as the contralateral hemisphere is used as a control. This is problematic and needs to be addressed.

As a result the authors focus on genotype effects in injured mice alone.

The authors need to include sham mice (sham injuries/cranial window and injections only), or at least demonstrate in this manuscript that sham mice (ipsilateral hemisphere with cranial window) show comparable changes with the contralateral hemisphere of injured mouse brains.

Additionally, data needs to be presented to show that there were no difference in the iGCL vs oGCL of the contralateral region.

There is no mention of injury effects within each genotype for APOE KO, APOE3 and APOE4 mice (Figure 2 and 3, 4A-B). For example no changes in spine density (between ipsi vs contra) was observed in WT, KO, E3 and E4 mice. In this case, the changes in the contra hemisphere were similar to the changes in the ipsi hemisphere, showing that the effects are driven by genotype and not injury. This should be highlighted and discussed.

Data in Figure 1 are included in graphs/analyses of data in Figure 2 and 4 which is redundant and repetitive.

It will be better to include the analyses of all groups on the same graph as in Fig 4E-G (This would be very busy but easier for the reader to compare across the different groups if clear labeling is used).

Rationale for including APOE-KO mice is not clear, as this is not relevant to humans (the main question is the influence of APOE3 vs APOE4). Moreover comparison of APOE-KO mice with WT also does not add much to our understanding as mice express a different form of APOE.

Why did the authors not collect cell counts in their analyses using the optical fractionator?

It was stated that astrocytes drive NSPCs to astrogenesis, it would add value to the paper if the authors obtained double staining for differentiated astrocytes as they did in their previous paper with nestin, ki-67 and GFAP etc... and APOE (Tensaouti et al., 2018).

If mice were administered Brdu, it would be good to include Brdu/Prox1 cell density in the different granular layers.

200: Says "mixed genotype and....." however the legend states that the data were from WT mice.

What was the rationale for injuries at 6 weeks, this is still very young and the brain is still developing in mice.

Discuss in some more detail, the rationale for the injection at 4 weeks post-injury. How does this relate to the timeline of events after CCI, and what could be missed from this timepoint of analyses.

211: is there another term that could be used instead of "spread" of adult granule neurons.

Discussion:

Their discussion needs to be expanded and re-written based on the comments above.

Reviewer #2: In the manuscript, the authors investigated the effects of ApoE gene in the regulation of the maturation of injury-induced adult-born neurons in the hippocampus following traumatic brain injury. Overall, the manuscript is interesting and contributes to the understanding of ApoE genes in the regulation of adult hippocampal neurogenesis after brain injury. However, some concerns need to be addressed.

1. In the study, mixed sex mice were used as the authors claimed. What is the percentage of each sex mice in each group? Are there the same numbers of female and male mice investigated in each genotype? It is known that there is sex difference in basal adult neurogenesis, however, the authors did not mention the numbers of male and female mice in each group they studied.

2. The authors should provide more information about the retrovirus carring eGFP they used in the study, it will help the readers understand the work more easily.

3. In result2, the authors claimed that they found significant interaction in both the iGCL and oGCL. However, in table1, the results shows the interaction of sholl analysis in oGCL is not significant.

4. In the two-way ANOVA they used to analyze the sholl anaylasis results, they provided the details of the results. However, what is the row factor in the analysis? They should specifically point out what is row factor and what is column factor.

5. In result4, they claimed that "the reduced spine density in ApoE-KO mice adult-born neurons is exaggerated after CCI". However, the result of the ApoE-KO adult born neurons without CCI is published in another paper. I am wondering if the authors conduct the two studies by using the mice purchased at the same time and performed the studies at the same time? If not, it is not very appropriate to compare these two results.

6. More details about the ApoE-KO, ApoE3 and ApoE4 mice will help the readers understand the manuscript better.

**********

6. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: Yes: Hao Wang

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email us at figures@plos.org. Please note that Supporting Information files do not need this step.

==============================

We would appreciate receiving your revised manuscript by February 2020. When you are ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter.

To enhance the reproducibility of your results, we recommend that if applicable you deposit your laboratory protocols in protocols.io, where a protocol can be assigned its own identifier (DOI) such that it can be cited independently in the future. For instructions see: http://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols

Please include the following items when submitting your revised manuscript:

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). This letter should be uploaded as separate file and labeled 'Response to Reviewers'.

A marked-up copy of your manuscript that highlights changes made to the original version. This file should be uploaded as separate file and labeled 'Revised Manuscript with Track Changes'.

An unmarked version of your revised paper without tracked changes. This file should be uploaded as separate file and labeled 'Manuscript'.

Please note while forming your response, if your article is accepted, you may have the opportunity to make the peer review history publicly available. The record will include editor decision letters (with reviews) and your responses to reviewer comments. If eligible, we will contact you to opt in or out.

We look forward to receiving your revised manuscript.

Kind regards,

Stephen D. Ginsberg, Ph.D.

Section Editor

PLOS ONE

Journal Requirements:

1. When submitting your revision, we need you to address these additional requirements. Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at http://www.journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and http://www.journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf

29 Jan 2020

Reviewer #1: INTRODUCTION

1)46-52: If TBI activates NSPCs to give rise to adult born granule neurons, and ablation of injury induced hippocampal adult born neurons impairs cognitive recovery after TBI, does this imply that the injury effect is neuroprotective? Would be helpful if the "controversy" or different ideas are explained a little further to lay the groundwork as it regards to the general consensus in the field.

We thank the reviewer for highlighting an ongoing controversy regarding injury-induced neurogenesis. It is clear that rodent modeling of TBI using a variety of injury modeling including the controlled cortical impact (CCI) model that we use here leads to neuronal cell death in the dentate with loss of both mature and immature DCX+ expressing neurons (Gao et al., 2008, Yu et al., 2008). It has also been shown that injury-induced activation of resident progenitors leads to neurogenesis that compensates for the loss of these vulnerable cell populations. These injury-induced neurons functionally integrate into the pre-existing circuitry (Villasana et al., 2015) and appear to be necessary for at least some aspects cognitive recovery in mice (Blaiss et al., 2011). In addition, it has been reported that these new neurons have morphologic aberrancies that may also increase seizure susceptibility and contribute to other aspects of brain injury pathophysiology (Danzer, 2018). Thus, it remains unclear what the overall effect of injury-induced neurogenesis following TBI might be, though its relevance to the overall adaptation for the hippocampus to traumatic brain injury is clear. We have added additional discussion around these points in the introduction (p3. lines 47 to 58).

2) Types of TBI are not addressed (in relation to APOE/TBI outcomes and TBI/neurogenesis effects).

We agree that we could make a clearer picture about the role of ApoE in human TBI as well as what is known about ApoE specifically around injury-induced neurogenesis. The ApoE4 allele is associated with small vessel disease and cognitive impairment on a vascular basis i.e. hypoxic-ischemic white matter damage and resulting dementia (Koizumi et al., 2018). ApoE4 is also associated with a greater incidence of moderate or severe contusions (Smith et al., 2006) as well as concussions (Merritt et al., 2018). Moreover, ApoE4 genotype combined with TBI is thought to increase the risk of developing Tauopathy and Alzheimer’s Disease (Cao et al., 2017) as well as post-traumatic epilepsy (Diaz-Arrastia et al., 2003) and impaired spontaneous blood brain barrier repair (Main et al., 2018). Neurogenesis is affected in an ApoE isoform-dependent manner after both ischemic stroke (Tobin et al., 2014) and controlled cortical impact (Hong et al., 2016) in mouse models but not after concussion (Wang et al., 2016). Recently a large meta-analysis in humans concluded that outcome following TBI occurs in an ApoE isoform dependent manner thus validating its importance in overall TBI but where the underlying mechanism remains entirely unknown (McFadyen et al., 2019. We have added additional discussion around these points in the introduction (p4. lines 80 to 89).

3) Effect of APOE4 in the context of TBI involves different mechanisms, which have been well explored molecularly in several studies. How much neurogenesis plays in the APOE4 dependent neurodegenerative or behavioral effects remains unknown?

We agree with the reviewer that it remains unclear how much neurogenesis plays in ApoE4-dependent behavioral deficits seen in any brain associated pathology. Our intent here is to highlight how injury-induced neurogenesis appears impaired in ApoE4 states in a manner that is similar to ApoE deficiency. Under normal conditions, mice lacking ApoE or expressing the E4 human allele have reduced neurogenesis and a smaller NSPC pool (Hong et al., 2016) as well as impaired dendritic maturation of surviving adult-born granule cell (Tensaouti et al., 2018). Moreover, injury-induced neurogenesis is greatly diminished in ApoE-deficient mice and decreased in ApoE4-expressing mice (Hong et al., 2016). Hippocampal-related cognitive performance is also impaired in healthy ApoE4 mice (Hartman et al., 2001; Peng et al., 2017; East et al., 2018) as well as in injured ApoE4 mice (Mannix et al., 2011; Teng et al., 2017). Furthermore, ApoE4 is the strongest genetic risk factor for late onset Alzheimer’s disease (Liu et al., 2013) where loss of neurogenesis has been linked with disease progression (Moreno-Jimenez et al., 2019). We have added in a number of references (noted above) both in the Introduction and Discussion to highlight these points.

4) Significant papers downplaying the role of neurogenesis in TBI are not referenced.

- Radoslaw Rola, Shinichiro Mizumatsu, Shinji Otsuka, Duncan R. Morhardt, Linda J. Noble-Haeusslein, Kelly Fishman, Matthew B. Potts, John R. Fike, Alterations in hippocampal neurogenesis following traumatic brain injury in mice, Volume 202, Issue 1, 2006, Pages 189-199, ISSN 0014-4886, https://doi.org/10.1016/j.expneurol.2006.05.034.

- R. Mark Richardson, Dong Sun, M. Ross Bullock, Neurogenesis After Traumatic Brain Injury, Neurosurgery Clinics of North America, Volume 18, Issue 1, 2007, Pages 169-181, ISSN 1042-3680, https://doi.org/10.1016/j.nec.2006.10.007.

- Kernie SG, Parent JM. Forebrain neurogenesis after focal Ischemic and traumatic brain injury. Neurobiol Dis. 2010;37(2):267–274. doi:10.1016/j.nbd.2009.11.002

We thank the reviewer for bringing these to our attention. We agree that it remains unclear exactly how important injury-induced neurogenesis might be following traumatic brain injury and based on the suggestion here, have now expanded on this point further and included the relevant citations (p3. lines 49 to 58).

Reviewer #1: Design/Methods/results

5)No sham mice have been included, as the contralateral hemisphere is used as a control. This is problematic and needs to be addressed. As a result the authors focus on genotype effects in injured mice alone. The authors need to include sham mice (sham injuries/cranial window and injections only), or at least demonstrate in this manuscript that sham mice (ipsilateral hemisphere with cranial window) show comparable changes with the contralateral hemisphere of injured mouse brains. Additionally, data needs to be presented to show that there were no difference in the iGCL vs oGCL of the contralateral region.

We recently published the dendritic arborizations and spine density in uninjured dentate gyrus neurons in wildtype and various ApoE conditions (Tensaouti et al., 2018). We compared these to the contralateral neurons in injured animals analyzed in the present manuscript and found no significant differences, and therefore believe this serves as an adequate control without bringing in data from our already published work. We have now made this point clear in the materials and methods as a justification for using the contralateral side as a control (p10. lines 199 t0 204). One hallmark of TBI in the hippocampus is the increase of neurogenesis and the outward migration of adult born granule cells in the ipsilateral dentate gyrus only. We only observe very rare oGCL neuron on the contralateral side and are therefore unable to compare them to iGCL neurons and therefore are not able to do this analysis. This is predicted from what we and others have shown following unilateral traumatic injury (Villasana et al., 2015).

6) There is no mention of injury effects within each genotype for APOE KO, APOE3 and APOE4 mice (Figure 2 and 3, 4A-B). For example no changes in spine density (between ipsi vs contra) was observed in WT, KO, E3 and E4 mice. In this case, the changes in the contra hemisphere were similar to the changes in the ipsi hemisphere, showing that the effects are driven by genotype and not injury. This should be highlighted and discussed.

We have added 2 supplementary figures to describe the intra-condition comparisons of dendritic morphology to highlight the changes driven by the injury in each genotype. Both the dendritic complexity and the spine density of injury-induced adult-born granule neurons were similar in the 3 studied areas (iGCL/oGCL and Contralateral hemispheres) for all 4 genotypes, revealing no injury-driven effect for these parameters. Thus, the differences observed in dendritic morphology and spine density when comparing WT with KO or E3 with E4 were strictly driven by genotype and are now highlighted in the discussion (p21-22 lines 383-389 and p24. lines 449 to 458).

7) Data in Figure 1 are included in graphs/analyses of data in Figure 2 and 4 which is redundant and repetitive. It will be better to include the analyses of all groups on the same graph as in Fig 4E-G (This would be very busy but easier for the reader to compare across the different groups if clear labeling is used).

We thank the reviewer for the comment and agree that concentrating the data without being redundant makes the story clearer. In this revised version we completely reworked how we present the data and have added in new data that we believe improve the manuscript. We have enhanced Figure 1 with new data showing ApoE expression in the various mice in various ApoE expressing states (see response to #8 below). We have now combined the original Figure 2 and 4 into one consolidated Figure 2 and similarly with original Figures 3 and 4 have been condensed into one consolidated Figure 3. In addition, we made a variety of what we believe are useful intra-group comparisons in Supplementary Figures 1 and 2. While useful for making obvious comparisons, the data are redundant with that presented now in Figures 2 and 3 and we therefore believe based on this reviewer’s comments that the main points of the story are now highlighted in a less confusing albeit highly concentrated manner.

8) Rationale for including APOE-KO mice is not clear, as this is not relevant to humans (the main question is the influence of APOE3 vs APOE4). Moreover comparison of APOE-KO mice with WT also does not add much to our understanding as mice express a different form of APOE.

We agree that examining ApoE-deficient mice is not as relevant in humans but it does provide insight into basic mechanisms of what ApoE does in the setting of injury-induced neurogenesis. The purpose of this study was two-fold: 1) to examine ApoE in injury-induced neurogenesis and to 2) to examine how human isoforms of ApoE function in the setting of injury-induced neurogenesis. Interestingly, we observe that ApoE deficiency nearly phenocopies the substitution of mouse ApoE with human ApoE4, thereby strengthening the hypothesis that ApoE4 works more as negative regulator of hippocampal neurogenesis. This is consistent with our previous published observations regarding ApoE and neurogenesis but the underlying mechanisms remain unclear (Yang et al., 2011, Hong et al., 2016, Tensaouti et al., 2018). We have added additional information around this in the discussion (p24. lines 449 to 458).

9) Why did the authors not collect cell counts in their analyses using the optical fractionator?

Retroviruses infect dividing cells stochastically, and therefore do not infect all dividing cells that are present in the area injected. Therefore, using classical quantification techniques such as with an optical fractionator are not accurate. The purpose of this particular study was to quantitatively assesses the characteristics of the dendritic arborizations and spine density, which are not dependent on quantitative cell counts. Since we have previously published quantitative cell counts of injury-induced neurogenesis in various ApoE states and do not feel it would add anything to the present story.

10) It was stated that astrocytes drive NSPCs to astrogenesis, it would add value to the paper if the authors obtained double staining for differentiated astrocytes as they did in their previous paper with nestin, ki-67 and GFAP etc... and APOE (Tensaouti et al., 2018).

We appreciate the suggestion and we have added in new data as suggested with a revised Figure 1 where we now show immunostaining for ApoE, GFAP and GFP in the dentate gyrus from the 4 genotypes analyzed.

11) If mice were administered Brdu, it would be good to include Brdu/Prox1 cell density in the different granular layers.

We did not administer Brdu to these animals as the work alluded to has been done by us previously and the focus here was in quantifying neuron development with Sholl analysis and spine density.

12) 200: Says "mixed genotype and....." however the legend states that the data were from WT mice.

Thank you for bringing this to our attention, the text has been corrected accordingly.

13) What was the rationale for injuries at 6 weeks, this is still very young and the brain is still developing in mice.

This is the timepoint we have chosen for most of our studies involving dentate gyrus neurogenesis as, although it is young in mice, it is a timepoint known to be well past the early childhood equivalent in humans. Moreover, the neurons were not analyzed until 10 weeks of age which is well into “early adulthood” of mice.

14) Discuss in some more detail, the rationale for the injection at 4 weeks post-injury. How does this relate to the timeline of events after CCI, and what could be missed from this timepoint of analyses.

Injections were done immediately after injury, but brains were analyzed 4 weeks later. We choose this 4-week timepoint deliberately, because it is the time necessary for an adult born granule cell to become fully mature and integrated to the pre-existing circuitry (Toni and Schinder, 2015). Here, we are interested in studying injury-induced adult-born granule cells that survived and matured and have integrated into the existing circuitry. These aspects of development are separated from peaks of proliferation, neural degeneration and inflammation, which occur much sooner (Zhou et al., 2012; Taib et al., 2017) as does the speed of dendritic development (Villasana et al., 2015).

15) 211: is there another term that could be used instead of "spread" of adult granule neurons.

As suggested, we have changed this to dendritic span.

16) Reviewer #1: Discussion: Their discussion needs to be expanded and re-written based on the comments above.

We appreciate this reviewer’s thoughtful and through critique. In addition to the specific points outlined in 1-15 above, the discussion has been reworked and expanded as suggested. Additions are now visible in blue.

Reviewer #2

In the manuscript, the authors investigated the effects of ApoE gene in the regulation of the maturation of injury-induced adult-born neurons in the hippocampus following traumatic brain injury. Overall, the manuscript is interesting and contributes to the understanding of ApoE genes in the regulation of adult hippocampal neurogenesis after brain injury. However, some concerns need to be addressed.

1) In the study, mixed sex mice were used as the authors claimed. What is the percentage of each sex mice in each group? Are there the same numbers of female and male mice investigated in each genotype? It is known that there is sex difference in basal adult neurogenesis, however, the authors did not mention the numbers of male and female mice in each group they studied.

Thank you for bringing this to our attention, we used a total of 16 animals, 4 in each group, with 2 males and 2 females in each case. The manuscript has been updated accordingly (p6, lines 110 to 111).

2) The authors should provide more information about the retrovirus carrying eGFP they used in the study, it will help the readers understand the work more easily

.

Retroviruses, as opposed to adenoviruses which infect post-mitotic cells and lentiviruses which inject both post-mitotic and mitotic cells, can only infect mitotic (actively dividing) cells. As a result, only neural stem and progenitor cells from the neuronal lineage can express GFP in the infected dentate gyrus. Other potentially dividing cells such as reactive astrocytes and microglia can morphologically be differentiated without ambiguity. Therefore, we chose to use a GFP-expressing retrovirus so that we can capture the progenitor cells when they are dividing and follow their development, which takes 3-4 weeks before they are fully mature, hence our choosing this timeframe after injection. We have made these points clearer in the materials and methods and results (see p7, lines 128 to 131).

3) In result2, the authors claimed that they found significant interaction in both the iGCL and oGCL. However, in table1, the results shows the interaction of sholl analysis in oGCL is not significant.

Thank you for bringing this to our attention, this has been corrected.

4) In the two-way ANOVA they used to analyze the sholl anaylasis results, they provided the details of the results. However, what is the row factor in the analysis? They should specifically point out what is row factor and what is column factor.

The Sholl Analysis represents the average number of intersections in each condition (Column) function of the distance to the soma (Row). We have updated the statistic table legend (p11, line 213).

5) In result4, they claimed that "the reduced spine density in ApoE-KO mice adult-born neurons is exaggerated after CCI". However, the result of the ApoE-KO adult born neurons without CCI is published in another paper. I am wondering if the authors conduct the two studies by using the mice purchased at the same time and performed the studies at the same time? If not, it is not very appropriate to compare these two results.

The two studies began at the same time and were conducted largely simultaneously and in the exact same conditions; only ApoE4 mice were purchased specially for the experiments, while WT, ApoE KO and ApoE3 mice are continuously generated by our lab. We have completely reworked the figures as suggested by Reviewer 1 to hopefully clarify these points and not to present redundant data. Further comparisons that do present some of these redundant data are now included in Supplementary Figures 1 and 2.

6) More details about the ApoE-KO, ApoE3 and ApoE4 mice will help the readers understand the manuscript better.

Thank you for the suggestion and a number of details about these mice have been added to the discussion including (in summary):

It has been shown that mice lacking ApoE or expressing ApoE4 exhibit cognitive impairments (Fuentes et al., 2018; Oitzl et al., 1997) (Masliah et al. 1997); (Grootendorst et al. 2001); (Peister et al. 2006); (Rodriguez et al. 2013); (Salomon-Zimri et al. 2015); (Peng et al. 2017); (East et al. 2018) while ApoE3 and WT mice are similar, particularly in spatial learning/memory and in olfactory memory, two neurogenic-dependent behaviors (p25. lines 449 to 458).

REFERENCES

1. Gao X, Deng-Bryant Y, Cho W, Carrico KM, Hall ED, Chen J. Selective death of newborn neurons in hippocampal dentate gyrus following moderate experimental traumatic brain injury. J Neurosci Res. 2008;86(10):2258-70.

2. Yu TS, Zhang G, Liebl DJ, Kernie SG. Traumatic brain injury-induced hippocampal neurogenesis requires activation of early nestin-expressing progenitors. J Neurosci. 2008;28(48):12901-12.

3. Villasana LE, Kim KN, Westbrook GL, Schnell E. Functional Integration of Adult-Born Hippocampal Neurons after Traumatic Brain Injury(1,2,3). eNeuro. 2015;2(5):ENEURO.0056-15.2015. Published 2015 Sep 28. doi:10.1523/ENEURO.0056-15.2015

4. Blaiss CA, Yu TS, Zhang G, Chen J, Dimchev G, Parada LF, et al. Temporally specified genetic ablation of neurogenesis impairs cognitive recovery after traumatic brain injury. J Neurosci. 2011;31(13):4906-16

5. Danzer SC. Contributions of Adult-Generated Granule Cells to Hippocampal Pathology in Temporal Lobe Epilepsy: A Neuronal Bestiary. Brain Plast. 2018;3(2):169-81.

6. Koizumi, K., Hattori, Y., Ahn, S.J. et al. Apoε4 disrupts neurovascular regulation and undermines white matter integrity and cognitive function. Nat Commun 9, 3816 (2018) doi:10.1038/s41467-018-06301-2

7. Smith C, Graham DI, Murray LS, Stewart J, Nicoll JA. Association of APOE e4 and cerebrovascular pathology in traumatic brain injury. J Neurol Neurosurg Psychiatry. 2006;77(3):363–366. doi:10.1136/jnnp.2005.074617

8. Victoria C Merritt, Amanda R Rabinowitz, Peter A Arnett, The Influence of the Apolipoprotein E (APOE) Gene on Subacute Post-Concussion Neurocognitive Performance in College Athletes, Archives of Clinical Neuropsychology, Volume 33, Issue 1, February 2018, Pages 36–46, https://doi.org/10.1093/arclin/acx051

9. Cao, J., Gaamouch, F.E., Meabon, J.S. et al. ApoE4-associated phospholipid dysregulation contributes to development of Tau hyper-phosphorylation after traumatic brain injury. Sci Rep 7, 11372 (2017) doi:10.1038/s41598-017-11654-7

10. Diaz-Arrastia R, Gong Y, Fair S, et al. Increased Risk of Late Posttraumatic Seizures Associated With Inheritance of APOE ϵ4 Allele. Arch Neurol. 2003;60(6):818–822. doi:https://doi.org/10.1001/archneur.60.6.818

11. Main, B.S., Villapol, S., Sloley, S.S. et al. Apolipoprotein E4 impairs spontaneous blood brain barrier repair following traumatic brain injury. Mol Neurodegeneration 13, 17 (2018) doi:10.1186/s13024-018-0249-5

12. Tobin MK, Bonds JA, Minshall RD, Pelligrino DA, Testai FD, Lazarov O. Neurogenesis and inflammation after ischemic stroke: what is known and where we go from here. J Cereb Blood Flow Metab. 2014;34(10):1573–1584. doi:10.1038/jcbfm.2014.130

13. Hong S, Washington PM, Kim A, Yang CP, Yu TS, Kernie SG. Apolipoprotein E Regulates Injury-Induced Activation of Hippocampal Neural Stem and Progenitor Cells. J Neurotrauma. 2016;33(4):362–374. doi:10.1089/neu.2014.3860

14. Wang X, Gao X, Michalski S, Zhao S, Chen J. Traumatic Brain Injury Severity Affects Neurogenesis in Adult Mouse Hippocampus. J Neurotrauma. 2016;33(8):721–733. doi:10.1089/neu.2015.4097

15. Tensaouti Y, Stephanz EP, Yu TS, Kernie SG. ApoE Regulates the Development of Adult Newborn Hippocampal Neurons. eNeuro. 2018;5(4).

16. R.E. Hartman, D.F. Wozniak, A. Nardi, et al.. Behavioral phenotyping of GFAP-apoE3 and -apoE4 transgenic mice: apoE4 mice show profound working memory impairments in the absence of Alzheimer's-like neuropathology. Exp Neurol, 170 (2) (2001), pp. 326-344

17. Peng KY, Mathews PM, Levy E, Wilson DA. Apolipoprotein E4 causes early olfactory network abnormalities and short-term olfactory memory impairments. Neuroscience. 2017;343:364–371. doi:10.1016/j.neuroscience.2016.12.004

18. East BS, Fleming G, Peng K, et al. Human Apolipoprotein E Genotype Differentially Affects Olfactory Behavior and Sensory Physiology in Mice. Neuroscience. 2018;380:103–110. doi:10.1016/j.neuroscience.2018.04.009

19. Mannix RC, Zhang J, Park J, et al. Age-dependent effect of apolipoprotein E4 on functional outcome after controlled cortical impact in mice. J Cereb Blood Flow Metab. 2011;31(1):351–361. doi:10.1038/jcbfm.2010.99

20. Teng, Z., Guo, Z., Zhong, J. et al. ApoE Influences the Blood-Brain Barrier Through the NF-κB/MMP-9 Pathway After Traumatic Brain Injury . Sci Rep 7, 6649 (2017) doi:10.1038/s41598-017-06932-3

21. Liu CC, Liu CC, Kanekiyo T, Xu H, Bu G. Apolipoprotein E and Alzheimer disease: risk, mechanisms and therapy [published correction appears in Nat Rev Neurol. 2013. doi: 10.1038/nmeurol.2013.32. Liu, Chia-Chan [corrected to Liu, Chia-Chen]]. Nat Rev Neurol. 2013;9(2):106–118. doi:10.1038/nrneurol.2012.263

22. Moreno-Jiménez, E.P., Flor-García, M., Terreros-Roncal, J. et al. Adult hippocampal neurogenesis is abundant in neurologically healthy subjects and drops sharply in patients with Alzheimer’s disease. Nat Med 25, 554–560 (2019) doi:10.1038/s41591-019-0375-9

23. Radoslaw Rola, Shinichiro Mizumatsu, Shinji Otsuka, Duncan R. Morhardt, Linda J. Noble-Haeusslein, Kelly Fishman, Matthew B. Potts, John R. Fike, Alterations in hippocampal neurogenesis following traumatic brain injury in mice, Volume 202, Issue 1, 2006, Pages 189-199, ISSN 0014-4886, https://doi.org/10.1016/j.expneurol.2006.05.034.

24. R. Mark Richardson, Dong Sun, M. Ross Bullock, Neurogenesis After Traumatic Brain Injury, Neurosurgery Clinics of North America, Volume 18, Issue 1, 2007, Pages 169-181, ISSN 1042-3680, https://doi.org/10.1016/j.nec.2006.10.007.

25. Kernie SG, Parent JM. Forebrain neurogenesis after focal Ischemic and traumatic brain injury. Neurobiol Dis. 2010;37(2):267–274. doi:10.1016/j.nbd.2009.11.002

26. Yang CP, Gilley JA, Zhang G, Kernie SG. ApoE is required for maintenance of the dentate gyrus neural progenitor pool. Development. 2011;138(20):4351-62.

27. Toni N, Schinder AF. Maturation and Functional Integration of New Granule Cells into the Adult Hippocampus. Cold Spring Harb Perspect Biol. 2015;8(1):a018903.

28. Zhou H, Chen L, Gao X, Luo B, Chen J. Moderate traumatic brain injury triggers rapid necrotic death of immature neurons in the hippocampus. J Neuropathol Exp Neurol. 2012;71(4):348–359. doi:10.1097/NEN.0b013e31824ea078

29. Taib T, Leconte C, Van Steenwinckel J, et al. Neuroinflammation, myelin and behavior: Temporal patterns following mild traumatic brain injury in mice. PLoS One. 2017;12(9):e0184811. Published 2017 Sep 14. doi:10.1371/journal.pone.0184811

30. Fuentes D, Fernández N, García Y, García T, Morales AR, Menéndez R. Age-Related Changes in the Behavior of Apolipoprotein E Knockout Mice. Behav Sci (Basel). 2018;8(3):33. Published 2018 Mar 3. doi:10.3390/bs8030033

31. Melly S. Oitzl, Monique Mulder, Paul J. Lucassen, Louis M. Havekes, Jeannette Grootendorst, E.Ron de Kloet, Severe learning deficits in apolipoprotein E-knockout mice in a water maze task, Brain Research, Volume 752, Issues 1–2, 1997, Pages 189-196, ISSN 0006-8993, https://doi.org/10.1016/S0006-8993(96)01448-5.Masliah E, Samuel W, Veinbergs I, Mallory M, Mante M, Saitoh T. 1997. Neurodegeneration and cognitive impairment in apoe-deficient mice is ameliorated by infusion of recombinant apoe. Brain Res. 751(2):307-314.

32. Grootendorst J, de Kloet ER, Dalm S, Oitzl MS. 2001. Reversal of cognitive deficit of apolipoprotein e knockout mice after repeated exposure to a common environmental experience. Neuroscience. 108(2):237-247.

33. Peister A, Zeitouni S, Pfankuch T, Reger RL, Prockop DJ, Raber J. 2006. Novel object recognition in apoe(-/-) mice improved by neonatal implantation of wild-type multipotential stromal cells. Exp Neurol. 201(1):266-269.

34. Rodriguez GA, Burns MP, Weeber EJ, Rebeck GW. 2013. Young apoe4 targeted replacement mice exhibit poor spatial learning and memory, with reduced dendritic spine density in the medial entorhinal cortex. Learn Mem. 20(5):256-266.

35. Salomon-Zimri S, Liraz O, Michaelson DM. 2015. Behavioral testing affects the phenotypic expression of apoe ε3 and apoe ε4 in targeted replacement mice and reduces the differences between them. Alzheimers Dement (Amst). 1(2):127-135.

36. Peng KY, Mathews PM, Levy E, Wilson DA. 2017. Apolipoprotein e4 causes early olfactory network abnormalities and short-term olfactory memory impairments. Neuroscience. 343:364-371.

37. East BS, Fleming G, Peng K, Olofsson JK, Levy E, Mathews PM, Wilson DA. 2018. Human apolipoprotein e genotype differentially affects olfactory behavior and sensory physiology in mice. Neuroscience. 380:103-110.

38. E. Masliah, W. Samuel, I. Veinbergs, M. Mallory, M. Mante, T. Saitoh. Neurodegeneration and cognitive impairment in apoE-deficient mice is ameliorated by infusion of recombinant apoE. Brain Res., 751 (1997), pp. 307-314

Submitted filename: plos_one_review_1_28_20_sk.docx

Click here for additional data file.

3 Feb 2020

Apolipoprotein E regulates the maturation of injury-induced adult-born hippocampal neurons following traumatic brain injury

PONE-D-19-27231R1

Dear Dr. Kernie,

We are pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it complies with all outstanding technical requirements.

Within one week, you will receive an e-mail containing information on the amendments required prior to publication. When all required modifications have been addressed, you will receive a formal acceptance letter and your manuscript will proceed to our production department and be scheduled for publication.

Shortly after the formal acceptance letter is sent, an invoice for payment will follow. To ensure an efficient production and billing process, please log into Editorial Manager at https://www.editorialmanager.com/pone/, click the "Update My Information" link at the top of the page, and update your user information. If you have any billing related questions, please contact our Author Billing department directly at authorbilling@plos.org.

If your institution or institutions have a press office, please notify them about your upcoming paper to enable them to help maximize its impact. If they will be preparing press materials for this manuscript, you must inform our press team as soon as possible and no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org.

With kind regards,

Stephen D. Ginsberg, Ph.D.

Section Editor

PLOS ONE

Comments to the Author

1. If the authors have adequately addressed your comments raised in a previous round of review and you feel that this manuscript is now acceptable for publication, you may indicate that here to bypass the “Comments to the Author” section, enter your conflict of interest statement in the “Confidential to Editor” section, and submit your "Accept" recommendation.

Reviewer #1: All comments have been addressed

Reviewer #2: All comments have been addressed

**********

2. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Yes

Reviewer #2: Yes

**********

3. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

Reviewer #2: Yes

**********

4. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

Reviewer #2: Yes

**********

5. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: Yes

**********

6. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: The authors have addressed the comments adequately and made substantial corrections to the original submission.

Reviewer #2: (No Response)

**********

7. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: Yes: Joseph O Ojo

Reviewer #2: Yes: Hao Wang

10 Feb 2020

PONE-D-19-27231R1

Apolipoprotein E regulates the maturation of injury-induced adult-born hippocampal neurons following traumatic brain injury

Dear Dr. Kernie:

I am pleased to inform you that your manuscript has been deemed suitable for publication in PLOS ONE. Congratulations! Your manuscript is now with our production department.

If your institution or institutions have a press office, please notify them about your upcoming paper at this point, to enable them to help maximize its impact. If they will be preparing press materials for this manuscript, please inform our press team within the next 48 hours. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information please contact onepress@plos.org.

For any other questions or concerns, please email plosone@plos.org.

Thank you for submitting your work to PLOS ONE.

With kind regards,

PLOS ONE Editorial Office Staff

on behalf of

Dr. Stephen D. Ginsberg

Section Editor

PLOS ONE