**3. Results**

### *3.1. Expression Pattern of TCTP in the Mouse Brain during Development*

We demonstrated that TCTP plays a critical role in survival at the mid-embryonic stage, but its role and functional mechanism in the regulation of neurogenesis at the early stage of brain development remains unknown. To address this question, we first examined whether TCTP is expressed in the developing brain and other populations of cells. TCTP RNA expression can be detected in the neural ectoderm and nervous system in mouse embryos at the E10.5 stage with in situ hybridization (Supplementary Figure S1), and the whole brain in E13.5, E16.5, P0.5, and P56 mice with qPCR (Figure 1A) with specific primers (Table 1), suggesting the presence of TCTP RNA from E13.5 to P10 involved in neural development. To gain insight into the TCTP function during brain development, we analyzed TCTP protein expression profiles in the CNS during development. As shown in Figure 1B, immunoblotting with a specific anti-TCTP antibody revealed that TCTP protein was highly expressed in whole brains from E13.5, E16.5, P0.5, and P10 but not mature mice (P56), as shown in the upper panel. The lower

panel shows a similar TCTP protein expression pattern from E12.5 to adult. A widespread TCTP protein expression pattern was confirmed by immunohistochemistry analysis in whole-brain sections from mice at E13.5 (Figure 1C), E16.5 (Figure 1D), and postnatal day 0.5 (Figure 1E), including the striatum, olfactory bulb, cerebral cortex, hippocampus, cerebellum, and brain stem. The quantification data are summarized in Figure 1F. Furthermore, TCTP protein at E13.5 detected by double immunofluorescence staining was highly expressed in the ventricular zone (Figure 1G), where neural progenitor cells are concentrated. We found that TCTP expression in the ventricular zone was dramatically decreased to a low level from E16.5 to P0.5 but increased in the cortical layer. The abundant expression pattern of TCTP in the brain suggests that it may play an important role in CNS development. To further investigate the role of TCTP in nervous system development, we disrupted TCTP expression with conditional knockout mice, specifically in the neural progenitor cells of the brain.

**Figure 1.** Expression pattern of translationally controlled tumor-associated protein (TCTP) from the embryonic to adult stage in the brain. (**A**) Relative TCTP mRNA expression at different stages in the

whole brain from wild-type mice was estimated by qRT-PCR (data presented as mean ± SEM, \* *p* < 0.05 compared with embryonic day 13.5 (E13.5), one-way ANOVA, *n* = 3–6 per group). (**B**) TCTP protein in the whole brain from control mice (*TCTPflox*/*flox*) at E12.5, 13.5, E15.5, 16.5, and postnatal day 0.5, 10, and 56 was estimated by Western blotting. (**C**–**E**) TCTP protein expression pattern in whole-brain sagittal sections from wild-type mice at E13.5 and coronal sections at E16.5 and P0.5, including the (a) cerebral cortex, (b) hippocampus, (c) cerebellum, (d) striatum, and (e) brain stem, detected by immunohistochemistry with DAB staining (brown stain indicates TCTP-positive cells) (*n* = 3 for each time point). (**F**) Summarized quantification of TCTP protein expression in (**E**). (**G**) Representative TCTP (red) and 4',6-diamidino-2-phenylindole (DAPI)/TCTP double immunofluorescence staining of brain sections from C57BL/6 mice at E13.5, E16.5, and P0.5; *n* = 3. The arrow indicates the ventricular zone area; the arrowhead indicates the cortical layer. Scale bars: C, 600 μm; D, 200 μm; E: a, c, d, e, 100 μm, b, 200 μm; G, 200 μm.

### *3.2. Generation of Conditional TCTP Knockout Mice*

To circumvent the embryonic lethality of complete disruption of the *TCTP* gene [15] and to examine the role of TCTP in CNS development, we used the Cre/loxP gene targeting system to generate neuronal progenitor cell-specific TCTP conditional mutants, which were obtained by breeding *TCTPf*/*f* mice with *Nestin-cre* transgenic mice from Jackson Lab [22,23] to produce TCTP-cKO mice (Figure 2A). The *Nestin-Cre* transgenic mice expressed Cre recombinase under the control of the rat *Nestin* promoter. The *Nestin-cre*-mediated recombination was first present in precursor cells of neurons and glia around embryonic day 11 and was detected in the cortical wall, cortical layers, ventricular zone (VZ), subventricular zone (SVZ), and intermediate zone (IZ) of the telencephalon and spinal cord of the growing population of postmitotic neurons in the developing central nervous system [27]. The generation and characterization studies of TCTP-cKO conditional mutants could help us to selectively assess the role of TCTP in neural precursor cells during central nervous system development.

Genotype analysis of progeny at P10 from *NestinCre*/+; *TCTPf*/+ mice crossed with *TCTPf*/*f* revealed that no homozygous offspring were found, suggesting embryonic or neonatal death of TCTP-cKO mice. Therefore, we checked the embryo variability for TCTP knockout mice. Homozygous offspring could be found at E9.5, E10.5, E13.5, E14.5, and E16.5. The TCTP disruption genotype *NestinCre*/+; *TCTPf*/*f* was confirmed by PCR in newborn mice, postnatal day 0.5 (Figure 2B). The four offspring genotypes (*Nestin*+/+; *TCTPf*/*f* , *Nestin*+/+; *TCTPf*/+, *NestinCre*/+; *TCTPf*/+, *NestinCre*/+; *TCTPf*/*f*) were identified in an expected Mendelian ratio (1:1:1:1). The *Nestin-cre*-derived deletion of TCTP was further demonstrated by TCTP mRNA level with qRT-PCR (Figure 2C), and protein expression with immunoblotting (Figure 2B, lower panel) and immunohistochemistry (Figure 2D). The sections at E16.5 were also detected by immunofluorescence double labeled with antibodies to TCTP to identify the *TCTP* deletion and antibodies against *Nestin* to identify neural precursor cells (Figure 2E). Mice heterozygous for the deleted allele of *TCTP* (*NestinCre*/+; *TCTPf*/+) were viable, fertile, and not morphologically different from wild-type (*NestinCre*/+; *TCTP*+/+) littermates.

**Figure 2.** Generation of conditional TCTP knockout mouse model. (**A**) Diagram of the targeting strategy for the generation of the conditional TCTP allele. The targeting construct contains *loxP* sites (arrowheads) flanking exons 3–4 and restriction sites EcoRI (E) and HindIII (H). (**B**) Genotyping with PCR, top panel, and Western blotting of TCTP expression, bottom panel, from *Nestin*+/+; *TCTPflox*/*flox* (Control), *NestinCre*/+; *TCTPflox*/*flox* (heterozygous), and <sup>N</sup>*estinCre*/+; *TCTPflox*/+ (cKO) mice at P0.5. (**C**) qRT-PCR of relative TCTP mRNA expression in the whole brain from control, heterozygous, and conditional knockout mice at different stages was estimated (data presented as mean ± SEM, \* *p* < 0.05, \*\*\* *p* < 0.001 compared with control, *#* < 0.05 compared with E13.5 cKO, one-way ANOVA, *n* = 3–11 per group). (**D**) Immunohistochemistry of TCTP protein from control and TCTP-cKO mice at P0.5. DG, dentate gyrus; Hipp, hippocampus; SVZ, subventricular zone. Scale bars: a, e, 400 μm; b, f, 200 μm; c, g, 40 μm; d, h, 100 μm. (**E**) Sections at E16.5 were also detected by double-labeled immunofluorescence with antibodies against TCTP to identify the deletion and Nestin to identify neural precursor cells. A representative image is shown for three independent experiments. Scale bars: 100 μm for right panel, 200 μm for left panel.

### *3.3. Loss of TCTP in Neuronal Progenitor Cells Resulted in Early Neonatal Death*

Compared with littermate controls, all the surviving TCTP-cKO mutant pups at P0.5 were smaller in body size (Figure 3A) and did not suckle well (Figure 3B, indicated by arrow), as demonstrated by the lack of milk in their stomachs. Abnormal behavior, particularly ataxia, was observed in cKO pups. Moreover, body weight, brain size, and brain weight were reduced in TCTP mutant pups examined at P0.5 when compared with littermate controls (Figure 3C,E,F). We found that the TCTP mutant mice exhibited early postnatal lethality phenotype at P1.5 to P2.5 (Figure 3D, Table 1). Mice heterozygous for the deleted allele of *TCTP* (*NestinCre*/+; *TCTPf*/+) were viable, fertile, and not morphologically different from wild-type (*NestinCre*/+; *TCTP*+/+) littermates. The perinatal death of TCTP-cKO mice might be caused by abnormal CNS development and malnutrition from a lack of milk in the stomach.

**Figure 3.** Phenotypes and survival rates of TCTP-cKO mice. (**A**) Body size of control and TCTP-cKO mice at P0.5. (**B**) Milk in stomachs of control and cKO mice at P0.5. (**C**) Body weight of control and cKO mice at P0.5 (data presented as mean ± SEM, \* *p* < 0.05, Student's *t* test, *n* = 3 per group). (**D**) Survival rate of offspring from *NestinCre*/+; *TCTPf*/+ mice crossed with *TCTPf*/*f* mice during development from P0.5 to P10.5. (**E**) Brain weight of control and cKO mice at P0.5 (data presented as mean ± SEM, \* *p* < 0.05, Student's *t* test, *n* = 4 per group). (**F**) Morphology of brains from control and cKO mice at E13.5, E16.5, and P0.5. Scale bar: F, 0.5 cm.


**Table 1.** Genotype analysis of o ffspring from *NestinCre*/+; *TCTPf*/+ mice crossed with *TCTPf*/*f* mice.

### *3.4. TCTP is Required for Cortical Neurogenesis*

The perinatal death of TCTP-cKO mice prompted us to further detect the brain morphology of cKO mice compared with littermate controls by hematoxylin and eosin Y staining. To assess the role of TCTP in brain development, brains were collected at E16.5, around mid-neurogenesis, and P0.5. This stage is characterized by a large population of Nestin-expressing progenitors within the VZ and subventricular zone (SVZ), early committed neuroblasts within the SVZ/intermediate zone (IZ), and a growing population of postmitotic neurons in the developing cortical plate (CP). Therefore, we examined the cortical plate, hippocampus, and lamination of SVZ in the cerebral cortex in TCTP-deficient brains compared with littermates (Figure 4A,B). The cortical plate and hippocampus region of the TCTP-cKO mouse brain was significantly decreased as expected. In contrast, the VZ/SVZ region in the cKO mouse brain was significantly increased. Enlargement of the VZ/SVZ region in TCTP-conditional null mice indicated that perhaps some progenitor cells cannot fully migrate and commit to di fferentiation or are unable to exit the cell cycle. TCTP-cKO mice at P0.5 also exhibited a size reduction in the cortical plate and hippocampus but broad and lower levels of staining in the proliferative SVZ and VZ compared with littermate controls (Figure 4C,D). The structural abnormality in the mutant brain at E16.5 and P0.5 suggests that the selective deletion of TCTP in neural progenitor cells significantly impaired the development of the cerebral cortex surrounding the lateral ventricle (Figure 4).

Furthermore, Tuj1 was expressed not only in newly committed immature postmitotic neurons but also in di fferentiated neurons and in some mitotically active neuronal precursors [28]. Therefore, the brain sections and isolated primary cortical neurons from control and TCTP conditional knockout mice at E16.5 or P0.5 were prepared for detection by double immunofluorescence staining with antibody against TCTP and Tuj1. We found that the expression of Tuj1 was reduced in the cortex and hippocampus of the brain sections (Figure 5A,B) and primary cortical neurons (Figure 5C,D) from TCTP-cKO mice at E16.5 and P0.5 by immunofluorescence analysis. The results exhibited decreased neurogenesis in the TCTP cKO mice compared with control mice (Figure 5A–D). To further investigate the impact of TCTP deletion on neurogenesis, we examined several cell-type-specific markers in the cerebral cortex, hippocampus, VZ, and SVZ. Nestin protein is expressed at high levels in cortical radial glia/neural progenitor cells (NPCs) [29]. In contrast, doublecortin (DCX) expression is low in NPCs but upregulated in postmitotic neurons [30]. Nestin and DCX are thus routinely used as markers to distinguish these mutually exclusive cell types. DCX, a marker for newly migrating populations, could a ffect neuronal migration by regulating the organization and stability of microtubules [31]. Compared to littermate controls, the brains of cKO mutant mice exhibited a significant reduction of neurogenesis marker DCX-positive neurons (Figure 5E,F) by immunohistochemistry analysis. Moreover, an in vitro study of primary cell culture indicated that high expression of TCTP was observed in both proliferating neural precursors (Nestin +) and mature neurons (MAP2 +) during brain development. Some TCTP-negative cells from mutant mice expressed Nestin and survived at first (Supplementary Figure S2A). In contrast, most surviving cells cultured from mutant mice expressed TCTP and MAP but not TCTP-negative cells (Supplementary Figure S2B,C). These results indicate that TCTP is involved in neurogenesis in the embryonic and early postnatal stages. Specifically, *Nestin-cre*-derived TCTP disruption resulted in decreased Tuj1-positive newly committed neurons, DCX, and Nestin-positive neural progenitors, indicating that TCTP is required for the survival of these

cell populations. Furthermore, we also demonstrated that a deficiency of TCTP possibly delayed or impaired neuronal proliferation and cell cycle progression.

**Figure 4.** Histological analysis of E16.5 embryos and P0.5 pups from control and cKO mice. (**A**) Control (*NestinCre*/+; *TCTPw*/*<sup>w</sup>*; a, c, e, g, and i) and cKO (*NestinCre*/+; *TCTPf*/*f* ; b, d, f, h, and j) brain sections from E16.5 embryos were stained with hematoxylin and eosin. (**B**) Bar graph summarizing quantification of the sizes of brain regions in (**A**). (**C**) Control (a, c, e, g, i, and k) and cKO (b, d, f, h, j, and l) brain sections from P0.5 pups were stained with hematoxylin and eosin. (**D**) Bar graph summarizing the quantification of the sizes of brain regions in (**C**), (data presented as mean ± SEM, \* *p* < 0.05, Student's *t* test, *n* = 3 per group). A representative image is shown for three independent experiments. MZ, marginal zone; SP, subplate. Scale bars on panel A: a, b, 400 μm; c, d, 200 μm; e–j, 100 μm. Scale bars on panel B: a, b, 400 μm; c, d, 200 μm; e–h, 100 μm; i–l, 40 μm.

**Figure 5.** Central nervous system (CNS) neurogenesis of TCTP-cKO and control mice was detected. (**A**) Double immunofluorescence staining detected TCTP (green) and Tuj1 (red) of brain sections at E16.5. Bottom panel shows magnification of panels a and b. (**B**) Bar graph summarizing the quantification of Tuj1-positive cells in (**A**). (**C**) TCTP (green) and Tuj1 (red) of primary cultured cortical neurons from control and TCTP-cKO mice at P0.5 were detected. (**D**) Bar graph summarizing the quantification of Tuj1-positive cells in (**C**). (**E**) The neurogenesis marker doublecortin (DCX) was detected in TCTP mutant brains from control and TCTP-cKO mice at P0.5. (**F**) Bar graph summarizing the quantification of DCX-positive cells in (**E**). (Data presented as mean ± SEM, \* *p* < 0.05, Student's *t* test, *n* = 3 per group.) Scale bars: A, 200 μm, B, 40 μm, C, a, e, 200 μm; b–d, f–h, 100 μm. A representative picture is shown for three independent experiments.

### *3.5. Decreased Cell Proliferation in Nestin-Cre-Derived TCTP-Deficient Mouse Brain*

We next examined whether the cellular proliferation and apoptosis involved the phenotype of the TCTP-cKO mutant mice. To address this issue, BrdU incorporation and TUNEL assays were determined. We examined whether loss of TCTP causes the defect in brain development and neuronal cell proliferation. The proportion of cells in the S phase of the cell cycle was determined by immunohistochemistry. High levels of BrdU expression were detected in the hippocampus, dentate gyrus, marginal zone, and subventricular zone of the cerebral cortex in the littermate control mice at P0.5; however, BrdU-positive cells were reduced in the TCTP mutant mice (Figure 6A). The quantification data at P0.5 are summarized in Figure 6B. A significant reduction of BrdU incorporation was also found in the cKO mice at E16.5 (Figure 6C), especially in the corpus callosum, but not at E13.5 (Figure 6D). The quantification data at E16.5 and E13.5 are summarized in Figure 6E. These results are consistent with the *Nestin-cre* activity, which was high in the preplate and low in the ventricular zone (VZ) at

E12.5, and high in the cortical plate (CP) and intermediate zone (IZ) and low in the VZ and SVZ at E14.5. Then, the scope of recombination expanded to neural stem cells (NSCs) and NPCs increased dramatically during E14.5 to E17.5 [32]. To determine whether decreased proliferation was correlated with perturbation of cell cycle progression, immunoblotting analysis was carried out to compare the expression levels of cyclins D2 and E2 between the *NestinCre*/+; *TCTPf*/*f* pups and littermate controls. We found that cyclin D2 and E2 expression was significantly suppressed in the TCTP-deficient brain (Figure 6F,G), which may delay or impair cell cycle progression and neuronal proliferation. These results were consistent with the BrdU incorporation assay and DCX immunohistochemistry staining. The reduced levels of cyclins D2 and E2 in *NestinCre*/+; *TCTPf*/*f* pups was also consistent with what we demonstrated: That mice with systemic disruption of TCTP exhibited decreased expression of cyclin D2 and E2 in E9.5 TCTP−/− embryos [15]. These results indicate that TCTP is required for cortical neurogenesis.

**Figure 6.** Cell proliferation and cyclin protein expression detected in brain sections and protein lysates from control and cKO mice. (**A**) Bromodeoxyuridine (BrdU) incorporation assay of control and TCTP-cKO mice was estimated at P0.5. (**B**) Bar graph summarizing the quantification of BrdU-positive

(+) cells in ( **A**). ( **C**,**D**) BrdU incorporation in brain sections from control and TCTP-cKO mice was also detected at E16.5 and E13.5 (**E**) Bar graph summarizing the quantification of BrdU-positive cells in ( **C**,**D**). (Values are mean ± SEM, Student's *t* test, \* *p* < 0.05 compared with control group; *n* = 3). CC. corpus callosum; DG, dentate gyrus; HIP, hippocampus; SVZ, subventricular zone. Scale bars: A: a, b, 200 μm; c, d, e, g, h, 40 μm; f, 100 μm. C: 40 μm; D: top panel, 400 μm; bottom panel, 40 μm. (**F**) Western blotting of cyclins D2 and E2 and TCTP in control and TCTP-cKO mice at P0.5. ( **G**) Bar graph summarizing quantification of Western blotting relative density in (**E**). Relative density is presented as the ratio of band intensity of target protein to internal control-actin. Values are mean ± SEM. Data were statistically analyzed by student's *t* test; \* *p* < 0.05, \*\* *p* < 0.01, \*\*\* *p* < 0.001 compared with control group; *n* = 3).

### *3.6. Disruption of TCTP-Induced Cell Apoptosis and Cell Autonomous Behavior*

Next, we determined whether cell-specific disruption of TCTP expression leading to increased apoptotic cell death may contribute to the decrease in di fferent kinds of neurons and neonatal death. To address this hypothesis, TUNEL assay was conducted at postnatal day 0.5. TUNEL staining was performed on coronal sections of the brain, including the hippocampus and corpus callosum. Increased TUNEL-positive cells (brown staining) were detected in *Nestin-cre*-derived TCTP-deficient brain compared with littermate controls (Figure 7A). The apoptotic cells were significantly increased in the cerebral cortex, subventricular zone, hippocampus, and corpus callosum in TCTP-cKO mutant compared with controls at P0.5 (Figure 7A,B) and E16.5 (Figure 7C) but not E13.5 (Figure 7D). The quantification data at E16.5 and E13.5 are summarized in Figure 7E.

We further evaluated whether the observed neuronal cell loss surrounding the mouse lateral ventricle resulted from apoptosis with another apoptosis marker. Coronal sections at P0.5 from *Nestin-cre*-mediated TCTP mutant brains and from control littermates were detected by immunohistochemistry with antibody agonist active caspase 3 (AC3), a hallmark of classical apoptotic cell death [18]. Numerous active caspase 3-positive cells were present throughout the cortical plate, VZ/SVZ, CP, and hippocampus of TCTP mutant mice, whereas only an occasional apoptotic cell was observed in littermate controls (Figure 8A,C). These data were consistent with the decrease of Tuj1- and DCX-positive neurons (Figure 5A,C,E) in mutant mice, suggesting that the increased apoptosis also involved the phenotype of the TCTP-cKO mutant mice. Immunohistochemistry with primary neuronal cultures was used to clarify whether the cell death of TCTP conditional knockout mice was cell autonomous or nonautonomous. The primary neuronal cultures from the TCTP-cKO mutant mice at E16.5 showed enhanced apoptotic cell death compared with control mice (Figure 8B,D). TCTP-negative cells (arrowhead), but not TCTP-positive cells (arrow), exhibited active caspase-3 positive signal (Figure 8B). This massive cell death by apoptosis observed in brain sections in vivo and TCTP-negative cells in vitro indicated that the cell death from *Nestin-cre*-derived TCTP disruption specifically in neuronal progenitor cells was caused by a cell autonomous mechanism. Again, TCTP is required for cortical neurogenesis.

**Figure 7.** Cell apoptosis was detected in brain sections from control and cKO mice. (**A**) TUNEL staining was performed on coronal sections of the brain at P0.5, including the (a,b) hippocampus, (c,d) corpus callosum, (e,f) subventricular zone, and (g,h) hippocampus CA3. (**B**) Bar graph summarizing the quantification of TUNEL-positive (+) cells in (**A**). (Values are mean ± SEM, Student's *t* test, \* *p* < 0.05, \*\* *p* < 0.01 compared with control group; *n* = 3). (**C**,**D**) Apoptosis of brain sections from TCTP-cKO and control mice at E16.5 (*n* = 3) and E13.5 (*n* = 4) was also detected by TUNEL staining assay. Boxes indicate magnification. (**E**) Bar graph summarizing quantification of TUNEL-positive cells in (**C**,**D**). (Values are mean ± SEM, Student's *t* test, \* *p* < 0.05 compared with control group). Hipp, hippocampus; SVZ, subventricular zone. Scale bar: A, C, 40 μm; D, 100 μm.

**Figure 8.** Caspase cleavage apoptosis was detected by immunohistochemistry, immunofluorescence, and Western blot. Apoptotic and antiapoptotic related protein levels were detected by Western blot. (**A**) Cerebral cortex and hippocampus of control and TCTP-cKO mice were stained for active caspase-3 (AC3) by immunohistochemistry at P0.5. (**B**) Double immunofluorescence staining for TCTP (green) and AC3 (red) of primary neuronal cell from control and TCTP-cKO mice at E16.5. The arrow indicates TCTP-positive cells, and the arrowhead indicates TCTP-negative cells. (**C**) Bar graph summarizing the quantification of active caspase-3 staining in (**A**). (**D**) Bar graph summarizing the quantification of active caspase-3 staining in (**B**). (Data presented as mean ± SEM, \* *p* ≤ 0.05, Student's *t* test, *n* = 3 per group). (**E**) Protein levels of cleaved caspase-3, bax, hax-1, bcl-xl, mcl-1, and TCTP in control and TCTP-cKO mice were measured by Western blotting at P0.5. (**F**) Bar graph summarizing quantification of Western blot in (**E**). Relative density is presented as the ratio of band intensity of target protein to the internal control, β-actin. (Data presented as mean ± SEM, \*\*\* *p* < 0.001, Student's *t* test, *n* = 3 per group). Scale bars: A, 100 μm; B, 40 μm.

Apoptotic cell death in the brain of TCTP-cKO mutant mice was also confirmed with the increased cleaved caspase-3 by Western blot. Among the interacting proteins of TCTP, we first investigated the expression of mMcl-1 [9] and Bcl-xL [10], both antiapoptotic members of the Bcl-2 family. TCTP protects from apoptotic cell death by antagonizing bax function in cooperation with mMcl-1 and Bcl-xL [33]. The *Nestin-cre*-driven TCTP knockout mice exhibited decreased expression levels of mMcl-1 and Bcl-xL but not bax protein (Figure 8E,F) compared with littermate controls. The disruption of TCTP caused Bcl-xL protein degradation (Figure 8E,F) but not decreased DNA transcription.

Antiapoptotic protein hax-1 was also reduced in mutant pups. These results demonstrate that TCTP plays a critical role in the developing nervous system when using TCTP-cKO mice. The increased apoptosis and decreased proliferation may contribute to early neonatal death. The phenotype of increased neuronal loss through apoptotic cell death in the brain of TCTP-cKO knockout mice might be attributed to the decreased cyclin D2, Mcl-1, Bcl-xL, and hax-1 expression. Therefore, TCTP is essential for the maintenance of neural precursor cell survival and di fferentiation during CNS development.

### *3.7. Conditional Deletion of TCTP in Neuron Progenitor Cells Resulted in Decreased Cell Survival and Suppression of Transcription Factor Oct4 Expression in Cortical Progenitor Cultures*

To further confirm whether neural precursor cell death observed in vivo and in vitro is caused by a cell autonomous or nonautonomous mechanism, we cultured primary cortical progenitor cells and measured their survival in vitro as an indicator of cell survival using MTT assay. Cortical progenitor cell cultures from *Nestin-cre*-mediated TCTP mutant embryos at E15.5 did not exhibit significant di fferences in cell aggregates, dendrite outgrowth, and di fferentiation at 24 h compared with littermate controls in vitro. After 3 days of culture, decreased cell viability and cell numbers were observed in TCTP-deficient cells (Figure 9A). Quantification of total cells per aggregation revealed a significant four-fold reduction in TCTP-deficient cultures, indicating that the cell death from *Nestin-cre*-derived TCTP disruption is caused by a cell autonomous mechanism.

Niwa et al. showed that a critical amount of octamer-binding transcription factor 3/4 (Oct3/4) is required to sustain stem cell self-renewal, and up- or downregulation induced divergent developmental programs [34]. In addition, a previous study showed that Tpt1 activates transcription through binding to the regulatory region of the mouse *Oct4* gene in transplanted somatic nuclei in *Xenopus laevis* oocytes [35]. To examine whether the self-renewal of TCTP mutant mouse cortical progenitor cells may encounter a problem in undi fferentiated embryonic stem cells, TCTP-modulated Oct4 expression was examined. Cortical progenitor cells were cultured from mouse brain at E15.5 and stained with antibody for Oct4 after 3 days of culture. Oct4-positive cells were detected in control groups, but a decrease was observed in the *Nestin-cre*-derived TCTP-deficient cell culture (Figure 9B). Moreover, we found significantly fewer cells and failure of di fferentiation of neuron dendrites in cortical progenitor cell culture. These results indicated that the loss of TCTP decreased cell numbers, suppressed transcription factor Oct4 expression, and disrupted cell di fferentiation in cortical progenitor cells cultured from *TCTP* mutant. The decreased Oct4 protein expression of TCTP-cKO mice was further confirmed in vivo by immunohistochemistry at E13.5 (Figure 9C) and P0.5 (Figure 9D,E) and Western blotting as early at E12.5 (Figure 9F) compared with littermate controls. Furthermore, we also studied the mRNA expression of Oct4 and cell survival-related genes with specific primers (Table 2). A significant decrease in Oct4 and Mcl-1 but not hax-1 and Bcl-xL mRNA in TCTP-cKO mice relative to control mice was observed (Figure 10). We expect that cells of cKO mice are needed for higher antiapoptotic activity, such as Bcl-xL and mcl-1, by the enhancement of mRNA expression levels. Bcl-xL protein was decreased, but Bcl-xL mRNA was increased in cKO mice at P0.5 (Figures 8C and 10E). On the other hand, mcl-1 was decreased in both protein and mRNA levels at P0.5 (Figures 8C and 10A).

**Figure 9.** Cell viability and octamer-binding transcription factor 4 (Oct4) protein expression were detected in primary cortical progenitor cultures and brain sections. (**A**) Cell viability and cell number of neuron progenitor cell cultures from TCTP-deficient mice and littermate controls at E15.5 were detected by MTT assay. (\*\*\* *p* < 0.001, Student's *t* test, *n* = 3 per group). (**B**) TCTP and Oct4 expression was detected by immunohistochemistry with antibody followed by DAB staining and then counterstaining with hematoxylin. (**C**,**D**) Oct4 expression of control and TCTP-cKO mice at E13.5 and P0.5 was detected by immunohistochemistry. A representative picture is shown for two independent experiments. (**E**) Bar graph summarizing quantification of Oct4-positive (+) cells in (**D**), (Data presented as mean ± SEM, \* *p* < 0.05, \*\* *p* < 0.01, Student's *t* test, *n* = 3 per group). (**F**) Protein level of Oct4 in control and TCTP-cKO mice was measured by Western blotting. Scale bar: A, 40 μm; C: a, d, 400 μm, b, c, e, f, 100 μm; D: a, b, d, e, 100 μm, c, f, 40 μm.

**Figure 10.** mRNA expression pattern of cell survival-related genes of TCTP-cKO mice compared with littermate controls. (**A**–**F**) Mcl-1, Hax-1, Oct-4 pair1, Oct-4 pair2, Bcl-xL, and TCTP mRNA expression from control and TCTP-cKO mice at E13.5, E16.5, and P0.5 were detected by qPCR. (Data presented as mean ± SEM, \* *p* < 0.05, \*\* *p* < 0.01, \*\*\* *p* < 0.001, Student's *t* test, *n* = 4–11 per group).



\* The pair of primers are located on exon 1 of OCT4 gene. # The pair of primers are located on exon 4 of OCT4 gene.

Taken together, our results demonstrate that TCTP is essential for cell survival during early neuronal and glial differentiation during CNS development. Thus, enhanced neuronal and functional loss and decreased cell numbers of Tuj1 and doublecortin-positive neurons mediated through Mcl-1, Bcl-xL, Oct4, and cyclin D2 and E2 suppression and caspase-3 cleavage activation resulting in increased apoptosis and decreased proliferation may contribute to the perinatal death of *TCTP* mutant mice.
