Dual Mkk4 and Mkk7 Gene Deletion in Adult Mouse Causes an Impairment of Hippocampal Immature Granule Cells

(1) Background: The c-Jun-NH2-terminal protein kinase (JNK) is a mitogen-activated protein kinase involved in regulating physiological processes in the central nervous system. However, the dual genetic deletion of Mkk4 and Mkk7 (upstream activators of JNK) in adult mice is not reported. The aim of this study was to induce the genetic deletion of Mkk4/Mkk7 in adult mice and analyze their effect in hippocampal neurogenesis. (2) Methods: To achieve this goal, Actin-CreERT2 (Cre+/−), Mkk4flox/flox, Mkk7flox/flox mice were created. The administration of tamoxifen in these 2-month-old mice induced the gene deletion (Actin-CreERT2 (Cre+/−), Mkk4∆/∆, Mkk7∆/∆ genotype), which was verified by PCR, Western blot, and immunohistochemistry techniques. (3) Results: The levels of MKK4/MKK7 at 7 and 14 days after tamoxifen administration were not eliminated totally in CNS, unlike what happens in the liver and heart. These data could be correlated with the high levels of these proteins in CNS. In the hippocampus, the deletion of Mkk4/Mkk7 induced a misalignment position of immature hippocampal neurons together with alterations in their dendritic architecture pattern and maturation process jointly to the diminution of JNK phosphorylation. (4) Conclusion: All these data supported that the MKK4/MKK7–JNK pathway has a role in adult neurogenic activity.


Introduction
The c-Jun NH2-terminal kinases (JNKs) are members of the Mitogen-Activated Protein Kinases (MAPKs) super-family. JNK activity regulates several cellular functions, such as cell growth, differentiation, survival, and apoptosis through their impact on gene expression, cytoskeletal protein dynamics, and cell death/survival pathways. In fact, the best-described mechanism linked to the JNK pathway signaling is its pro-apoptotic action following sustained or intense exposure to cellular stress (including oxidative, genotoxic, and osmotic stress) or pro-inflammatory cytokines such as tumor necrosis factor (TNF)-α and interleukin (IL)-1β. This allows regulating physiological and pathological processes Since Mkk4 and Mkk7 genetic disabling results in a lethal embryonic phenotype [7,19,21], we generated a conditional KO hemizygous Actin-Cre ERT2 (Cre +/− ), Mkk4 flox/flox Mkk7 flox/flox mice by crossing different mice colonies. Double KO Actin-Cre ERT2 (Cre +/− ), Mkk4 ∆/∆ , Mkk7 ∆/∆ mice were induced after tamoxifen administration at 2 months old. The most effective tamoxifen treatment to obtain the double deletion was a dose of 10 mg.
The hippocampus was used to elucidate the time necessary to diminish MKK4 and MKK7 protein levels. As shown by immunoblot, this decline was evident after 7 days of tamoxifen administration, and it was more noticed after 14 days ( Figure 2C). This result Since Mkk4 and Mkk7 genetic disabling results in a lethal embryonic phenotype [7,19,21], we generated a conditional KO hemizygous Actin-Cre ERT2 (Cre +/− ), Mkk4 flox/flox Mkk7 flox/flox mice by crossing different mice colonies. Double KO Actin-Cre ERT2 (Cre +/− ), Mkk4 ∆/∆ , Mkk7 ∆/∆ mice were induced after tamoxifen administration at 2 months old. The most effective tamoxifen treatment to obtain the double deletion was a dose of 10 mg.
The hippocampus was used to elucidate the time necessary to diminish MKK4 and MKK7 protein levels. As shown by immunoblot, this decline was evident after 7 days of tamoxifen administration, and it was more noticed after 14 days ( Figure 2C). This result was supported by the immunohistochemistries against MKK4 and MKK7 in the hippocampus of Actin-Cre ERT2 (Cre +/− ), Mkk4 ∆/∆ , Mkk7 ∆/∆ mice after 14 days of tamoxifen administration. Immunohistochemistries showed that MKK4 and MKK7 decline their expression in Actin-Cre ERT2 (Cre +/− ), Mkk4 ∆/∆ , Mkk7 ∆/∆ ( Figure 2D).
It is important to note that not complete removal of the proteins was achieved from both MKK4 and MKK7 in the hippocampus; for this reason, this experimental model resembles a knock-down. However, the animals presented phenotypic alterations after 16 days of tamoxifen administration, such as signs of distress, weight loss, and reduced activity. Since mortality was high after 20 days of tamoxifen administration, all experiments were carried out 14 days after tamoxifen administration. At this time, we achieved the maximum levels of MKK4 and MKK7 protein reduction without any phenotype alteration.
It is important to note that not complete removal of the proteins was achieved from both MKK4 and MKK7 in the hippocampus; for this reason, this experimental model resembles a knock-down. However, the animals presented phenotypic alterations after 16 days of tamoxifen administration, such as signs of distress, weight loss, and reduced activity. Since mortality was high after 20 days of tamoxifen administration, all experiments were carried out 14 days after tamoxifen administration. At this time, we achieved the maximum levels of MKK4 and MKK7 protein reduction without any phenotype alteration. The elimination of MKK4 and MKK7 proteins was screened in the liver, heart, and CNS of Actin-Cre ERT2 (Cre +/− ), Mkk4 ∆/∆ , Mkk7 ∆/∆ mice and their control mice groups ( Table  1). The analyses were done through Western blot, using antibodies against MKK4 ( Figure  3) and MKK7 ( Figure 4). We evaluated protein levels in WT and the other experimental The elimination of MKK4 and MKK7 proteins was screened in the liver, heart, and CNS of Actin-Cre ERT2 (Cre +/− ), Mkk4 ∆/∆ , Mkk7 ∆/∆ mice and their control mice groups ( Table 1). The analyses were done through Western blot, using antibodies against MKK4 ( Figure 3) and MKK7 (Figure 4). We evaluated protein levels in WT and the other experimental groups at 14 days after vehicle or tamoxifen administration. A reduction in MKK4 and MKK7 levels was only detected in Actin-Cre ERT2 (Cre +/− ), Mkk4 ∆/∆ , Mkk7 ∆/∆ mice (CRE (+) + T group). There was a higher reduction or apparent absence of these proteins in the heart and liver. The decreases in MKK4 levels achieved were ≈80% in cortex, ≈60% in hippocampus, ≈52.4% in cerebellum, ≈83.2% in liver, and ≈98% in heart ( Figure 3). Meanwhile, the expression levels of MKK7 dropped ≈66% in the hippocampus, ≈56% in cortex, and ≈60% in cerebellum of Actin-Cre ERT2 (Cre +/− ), Mkk4 ∆/∆ , Mkk7 ∆/∆ mice ( Figure 4). MKK7 disappeared in heart and liver after 14 days of tamoxifen administration (data not shown). groups at 14 days after vehicle or tamoxifen administration. A reduction in MKK4 and MKK7 levels was only detected in Actin-Cre ERT2 (Cre +/− ), Mkk4 ∆/∆ , Mkk7 ∆/∆ mice (CRE (+) + T group). There was a higher reduction or apparent absence of these proteins in the heart and liver. The decreases in MKK4 levels achieved were ≈80% in cortex, ≈60% in hippocampus, ≈52.4% in cerebellum, ≈83.2% in liver, and ≈98% in heart ( Figure 3). Meanwhile, the expression levels of MKK7 dropped ≈66% in the hippocampus, ≈56% in cortex, and ≈60% in cerebellum of Actin-Cre ERT2 (Cre +/− ), Mkk4 ∆/∆ , Mkk7 ∆/∆ mice ( Figure 4). MKK7 disappeared in heart and liver after 14 days of tamoxifen administration (data not shown).

Genetic Ablation of Mkk4 and Mkk7 Reduces Phosphorylation of JNK in the Hippocampus
The phosphorylation of JNK (pJNK) in the hippocampus was evaluated in WT, Actin-Cre ERT2 (

Discussion
In the present work, we generated a new adult murine model with a double deletion (Mkk4 ∆/∆ , Mkk7 ∆/∆ genotype). After 14 days of tamoxifen administration, Actin-Cre ERT2 (Cre +/− ), Mkk4 ∆/∆ , Mkk7 ∆/∆ adult mice showed a reduction of MKK4 and MKK7 proteins in the CNS and other tissues, such as the heart and liver. This diminution provoked a decrease in JNK phosphorylation, which correlated with alterations in the position, dendritic pattern, and differentiation of immature hippocampal neurons as well as with changes in the dendritic pattern of cortical neurons.

The Levels of MKK4 and MKK7 Are Not Equal in the Different Tissues
Actin-Cre ERT2 (Cre +/− ), Mkk4 ∆/∆ , Mkk7 ∆/∆ mice overcome the developmental drawbacks of conventional KOs for Mkk4 or Mkk7 [23] and even some conditional KO mice, allowing MKK4 and MKK7 protein reductions in adults. Although the decrease was induced in all the body, it was not equal in all tissues, which was probably because of differences in their physiological basal levels. Thus, while protein elimination was apparently total in the heart and liver, it was partial in the CNS, supporting that MKK4 and MKK7 levels are higher in the CNS than in other tissues, which is in accordance with the results obtained with Western blot in WT mice (Figure 1). In this line, Lee et al. showed that the levels of Mkk4 transcripts were high in the cerebral cortex, hypothalamus, hippocampus, and cerebellum of adult mice [24]. All these data emphasized that the MKK4/MKK7/JNK signaling pathway has an important role in the adult CNS [25]. Moreover, the different subcellular localizations of these proteins supports that they have distinct functions, and therefore, their levels vary in distinct tissues [21,23]. This is supported by Tournier et al., who found that the simultaneous disruption of the Mkk4 and Mkk7 genes was required to block JNK activation caused by the exposure of cells to environmental stress (e.g., ultraviolet radiation) [26]. However, with stimuli such as pro-inflammatory cytokines (e.g., TNF and IL-1), the disruption of the Mkk7 gene alone could prevent JNK activation.

MKK7 Plays an Essential Role in Heart and Liver Tissues
After analyzing the levels of MKK4 and MKK7 in different tissues of Actin-Cre ERT2 (Cre +/− ), Mkk4 ∆/∆ , Mkk7 ∆/∆ adult mice, we observed that the presence of MKK7 was reduced or absent in the heart and liver. These data are in accordance with those of Nishina et al., who evidenced low levels of MKK7 in embryonic tissue of mice, which was probably restricted to the skin, lung epithelium, and epithelial layers lining the olfactory cavity developing teeth. By contrast, MKK4 was ubiquitous and with high basal levels [21]. Despite the scarce levels of MKK7 detected in heart and liver, several studies support the notion that this protein has a critical role in these tissues, both in embryonic development and adulthood [7,27]. Indeed, Ooshio et al., through hepatocyte and hematopoietic cellspecific deletion of Mkk7, using Albumin (Alb)-Cre and Myxovirus resistance protein-1 (Mx1)-Cre line, evidenced that MKK7 is essential for wound-healing processes following parenchymal destruction by carbon tetrachloride (CCl 4 ) in the liver [27]. In addition, MKK7 suppress branching morphogenesis through the modulation of hepatocyte-extracellular matrix interaction.
Concerning MKK7 and cardiac tissue, Liu et al. revealed an essential protective role of this protein in the heart from hypertrophic insults in cardiomyocytes, hence preventing the transition to heart failure [28].

The Levels of JNK Phosphorylation
Were Decreased in Actin-Cre ERT2 (Cre +/− ), Mkk4 ∆/∆ , Mkk7 ∆/∆ Mice MAP kinase cascade, which senses cellular and extracellular stress, conveys cellular response to regulate cell fate. The timing and duration of JNK activation determines whether cells proliferate or adapt to metabolic or toxic stress or undergo programmed cell death instead, such as apoptosis, necrosis, and even other forms of cell death. MKK4/MKK7 proteins have a role in the control of JNK activation by interacting with JNK via D-motif, phosphorylating JNK [29]. Since the levels of pJNK are correlated with the activity of this signaling pathway, and they are reduced with the Mkk4 and Mkk7 deletion, we circumscribed the analysis when Actin-Cre ERT2 (Cre +/− ), Mkk4 ∆/∆ , Mkk7 ∆/∆ mice had a significant reduction of JNK phosphorylation at 14 days after tamoxifen administration. These data reinforce that Mkk4 and Mkk7 gene deletion correlated with JNK activity diminution.

The Deletion of Mkk4 and Mkk7 Gene Alters Immature Hippocampal Neurons
Tangential-to-radial migration has been described for immature hippocampal neurons [30,31]. First, neuroblasts migrate tangentially after the last division from neuron stem cell clusters through the SGZ, and then apical dendrites extend toward the molecular layer [32]. In agreement with this, neuroblasts are lined up in the SGZ and have their apical dendrites projected in the radial direction both in WT and Actin-Cre ERT2 (Cre −/− ), Mkk4 flox/flox , Mkk7 flox/flox mice. However, in Actin-Cre ERT2 (Cre +/− ), Mkk4 ∆/∆ , Mkk7 ∆/∆ mice, these cells were disarranged, since they were displaced over the GCL, maintaining immature cell markers. These finding suggest that MKK4 and MKK7 have a role in adult neuroblast migration and differentiation processes, in accordance with the results obtained by Smith, Coker, and Tucker, who identified that the JNK signaling pathway is a regulator of branching and nucleokinesis during the migration of cortical interneurons [33]. In this respect, Nestin-Cre, Mkk7 flox/flox mice showed severe defects along embryonic brain development in radial migration and axonal growth [8]. In addition, Nestin-Cre, Mkk4 flox/flox mice showed misalignment of cerebellar granule cells and defects in radial migration [34]. However, no changes in cell differentiation were identified in Mkk4 flox/flox or Mkk7 flox/flox mice under Nestin-Cre promoter in developing ages, as it occurs in our adult mice, which is probably due to the combinatorial effect of the double deletion. Moreover, the analyses with Nestin-Cre mice have the disadvantage that even though they survive after birth, eventually, they die at postnatal day 21.
The alterations detected in the dendritic projections of immature hippocampal neurons and mature cortical neurons of Mkk4 ∆/∆ and Mkk7 ∆/∆ mice support that the MKK4/MKK7-JNK signaling pathway has a role in the maintenance of the dendritic and axonal processes [35] in accordance with Bjorkblom et al., who found that JNK phosphorylation of MAP2 plays an important role in defining dendritic architecture in the brain [36].
In this line, different studies reported that JNK1 regulates neural architecture through the phosphorylation of cytoskeletal substrates [36][37][38]. Further studies should be done to determinate how MKK4/MKK7/JNK signaling is involved in all these neuronal processes.
To know the specific functions of the JNK signaling pathway in neural cell subpopulations, conditional KOs mice would be used with recombination under specific neural promoters instead of using the ubiquitous promoter Actin. In this way, there are CamKIIα-Cre mice that express Cre recombinase in postmitotic glutamatergic neurons of the CA1 hippocampus and layer V cerebral cortex [39,40] or Synapsyn I-Cre mice that drive the expression of Cre in general mature neurons [41]. Other Cre mice, such as Glial Fibrillary Acid Protein (GFAP)-Cre or GFAP-Cre ERT2 , will allow the recombination glial linage [42,43] involved in the homeostatic functions control in health and disease.

Tamoxifen Treatment
Actin-Cre ERT2 (Cre +/− ), Mkk4 flox/flox , Mkk7 flox/flox 2-month-old mice were used to obtain a double KO mouse Mkk4 and Mkk7 (Actin-Cre ERT2 (Cre +/− ), Mkk4 ∆/∆ Mkk7 ∆/∆ genotype). These mice have a Cre recombinase expressed under the Actin promoter and fused with human estrogen receptor that can be activated with tamoxifen as a selective estrogen receptor modulator. The CRE activation with tamoxifen allows controlling the specific time to delete floxed genes [45]. Tamoxifen was administered via orogastric gauge, at different doses and days, in order to evaluate the dose and time necessary to delete MKK4 and MKK7 proteins. The optimal dose found was 5 mg per day for two consecutive days. Tamoxifen (Sigma-Aldrich, Madrid, Spain) was dissolved in a solution containing 90% of sunflower oil and 10% ethanol. Actin-Cre ERT2 (Cre −/− ), Mkk4 flox/flox , Mkk7 flox/flox mice were used as controls and were treated with vehicle solution (90% of sunflower oil/10% ethanol) or tamoxifen for two consecutive days. After 3 days, the Actin-Cre ERT2 (Cre +/− ), Mkk4 ∆/∆ Mkk7 ∆/∆ genotype was detected. The protein elimination was screened in the heart, liver, and CNS at 3, 7, and 14 days after tamoxifen administration. The studies were done after 14 days of tamoxifen administration because beyond 16 days, the deterioration of treated animals was severe, since they showed signs of distress, weight loss, reduced activity, and after 20 days of treatment, there was high mortality (50%). Animals of each genotype used are shown in Table 1. Tamoxifen treatment is shown in Scheme 1.  Cre ERT2 (Cre +/− ) mice (see Supplementary Figure S1). Thus, hemizygous (1) Actin-Cre ERT2 (Cre +/− ), Mkk4 flox/flox , Mkk7 flox/flox , (2) Actin-Cre ERT2 (Cre −/− ), Mkk4 flox/flox , Mkk7 flox/flox , and (3) C57BL/6 (WT mice) were used in this study. All mice were housed in constant and controlled environments during the experiments with a light/dark cycle of 12 h. The mice had free access to food and water. The experiments were conducted in accordance with the Council of Europe Directive 2010/63. The procedure was registered and accepted by the Catalan Government Decree 214/97, 30 July 2020, the University of Barcelona, and the Animal Experimentation Ethics Committee.

Tamoxifen Treatment
Actin-Cre ERT2 (Cre +/− ), Mkk4 flox/flox , Mkk7 flox/flox 2-month-old mice were used to obtain a double KO mouse Mkk4 and Mkk7 (Actin-Cre ERT2 (Cre +/− ), Mkk4 ∆/∆ Mkk7 ∆/∆ genotype). These mice have a Cre recombinase expressed under the Actin promoter and fused with human estrogen receptor that can be activated with tamoxifen as a selective estrogen receptor modulator. The CRE activation with tamoxifen allows controlling the specific time to delete floxed genes [45]. Tamoxifen was administered via orogastric gauge, at different doses and days, in order to evaluate the dose and time necessary to delete MKK4 and MKK7 proteins. The optimal dose found was 5 mg per day for two consecutive days. Tamoxifen (Sigma-Aldrich, Madrid, Spain) was dissolved in a solution containing 90% of sunflower oil and 10% ethanol. Actin-Cre ERT2 (Cre −/− ), Mkk4 flox/flox , Mkk7 flox/flox mice were used as controls and were treated with vehicle solution (90% of sunflower oil/10% ethanol) or tamoxifen for two consecutive days. After 3 days, the Actin-Cre ERT2 (Cre +/− ), Mkk4 ∆/∆ Mkk7 ∆/∆ genotype was detected. The protein elimination was screened in the heart, liver, and CNS at 3, 7, and 14 days after tamoxifen administration. The studies were done after 14 days of tamoxifen administration because beyond 16 days, the deterioration of treated animals was severe, since they showed signs of distress, weight loss, reduced activity, and after 20 days of treatment, there was high mortality (50%). Animals of each genotype used are shown in Table 1. Tamoxifen treatment is shown in Scheme 1.

Data Analysis
Student's t-test was performed to compare two conditions, and one-way ANOVA post hoc Fisher's Least Significant Difference (LSD) tests were used for comparison in 3 or more conditions. Level of significance was fixed at α = 0.05. Both statistical analyses and graphs were created with the Graph Pad InStat software V5.0 (Graph Pad Software Inc., San Diego, CA, USA).

Conclusions
The new transgenic Actin-Cre ERT2 (Cre +/− ), Mkk4 flox/flox , Mkk7 flox/flox mice allow inducing the conditional deletion of Mkk4 and Mkk7 genes in adults, hence overcoming the lethality induced with other KOs. Thus, these types of mice would allow studying the specific functions of MKK4 and MKK7 proteins in adult organisms. Specifically, here, we identified the role that the MKK4/MKK7/JNK signaling pathway plays to control the positioning, morphology, and differentiation of the immature hippocampal subpopulation. This approach will make it possible to control adult pathways through the modulation of specific proteins that can be used as targets.