A Zdhhc22-mCherry BAC transgenic mouse line reveals neuronal expression dynamics in the developing and adult CNS

Abstract

The DHHC palmitoyltransferase Zdhhc22 is known to play a role in neuronal differentiation, synaptic regulation, and brain development. While transcriptomic data hint at region-specific expression, its exact spatiotemporal and cell-type distribution in the mammalian brain is unclear. For this purpose, we generated a bacterial artificial chromosome (BAC) transgenic mouse line that expresses the mCherry fluorescent reporter driven by the Zdhhc22 promoter. We then analyzed Zdhhc22 expression from embryonic day 13.5 (E13.5) through adulthood. mCherry fluorescence was detected in many brain regions, including the cortex, thalamus, midbrain, piriform cortex, and brainstem. Interestingly, a dynamic developmental gene expression pattern was observed: Zdhhc22 expression was initially restricted to the cortical marginal zone between E13.5 and E15.5, it then expanded into deeper cortical layers by E17.5, and at postnatal day 0 (P0), it persisted in deep layers while also appearing in a new subset of cortical plate neurons. Through co-immunostaining, mCherry expression was found to be predominantly neuronal, showing strong co-localization with NeuN and minimal overlap with glial cells. In the cortex, Zdhhc22 expression showed no co-localization with CUX1 or CTIP2 but did partially overlap with FOG2, a marker for layer VI pyramidal neurons. A particularly striking finding was that nearly all marginal zone mCherry-positive cells co-expressed RELN, identifying them as Cajal–Retzius cells. This neuronal specificity was maintained in the adult brain. Our findings validate the Zdhhc22-mCherry BAC transgenic line as a faithful model of endogenous Zdhhc22 expression, providing invaluable insight into its cellular specificity and a powerful new tool for future research.

Supplementary Information

The online version contains supplementary material available at 10.1007/s00429-026-03075-y.

Introduction

Protein palmitoylation, a reversible lipid modification, plays a crucial role in regulating the trafficking, stability, and subcellular localization of numerous neuronal proteins (Greaves and Chamberlain 2007). This process is catalyzed by a family of palmitoyl acyltransferases, all of which contain a conserved DHHC (Asp-His-His-Cys) motif. These enzymes have been identified as key regulators of nervous system development, neuronal differentiation, and synaptic plasticity (Roth et al. 2002; Fukata and Fukata 2010). Given that dysregulation of protein palmitoylation is implicated in several neurological disorders, including intellectual disability, epilepsy, and neurodegenerative diseases, the precise spatiotemporal control of DHHC enzyme activity is clearly of paramount importance (Fukata and Fukata 2010).

Zdhhc22 encodes a DHHC family member whose function remains largely uncharacterized. In contrast to other DHHC enzymes like Zdhhc5 and Zdhhc8, which have been extensively investigated for their roles in dendritic growth, receptor trafficking, and synapse formation (Brigidi et al. 2015; Collura et al. 2020; Liu et al. 2025) little is known about Zdhhc22. Initial molecular cloning studies identified ZDHHC22 as a Golgi-localized protein with predicted palmitoyltransferase activity (Ohno et al. 2006). Although bulk expression analyses and in situ hybridization techniques suggest that Zdhhc22 transcripts are enriched in specific regions of the central nervous system (CNS), including the cerebral cortex, hippocampus, midbrain, pons, and medulla (Globa and Bamji 2017), the cell-type specificity of its expression and its developmental regulation remain largely unknown.

Bioinformatics and functional studies have recently underscored the potential significance of Zdhhc22. For instance, Kim et al. (2018) showed that ZDHHC22 regulates the palmitoylation and secretion of the matricellular protein CCN3 in Neuro2a cells. Additionally, large-scale transcriptomic analyses have recently implicated ZDHHC22 in neurodegenerative disease pathways, such as Alzheimer’s disease, suggesting a role in immune modulation and neuronal dysfunction (Mao et al. 2025). Despite these compelling clues, a detailed description of Zdhhc22 expression at the cellular and anatomical levels during brain development is lacking. Acquiring this knowledge is critical for a deeper understanding of how Zdhhc22 may contribute to neuronal maturation, circuit formation, and synaptic activity.

To directly address this knowledge gap, we created a bacterial artificial chromosome (BAC) transgenic mouse line in which the fluorescent reporter mCherry is expressed under the control of the Zdhhc22 promoter. This innovative strategy overcomes the limitations of traditional mRNA-based methods by allowing for the direct in situ visualization and precise mapping of Zdhhc22-expressing cells across various developmental stages. We analyzed expression from embryonic day 13.5 (E13.5) through adulthood, combining fluorescence microscopy with a panel of established neuronal and glial markers.

This methodological approach provides several advantages: (1) the inclusion of large genomic regulatory elements within the BAC promotes the faithful recapitulation of endogenous expression, (2) the use of a fluorescent reporter facilitates high-resolution mapping in intact brain tissue, and (3) co-immunostaining with cell-type-specific markers enables the definitive identification of the neuronal populations expressing Zdhhc22.

In this work, we present the generation and thorough characterization of a new Zdhhc22-mCherry BAC transgenic mouse line. Our analysis of its spatiotemporal expression pattern across brain regions and developmental stages provides a critical resource. We anticipate this model will serve as a valuable and reliable experimental tool for future studies aiming to define the cellular distribution and function of Zdhhc22 in the CNS, thereby advancing our understanding of its role in neuronal development.

Methods

Animals

Animal experiments were performed in full compliance with the ethical guidelines of the Institutional Animal Care and Use Committee of the Asan Institute for Life Sciences (protocol No. 2018-12-256). We utilized C57BL/6N mice (Orient Bio Co., Korea) for pronuclear injection and the collection of fertilized oocytes. ICR strain females were used as surrogate mothers for embryo transfer.

Generation and characterization of Zdhhc22-mCherry BAC transgenic mice

We constructed the targeting vector by first amplifying the mCherry cDNA via PCR with the mCherry-Fwd (5′-AATAGCTAGCATGTTGAGCAAGGGCGAG-3′) and mCherry-Rev (5′-GGACTAGTATGTTGAGCAAGGGCGAG-3′) primers. The resulting product was digested with NheI and SpeI before being subcloned into the NheI site of the PL452 vector, yielding PL452-mCherry. Next, the left and right homology arms of the Zdhhc22 locus were PCR-amplified employing the corresponding Forward (5′- AATAGCTAGCatgttgagcaagggcgagga-3′) and Reverse (5′-GGACTAGTctacttgtacagctcgtcc-3′) primers, digested with the appropriate restriction enzymes, and sequentially subcloned into the 5′ and 3′ regions of the PL452-mCherry vector to complete the targeting construct. The finished vector was excised with KpnI and NotI and introduced into the E. coli strain SW102, which contains the RP23-74L10 BAC clone, to perform homologous recombination.

After isolating recombined BAC DNA from targeted clones, we removed the Neo cassette using Cre recombinase. The purified recombinant Zdhhc22-mCherry BAC DNA was subsequently transformed into E. coli DH10b for amplification and large-scale preparation. To generate transgenic founder lines, the resulting BAC transgene was microinjected into the pronuclei of fertilized mouse oocytes.

Genotyping was performed by extracting genomic DNA from tail biopsies of newborn pups, followed by PCR using mCherry-specific primers to confirm the presence of the transgene.

Immunohistochemical analysis of Zdhhc22 expression

Immunohistochemical analysis was carried out following a previously published protocol with slight modifications (Yang et al. 2020). For embryonic and early postnatal analysis, E13.5–E17.5 and postnatal day 0 (P0) brains were rapidly dissected, fixed in 4% paraformaldehyde (PFA) overnight at 4 °C, and subsequently washed in phosphate-buffered saline before being embedded in 2% agarose gel. Adult brains (P56) were collected after deep anesthesia and transcardial perfusion with 4% PFA, followed by dissection. All brain tissues were then sectioned coronally at 70 µm on a vibratome (Leica, Wetzlar, Germany). Free-floating sections were sequentially incubated in a blocking buffer, primary antibodies (listed in Table 1), and appropriate fluorophore-conjugated secondary antibodies (Jackson ImmunoResearch Laboratories). Sections were counterstained with DAPI for nuclear visualization.

Table 1.

List of antibodies used in the study

Antibody	Cat #	RRID	Supplier	Dilution
Rabbit anti-mCherry	Ab167453	AB_2571870	Abcam	1:1000
Rat anti-mCherry	M11217	AB_2536611	Thermo Fisher Scientific	1:1000
Rabbit anti-FOXP4	ABE74	AB_10617521	Millipore	1:1000
Rabbit anti-PAX2	71–6000	AB_2533990	Thermo Fisher Scientific	1:500
Rabbit anti-PAX6	Ab2237	AB_1587367	Millipore	1:500
Rabbit anti-S100	Ab868	AB_306716	Abcam	1:1000
Rabbit anti-OLIG2	Ab9610	AB_570666	Sigma-Aldrich	1:500
Mouse anti-RELN	MAB5364	AB_2179313	Millipore	1:200
Rabbit anti-CUX1	Sc-13024	AB_2261231	Santa Cruz Biotechnology	1:250
Rat anti-CTIP2	Ab18465	AB_2064130	Abcam	1:250
Rabbit anti-FOG2	Sc-10755	AB_2218978	Santa Cruz Biotechnology	1:500
Mouse anti-GAD67	MAB5406	AB_2278725	Millipore	1:200
Rabbit anti-PSD95	APZ-009	AB_2341060	Alomone Labs	1:100

Confocal image acquisition

Confocal images were acquired using a C2 Confocal microscope (Nikon, Tokyo, Japan) equipped with 10x, 20x, and 60 × objectives. For whole-section overviews (10x), shallow Z-stacks were collected (2.8 µm step; 5–10 slices). For higher-magnification imaging (20 × and 60x), Z-stacks were acquired with step sizes of 0.4–0.8 µm for 20 × images (10–25 slices) and 0.1 µm for 60 × images (40–60 slices). Maximum-intensity projections are shown unless otherwise noted, and XZ/YZ orthogonal views were generated by reslicing the corresponding Z-stacks when presented.

Results

Spatiotemporal expression of Zdhhc22-mCherry during brain development

To investigate the spatiotemporal expression of Zdhhc22 during CNS development, we generated a BAC transgene. In this construct, an mCherry reporter was inserted into a BAC clone containing the complete Zdhhc22 genomic locus (Fig. 1A). We then established transgenic mice with this construct and examined reporter expression from embryonic day 13.5 (E13.5) to postnatal day 0 (P0).

Fig. 1.

Zdhhc22-mCherry BAC construct and developmental expression pattern. A The Zdhhc22-mCherry BAC construct. The mCherry reporter cassette was inserted into a BAC clone containing the full Zdhhc22 genomic locus. B Whole-brain sagittal sections show the developmental progression of Zdhhc22-mCherry expression at E13.5, E15.5, E17.5, and P0. At E13.5, mCherry fluorescence is largely restricted to the marginal zone (MZ). By E15.5, signals become more prominent and remain localized to the superficial cortical domain. At E17.5, expression extends into deeper cortical regions. By P0, robust mCherry labeling is visible in the cortex (Ctx), hippocampus (Hp), thalamus (Th), olfactory bulb (OB), caudoputamen (CP), superior/inferior colliculus (SC/IC), cerebellum (CB), pons, and medulla (Me). Scale bars: 1 mm. B′ Higher-magnification cortical images illustrating laminar distribution across developmental stages. At E13.5, mCherry is confined to the MZ above the cortical plate (CP), intermediate zone (IZ), and SVZ/VZ. At E15.5, expression remains enriched in the MZ. By E17.5, signals extend into deeper developing layers. At P0, mCherry-positive cells occupy the MZ, upper layers (L2–4), L5, and L6. Scale bars: 100 μm

Our analysis revealed mCherry fluorescence in numerous brain regions, including the cortex, midbrain, cerebellum, thalamus, olfactory bulb, and brainstem (Fig. 1B). Early in development (E13.5 and E15.5), expression was tightly restricted to the marginal zone of the cortex. By E17.5, the mCherry signal had expanded into the deep cortical layers, and at P0, expression was maintained in these layers while also appearing in a new subset of cells in the cortical plate (Fig. 1B′). In the developing cerebellum at P0, mCherry was selectively expressed in the Purkinje cell layer (Supplementary Fig. 1).

These data establish a developmental timeline of Zdhhc22 expression that is difficult to infer from transcriptomic datasets alone.

Zdhhc22-mCherry is specifically expressed in cortical neurons and Cajal–Retzius cells

To define cell-type specificity, we performed co-immunostaining in P0 and P7 cortex. mCherry-positive cells in middle and deep layers robustly co-localized with NeuN, whereas upper-layer mCherry signal showed limited NeuN overlap at early stages (Fig. 2A,B). Co-staining with glial markers indicated minimal overlap overall: at P0, a small subset of mCherry-positive cells co-expressed S100 or OLIG2, but by P7 these double-positive cells were rare and largely restricted to superficial regions, with essentially none in middle/deep layers (Fig. 2A, B). mCherry did not co-localize with GAD67, indicating exclusion from interneuron populations (Supplementary Fig. 2).

Fig. 2.

Cell-type-specific Zdhhc22 expression in the developing cortex. A Co-immunostaining of P0 cortex with NeuN, S100, and OLIG2. mCherry-positive cells in the middle and deep cortical layers show strong co-localization with NeuN, confirming neuronal identity. A small number of mCherry-positive cells co-express S100 across the cortex, although most signals do not overlap. Occasional co-localization with OLIG2 is observed in both upper and deeper layers. High-magnification images of the upper, middle, and deep layers are shown below. Scale bars: 100 μm (top), 50 μm (bottom). B Co-immunostaining of P7 cortex. mCherry-positive neurons again co-localize with NeuN in middle and deep layers. Only a very small number of S100-positive glial cells overlap with mCherry, mainly in the upper layers. Rare OLIG2 co-localization is observed, with no overlap in the middle or deep layers. Representative high-magnification images across cortical depth are shown. Scale bars: 100 μm (top), 50 μm (bottom). (C, C′) At P7, nearly all mCherry-positive cells in the marginal zone (MZ) co-express RELN, identifying them as Cajal–Retzius cells. C′ shows a high-magnification view of MZ labeling and orthogonal XZ/YZ projections. Scale bars: 100 μm

A prominent feature was marginal zone expression. At P7, nearly all marginal zone mCherry-positive cells co-expressed RELN, identifying them as Cajal–Retzius cells (Fig. 2C, C′). We next assessed cortical pyramidal layer markers. mCherry-positive neurons did not co-localize with CUX1 or CTIP2 but partially overlapped with FOG2, consistent with enrichment in layer VI pyramidal neurons (Supplementary Fig. 3A). Together, these results demonstrate that Zdhhc22 expression during cortical development is strongly neuron-biased, prominently marking Cajal–Retzius cells and a subset of deep-layer pyramidal neurons.

Zdhhc22-mCherry expression is neuron-specific in the adult brain

To reconcile a discrepancy in expression patterns, where previous mouse transcriptomic data suggested low or absent adult cortical Zdhhc22 expression (Wild et al. 2022) but human data indicated its persistence (Kang et al. 2011), we performed co-immunostaining on P56 Zdhhc22-mCherry BAC transgenic mouse brains. Using the neuronal marker NeuN and the glial marker S100, we found that mCherry fluorescence was indeed present in adult regions such as the cerebral cortex, piriform cortex, and thalamus, where it robustly co-localized with NeuN (Fig. 3A, A′). Critically, no co-localization with S100 was observed (Fig. 3B, B′).

Fig. 3.

Zdhhc22 is expressed in adult neurons but not glia. (A, A′) P56 brain sections from Zdhhc22-mCherry BAC transgenic mice were co-immunostained with the neuronal marker NeuN. mCherry fluorescence was detected in multiple regions, including the cerebral cortex (Ctx), piriform cortex (PC), and thalamus (Th), showing robust co-localization with NeuN. (B, B′) Co-staining with the glial marker S100. This revealed the complete absence of overlap between mCherry and S100 signals in all regions studied. Scale bars: 1 mm (A, B); 100 μm (A’, B’); 50 μm (enlarged views)

To further characterize the laminar distribution of Zdhhc22-expressing neurons in the adult cortex, we examined co-expression with cortical layer markers (CUX1, CTIP2, and FOG2). Interestingly, unlike the developmental pattern observed at P7, mCherry-positive neurons were broadly distributed across cortical layers in adulthood (Supplementary Fig. 3B). To determine the neuronal subtype characteristics of these adult mCherry-expressing cells, we next assessed co-localization with the postsynaptic marker PSD95. Remarkably, mCherry-positive neurons in the adult cortex, piriform cortex, and thalamus showed extensive and near-complete co-localization with PSD95, supporting the postsynaptic neuronal identity of Zdhhc22-expressing cells (Supplementary Fig. 4).

These findings align with the human data, definitively showing that Zdhhc22 expression in the adult mouse brain is exclusively neuronal, with no expression in glial populations.

Discussion

In this work, we present a detailed characterization of a Zdhhc22-mCherry BAC transgenic mouse line as a new tool for investigating the spatiotemporal expression of Zdhhc22 in the CNS. Faithful recapitulation of endogenous expression by the mCherry reporter enabled precise mapping of Zdhhc22-positive cells across developmental stages. We show that Zdhhc22 expression is strongly biased toward neuronal populations, with minimal overlap with glial lineages. Furthermore, we identify a restricted expression pattern in deep-layer pyramidal neurons and Cajal–Retzius cells. Zdhhc22 expression also persists into adulthood, where it remains predominantly neuronal across multiple brain regions.

Single-cell RNA sequencing (scRNA-seq) has greatly expanded our understanding of cell-type-specific gene expression (Svensson et al. 2018). However, technical limitations such as dropout effects and reduced sensitivity for spatially restricted or low-abundance transcripts can lead to underestimation of certain genes. This may explain why previous scRNA-seq datasets reported little or no Zdhhc22 expression in the mouse cortex (Wild et al. 2022). In contrast, human bulk transcriptomic and in situ hybridization studies have consistently shown persistent ZDHHC22 expression throughout the lifespan (Kang et al. 2011). Our BAC-based reporter analysis provides in vivo anatomical evidence that helps reconcile these discrepancies.

The BAC-based reporter strategy offers clear advantages for faithfully reflecting endogenous gene regulation, as BAC constructs contain extensive native regulatory elements (Yang and Seed 2003). Using this approach, we show that Zdhhc22 expression is initially confined to the cortical marginal zone during early embryogenesis, expands into deep cortical layers by late embryonic stages, and is maintained postnatally, with additional expression in subsets of cortical plate neurons at birth. This work provides a refined developmental timeline for Zdhhc22 expression that was previously difficult to infer from transcriptomic datasets alone.

A notable finding of this study is the expression of Zdhhc22 in Cajal–Retzius cells, a transient neuronal population in the marginal zone that plays a critical role in cortical lamination through Reelin secretion (Soriano and Del Río 2005; Vílchez-Acosta et al. 2022). Our finding that Zdhhc22 co-expresses with Reelin indicates that Zdhhc22 is present in Cajal–Retzius cells and raises the possibility that palmitoylation-related pathways may be relevant in this cell population. However, whether Zdhhc22 directly regulates Reelin processing, secretion, or signaling remains to be determined by future functional studies.

The neuronal specificity of Zdhhc22 expression is further underscored by its laminar distribution within the cortex. Zdhhc22 expression was not detected in CUX1-positive superficial-layer pyramidal neurons and showed partial overlap with the layer VI pyramidal neuron marker FOG2. This pattern indicates that Zdhhc22 is preferentially expressed in a subset of deep-layer pyramidal neurons, including corticothalamic projection neurons, during cortical development. Nevertheless, the present study does not directly assess the functional role of Zdhhc22 in corticothalamic circuit formation or maturation. Although this expression pattern is consistent with reports implicating other DHHC palmitoyltransferases in synaptic organization and circuit assembly (Fukata and Fukata 2010; Tian et al. 2012), functional validation will require targeted experimental approaches.

Beyond development, Zdhhc22 expression persists in the adult brain and remains confined to neuronal populations. Dysregulation of palmitoylation has been implicated in neurodevelopmental and neurodegenerative disorders (Zaręba-Kozioł et al. 2018), and recent transcriptomic studies have identified ZDHHC22 as a potential disease-associated factor (Mao et al. 2025; Qian et al. 2025). The present findings provide an anatomical framework that may facilitate future investigations into how dysregulation of Zdhhc22 could contribute to neurological disease.

A limitation of this study is that it focuses on anatomical and cell-type-specific expression using a BAC reporter system. Functional roles of Zdhhc22 in neuronal development, synaptic regulation, or disease-associated processes were not directly examined and will require genetic, biochemical, and physiological analyses in future studies.

In summary, we establish the Zdhhc22-mCherry BAC transgenic mouse as a reliable model for visualizing Zdhhc22 expression in vivo. Our data demonstrate that Zdhhc22 expression is preferentially neuronal, marking Cajal–Retzius cells and a subset of deep-layer pyramidal neurons while being largely absent from glial lineages. This study provides a refined anatomical foundation for future functional investigations into the role of Zdhhc22 in brain development and disease.

Supplementary Information

Below is the link to the electronic supplementary material.

Author contributions

Conceptualization: S.S., S.J. Methodology: H.Y., J.R., H.J.B. Resources: K.N., J.C. Writing: S.S., H.Y., S.J., J.R.

Funding

This study was supported by the Korea Research Institute of Bioscience and Biotechnology (KRIBB) Research Initiative Program, KGM1012521, KGM1012521, KGM1012521, The Basic Science Research Program through the National Research Foundation of Korea (NRF), funded by the Ministry of Education, 2020R1A6A1A06046235, 2020R1A6A1A06046235.

Data availability

Data are available from the corresponding author upon reasonable request.

Declarations

Conflict of interest

The authors declare no competing interests.

Ethical approval

IACUC protocol No. 2018–12-256.