BRAP maintains SEC24D stability to regulate ER stress and promote cervical adenocarcinoma progression

Abstract

Background

Cervical adenocarcinoma (ADC) exhibits a poorer prognosis than cervical squamous cell carcinoma and lacks effective targeted therapies. Bombesin receptor- activated protein (BRAP, encoded by C6orf89) has been implicated in cell proliferation and stress responses in multiple biological contexts; however, its role in ADC progression and endoplasmic reticulum (ER) homeostasis remains unknown.

Methods

BRAP expression was evaluated in human cervical adenocarcinoma (ADC) tissues by immunohistochemistry and western blotting, and its clinicopathological significance was analyzed. BRAP function was investigated by siRNA-mediated knockdown in HeLa and SiHa cells using proliferation, apoptosis, migration, invasion, and epithelial–mesenchymal transition (EMT) assays, and further validated in xenograft models. Ultrastructural alterations were examined by transmission electron microscopy. BRAP-interacting proteins were identified by yeast two-hybrid screening and validated by co-immunoprecipitation. ER stress, ER exit site (ERES) homeostasis, and autophagic flux were evaluated, and rescue experiments were performed by re-expression of BRAP or SEC24D.

Results

BRAP was markedly upregulated in cervical adenocarcinoma tissues and was associated with poor tumor differentiation. BRAP depletion selectively inhibited proliferation, migration, invasion, and EMT while promoting apoptosis in ADC-derived HeLa cells, but not in squamous cell carcinoma-derived SiHa cells. Mechanistically, BRAP deficiency disrupted ER homeostasis and selectively activated the PERK–ATF4–DDIT3/CHOP branch of the unfolded protein response (UPR), whereas pharmacological inhibition with GSK2606414 largely reversed these phenotypic alterations. We identified the COPII cargo adaptor SEC24D as a novel BRAP-interacting protein. BRAP maintained SEC24D protein stability by preventing excessive autophagy–lysosome-mediated degradation, thereby preserving ERES number and function, as reflected by SEC16A-positive ERES and peripheral LMAN1-positive structures, and maintaining ER–Golgi trafficking homeostasis. Restoration of SEC24D rescued ERES defects, attenuated PERK-dependent UPR activation, and reversed the impaired proliferation, migration, invasion, apoptosis, and EMT induced by BRAP depletion.

Conclusion

BRAP promotes cervical adenocarcinoma progression by maintaining SEC24D stability and preserving ERES organization and function, thereby sustaining COPII-dependent ER export and ER homeostasis. Disruption of the BRAP–SEC24D axis impairs ERES integrity, activates PERK-dependent ER stress, and suppresses malignant phenotypes. These findings identify BRAP as a previously unrecognized regulator of ER homeostasis and establish the BRAP–SEC24D signaling axis as a potential therapeutic target for cervical adenocarcinoma.

Supplementary information

The online version contains supplementary material available at https://doi.org/10.1186/s12967-026-08780-0.

Introduction

Cervical cancer (CC) is the fourth most common malignancy among women worldwide, accounting for approximately 604,000 new cases and 342,000 deaths annually [1, 2]. Although widespread implementation of human papillomavirus (HPV) vaccination and cervical screening programs has substantially reduced the overall incidence of cervical cancer, the proportion of cervical adenocarcinoma (ADC) has continued to increase, particularly among women younger than 35 years of age [3]. ADC arises from the endocervical glandular epithelium and currently accounts for approximately 20–25% of all cervical cancers. Compared with cervical squamous cell carcinoma (SCC), ADC exhibits distinct clinicopathological characteristics, molecular alterations, and therapeutic responses [4, 5]. Importantly, ADC is less sensitive to conventional treatment modalities, including surgery, radiotherapy, and platinum-based chemotherapy, and is generally associated with poorer clinical outcomes than SCC [6]. Despite its increasing clinical burden, the molecular mechanisms driving ADC progression remain incompletely understood, and no effective ADC-specific therapeutic targets have yet been established.

Endoplasmic reticulum (ER) homeostasis is essential for maintaining protein folding, processing, and secretion in highly secretory epithelial cells. Disruption of ER homeostasis leads to the accumulation of unfolded or misfolded proteins within the ER lumen, a condition known as ER stress [7]. To restore proteostasis, cells activate the unfolded protein response (UPR), which is mediated by three canonical signaling branches initiated by PERK, IRE1α, and ATF6 [8]. While transient UPR activation promotes cellular adaptation and survival, persistent or excessive activation—particularly of the PERK–ATF4–DDIT3/CHOP pathway—triggers apoptosis [9, 10]. Beyond its role in maintaining ER homeostasis, accumulating evidence indicates that dysregulated UPR signaling contributes to tumor progression by modulating cellular processes such as proliferation, epithelial–mesenchymal transition (EMT), and apoptosis in a context-dependent manner [11–13]. Among the three UPR branches, PERK signaling has emerged as a central regulator of cell fate determination and has been implicated in cancer progression through its effects on both apoptosis and EMT [12].

Maintenance of ER homeostasis depends not only on adaptive UPR signaling but also on efficient protein export from the ER. Newly synthesized proteins are transported from the ER to the Golgi through the secretory pathway, a process initiated at specialized ER exit sites (ERES), where COPII-coated vesicles are assembleds [14, 15]. Among the COPII components, SEC24D functions as a major cargo adaptor responsible for selective cargo recognition and efficient protein export from the ER [16]. Recent studies have further demonstrated that ER quality control is tightly integrated with autophagy, in which components of the COPII machinery cooperate with autophagy receptors to regulate ER turnover and maintain secretory pathway homeostasis [17]. However, the upstream mechanisms coordinating SEC24D-mediated ER export, ER homeostasis, and malignant progression remain largely unknown.

Bombesin receptor–activated protein (BRAP), encoded by the human C6orf89 gene, is expressed in bronchial epithelial cells, neurons, macrophages, endothelial cells, and renal tubular epithelial cells [18–22]. Previous studies have implicated Bombesin receptor–activated protein in cell proliferation, oxidative stress regulation, and transcriptional control [18, 23, 24]. Its mouse homologue promotes renal fibrosis by enhancing EMT in tubular epithelial cells [20]. Preliminary observations suggest that BRAP is abnormally upregulated in several human malignancies, including cervical adenocarcinoma. In addition, transmission electron microscopy of Hela cells revealed that BRAP knockdown induces marked swelling and dilation of the endoplasmic reticulum, which is a characteristic feature of endoplasmic reticulum stress [25]. The molecular mechanisms underlying these changes remain unknown.

To explore the molecular function of BRAP, we previously performed a yeast two-hybrid (Y2H) screen and identified several proteins involved in ER-to-Golgi trafficking as potential BRAP-interacting candidates, including SEC23B, SEC24B, SEC24C, SEC24D, and TSG101 [19]. Subsequent validation demonstrated that only SEC24D specifically interacted with the C-terminal region of BRAP, suggesting a functional association between BRAP and SEC24D.

However, whether BRAP regulates ER homeostasis through SEC24D-dependent ER exit site organization and ER-to-Golgi trafficking, and whether this mechanism contributes to cervical adenocarcinoma progression, remains unknown.

In the present study, we investigated the biological function of BRAP in cervical adenocarcinoma and explored its role in regulating ER homeostasis and tumor progression. We demonstrate that BRAP interacts with SEC24D to maintain ER exit site integrity. Loss of BRAP promotes autophagic degradation of SEC24D, disrupts ER exit site organization, activates the PERK branch of the UPR, and consequently promotes apoptosis while suppressing EMT. These findings identify the BRAP–SEC24D axis as a previously unrecognized mechanism linking ER homeostasis to cervical adenocarcinoma progression and suggest a potential therapeutic vulnerability for this aggressive histological subtype.

Method

Mammalian cell culture and transfection

HeLa (human cervical adenocarcinoma), SiHa (human cervical squamous cell carcinoma), HT-3 (human cervical carcinoma), and HaCat (immortalized human keratinocyte) cell lines were used in this study. All cell lines were obtained from the Cancer Research Institute, School of Basic Medical Sciences, Central South University (Changsha, China). Cells were cultured in Dulbecco’s modified Eagle’s medium (DMEM; Gibco, Carlsbad, CA, USA) supplemented with 10% fetal bovine serum (FBS), 100 U/mL penicillin, and 100 μg/mL streptomycin at 37 °C in a humidified incubator containing 5% CO₂.

Cells were seeded into 6-well plates at a density of 2.5 × 10⁵ cells per well and cultured overnight to reach approximately 70–80% confluence before transfection. The pcDNA3.1-SEC24D overexpression plasmid was generated by homologous recombination. Two independent siRNAs targeting C6orf89 (Santa Cruz Biotechnology, Dallas, TX, USA) were used: sc-95381A (sense, 5′-GGAUGUCGUUGACUCAGAUTT-3′; antisense, 5′-AUCUGAGUCAACGACAUCCTT-3′) and sc-95381B (sense, 5′-CAGUCACUUUCGUAGUGAATT-3′; antisense, 5′-UUCACUACGAAAGUGACUGTT-3′).

Transient transfection was performed using Lipofectamine™ 3000 (Invitrogen, Carlsbad, CA, USA) according to the manufacturer’s instructions. For each well of a 6-well plate, 2.5 μg of pcDNA3.1-SEC24D plasmid (or empty vector) and/or 100 pmol siRNA were transfected using Opti-MEM™ and P3000 reagent. For co-transfection experiments, plasmids and siRNAs were mixed before complex formation with the transfection reagent. The culture medium was replaced with fresh complete medium 6 h after transfection, and cells were harvested 48 h after transfection for subsequent experiments.

Patient selection and follow-up

A total of 61 treatment-naïve patients with histologically confirmed cervical adenocarcinoma who underwent surgical resection at The Second Xiangya Hospital, Central South University, between February 2020 and February 2023 were enrolled in this study. Formalin-fixed, paraffin-embedded (FFPE) tumor specimens were obtained from the archives of the Department of Pathology, The Second Xiangya Hospital, Central South University. The study was approved by the Institutional Ethics Committee of the School of Basic Medical Sciences, Central South University, Changsha, China (Approval No. 2022-KT118), and was conducted in accordance with the Declaration of Helsinki. Written informed consent was obtained from all participants. All tumors were classified according to the International Endocervical Adenocarcinoma Criteria and Classification (IECC) [26]. Patients with neuroendocrine carcinoma, adenosquamous carcinoma, or a history of neoadjuvant chemotherapy or radiotherapy were excluded. The histopathological diagnosis was independently confirmed by at least two experienced pathologists. Clinicopathological information, including age, histological differentiation, and International Federation of Gynecology and Obstetrics (FIGO) stage, was retrieved from the corresponding medical records. The median age of the patients was 49 years (range, 30–64 years). Detailed clinicopathological characteristics are summarized in Table 1.

Table 1.

Clinical characteristics of cervical adenocarcinoma patients

Characteristics	N	%
Total patients
Age (median 49 years, range 30–64 years)
≤49	31	50.8
＞49	30	49.2
Differentiation
Well-differentiated	18	29.5
Moderately differentiated	20	32.8
Poorly differentiated	23	37.7
Stage
I	39	63.9
II	18	29.5
III	4	6.6
BRAP expression-H-Score
Negative	25	41
Positive	36	59

In the present study, a total of 61 patients with cervical adenocarcinoma were included. The median age was 49 years (range, 30–64 years), with slightly more than half of the patients (50.8%) aged ≤ 49 years. Histological examination showed that 29.5%, 32.8%, and 37.7% of tumors were well, moderately, and poorly differentiated, respectively. According to the FIGO staging system, the majority of patients (63.9%) had stage I disease, followed by stage II (29.5%) and stage III (6.6%). Based on immunohistochemical H-score evaluation [27], BRAP expression was positive in 59.0% (36/61) of tumors and negative in 41.0% (25/61) of tumors (Table 1).

Immunohistochemistry and IHC scoring

Formalin-fixed, paraffin-embedded tissue sections (4 μm) were subjected to immunohistochemical staining using an immunohistochemistry detection kit (cat. no. DD23, Tongling, China). After deparaffinization, antigen retrieval, and blocking of endogenous peroxidase activity, sections were incubated overnight at 4 °C with primary antibodies, followed by incubation with the corresponding secondary antibody according to the manufacturer’s instructions. Sections were counterstained with hematoxylin and examined under a light microscope. BRAP immunoreactivity was evaluated independently by two experienced pathologists who were blinded to the clinicopathological information. Staining intensity was scored as follows: 1, negative; 2, light yellow; 3, deep yellow; and 4, yellow-brown. The percentage of positive tumor cells was scored as follows: 1, 0–25%; 2, 25–50%; 3, 50–75%; and 4, 75–100%. The final immunohistochemical (IHC) score was calculated by multiplying the intensity and proportion scores (range, 1–16), as previously described [27].

Construction of BRAP-knockdown HeLa stable cell line

Stable BRAP-knockdown HeLa cells were generated by lentiviral transduction using a short hairpin RNA (shRNA) targeting C6orf89 (shC6orf89), with a non-targeting shRNA (shCtrl) serving as the negative control. Lentiviral particles were produced in HEK293T cells by co-transfecting the shRNA expression vector with the packaging plasmids pMDLg/pRRE, pRSV-Rev, and pVSV-G. HeLa cells were seeded in 24-well plates at a density of 3–5 × 10⁴ cells per well and infected the following day at a multiplicity of infection (MOI) of 10 in the presence of 5 μg/mL polybrene. The culture medium was replaced with fresh complete medium after 16–20 h. Puromycin selection (2 μg/mL) was initiated 72 h after infection and maintained for 2 weeks to establish stable polyclonal cell lines. Cells were subsequently maintained in complete DMEM containing 1 μg/mL puromycin. Knockdown efficiency was verified by western blot analysis. The shRNA sequence targeting C6orf89 was 5′-CCGAAGACATTGTCAGTCTGT-3′ (TRCN0000122768).

Transient transfection was performed as described above using Lipofectamine™ 3000. Stable BRAP-knockdown HeLa cells were transfected with pcDNA3.1-BRAP or the empty vector and harvested 48 h later for subsequent experiments.

CCK-8 cell viability assay

Cell viability was assessed using the Cell Counting Kit-8 (CCK-8; cat. no. C0009S, Beyotime, Shanghai, China). Cells were seeded into 96-well plates at a density of 3 × 10³ cells per well and cultured overnight before transient transfection with the indicated siRNAs or plasmids using Lipofectamine 3000 (Thermo Fisher Scientific) according to the manufacturer’s instructions. Five independent biological replicates were performed, with three technical replicates for each condition. At the indicated time points after transfection, 10 μL of CCK-8 reagent was added to each well and incubated for 2 h at 37 °C. Absorbance was measured at 450 nm using an EnSpire multimode plate reader (PerkinElmer), and cell viability was expressed as the relative absorbance normalized to the corresponding control group.

Wound healing assay

To evaluate the migratory capacity of HeLa and SiHa cells, a wound healing (scratch) assay was performed. Cells were seeded in 6-well plates and transfected with either C6orf89-specific siRNA or the pcDNA3.1-SEC24D plasmid as described above. Forty-eight hours after transfection, when the cells reached approximately 90–100% confluence, cells were treated with mitomycin C (10 μg/mL) for 2 h to inhibit cell proliferation. A linear wound was then generated using a sterile 10-μL pipette tip. Detached cells were removed by washing three times with Phosphate Buffered Saline (PBS), and the cells were subsequently cultured in serum-free medium. Images of the wounded area were captured immediately after scratching (0 h) and again at 48 h using an inverted microscope. Wound widths were measured at both time points, and the migration index was calculated as follows: Migration index (%) = [(W₀ − W₄₈)/W₀] × 100, where W₀ and W₄₈ represent the wound widths at 0 h and 48 h, respectively, as previously described [28]. At least five randomly selected fields were analyzed for each well using ImageJ software. Three independent biological replicates were performed.

Colony formation assay

The clonogenic potential of cells was evaluated using a colony formation assay. Cells were trypsinized into single-cell suspensions and seeded into 6-well plates at a density of 1,000 cells per well. After overnight attachment, cells were transfected with either C6orf89-specific siRNA or negative control siRNA as described above. Forty-eight hours after transfection, the culture medium was replaced with fresh complete medium, and the cells were cultured for an additional 12 days with medium changes every 3 days. At the end of the 14-day culture period, colonies were washed twice with PBS, fixed with 4% paraformaldehyde for 10 min, and stained with 0.1% crystal violet for 10 min at room temperature. After air-drying, colony images were captured using a digital camera. Colonies containing more than 50 cells were counted using ImageJ software. Three independent biological experiments were performed.

Transwell migration and invasion assays

Cell migration and invasion were evaluated using 24-well Transwell chambers with 8.0-μm pore polycarbonate membranes (Corning, NY, USA). For the invasion assay, the upper surface of the membranes was coated with Matrigel (100 μg/mL; BD Biosciences) and incubated at 37 °C for 1 h to allow gelation, whereas membranes used for the migration assay were left uncoated. Forty-eight hours after transfection, cells were serum-starved for 12 h, harvested, and resuspended in serum-free DMEM at a density of 1 × 10⁵ cells/mL. A total of 1 × 10⁴ cells in 100 μL of serum-free medium was added to the upper chamber, while 600 μL of complete medium containing 10% fetal bovine serum (FBS) was added to the lower chamber as a chemoattractant. Cells were incubated for 24 h (migration assay) or 48 h (invasion assay) at 37 °C in a humidified atmosphere containing 5% CO₂. After incubation, cells remaining on the upper surface of the membrane were gently removed with a cotton swab. Cells that had migrated or invaded to the lower surface were fixed with 4% paraformaldehyde for 30 min and stained with 0.1% crystal violet for 10 min at room temperature. Images were captured using an inverted microscope (Olympus, Tokyo, Japan). For each well, five randomly selected microscopic fields (200× magnification) were counted, and the average number of migrated or invaded cells was calculated. Three independent biological experiments were performed.

In vivo xenografts

All animal experiments were approved by the Institutional Animal Care and Research Advisory Committee of Central South University (Approval No. XMSB-2022-0003) and performed in accordance with institutional guidelines. Four-week-old male BALB/c nude mice were randomly assigned to the indicated experimental groups (n = 5 per group). Each mouse was subcutaneously injected with 3 × 10⁶ cells. Tumor length (L) and width (W) were measured every 2 days using a Vernier caliper by an investigator blinded to the group allocation, and tumor volume was calculated as (L × W²)/2. Twenty-one days after inoculation, mice were euthanized, and tumor tissues were excised for subsequent analyses.

For longitudinal tumor growth analysis, tumor volumes from mice that developed measurable tumors were analyzed using two-way repeated-measures ANOVA in GraphPad Prism, with time as the repeated-measures factor and experimental group as the between-group factor, followed by Šídák’s multiple-comparisons test. Endpoint tumor weights and final tumor volumes were analyzed using an unpaired two-tailed Student’s t-test. Animals that failed to develop measurable tumors were excluded from tumor volume and weight analyses.

EdU proliferation assay

Cell proliferation was evaluated using a 5-Ethynyl-2’-deoxyuridine (EdU) Cell Proliferation Kit (cat. no. C0078S, Beyotime, Shanghai, China) according to the manufacturer’s instructions. Forty-eight hours after infection, cells were incubated with EdU working solution for 2 h at 37 °C, fixed with 4% paraformaldehyde for 15 min, and permeabilized with permeabilization buffer (cat. no. P0097, Beyotime) for 10 min at room temperature. After washing with PBS, the Click reaction mixture was added and incubated for 30 min at room temperature according to the manufacturer’s instructions. Fluorescence images were captured using a fluorescence microscope (Ti-S, Nikon, Japan). For each well, at least five randomly selected microscopic fields were analyzed, and the percentage of EdU-positive cells was calculated. Three independent biological experiments were performed.

Cell cycle analysis

Cell cycle analysis was performed using a Cell Cycle Analysis Kit (cat. no. C1052, Beyotime, Shanghai, China) according to the manufacturer’s instructions. Cells were seeded into 6-well plates at a density of 1 × 10⁵ cells per well and cultured overnight in complete DMEM supplemented with 10% fetal bovine serum (FBS). Forty-eight hours after transfection, cells were harvested by trypsinization, washed twice with PBS, and fixed in 70% ice-cold ethanol at 4 °C for 24 h. The fixed cells were then washed with PBS, incubated with DNase-free RNase A (50 μg/mL), and stained with propidium iodide (PI) for 30 min at room temperature in the dark. DNA content was analyzed using a Cytek® Northern Lights™ flow cytometer (cat. no. N7-00008-0A; Cytek Biosciences, China), and cell cycle distribution was determined using the manufacturer’s analysis software. All experiments were performed with three independent biological replicates.

Quantitative real-time PCR (RT-qPCR)

Total RNA was extracted from cells using RNAiso Plus (cat. no. 9108, Takara, Japan) according to the manufacturer’s instructions as described previously [19]. 1 μg of total RNA was reverse-transcribed into complementary DNA (cDNA) using the PrimeScript^TM RT Reagent Kit with gDNA Eraser (cat. no. RR047B, Takara). Quantitative real-time PCR was performed using SYBR Premix Ex Taq II^TM (cat. no. RR420A, Takara) on a CFX96 Touch Real-Time PCR Detection System (Bio-Rad). All reactions were performed in triplicate. Relative mRNA expression levels were calculated using the 2^−ΔΔCt method, with human β-actin as the internal reference gene. Primer sequences are listed in Table 2.

Table 2.

Primer sequences for RT-qPCR

Primer	Forward (5’-3’)	Reverse (5’-3’)
β-actin	TTCCAGCCTTCCTTCCTGGG	TTGCGCTCAGGAGGAGCAAT
C6orf89	ATCTTGCTCACTGCCTAC	GCTCTGACTCATTCTGCT
SLUG	CAACGCCTCCAAAAAGCCAA	ACTCACTCGCCCCAAAGATG
ZEB1	AGAGCGCTAGCTGCCAATAA	GGGCGGTGTAGAATCAGAGT
ZEB2	TACACGTTTGCCTACCGCAC	ACTTGCGATTACCTGCTCCT
TWIST1	TCAAGAGGTCGTGCCAATCA	ATGGTTTTGCAGGCCAGTTT
SEC24D	CACCTGTCAACAATGTGGCA	GCAAAATGGAAGGCTGTGGA

Apoptosis assay

Apoptosis was assessed using an Annexin V-FITC Apoptosis Detection Kit (cat. no. 556547, BD Pharmingen) according to the manufacturer’s instructions. Forty-eight hours after transfection, the culture supernatant containing floating cells was collected. Adherent cells were then detached by trypsinization without EDTA and combined with the corresponding floating cells. The collected cells were washed twice with cold PBS and resuspended in 100 μL of Annexin V binding buffer. The cells were then incubated with Annexin V-FITC and propidium iodide (PI) for 15 min at room temperature in the dark. Apoptotic cells were analyzed using a Cytek® Northern Lights™ flow cytometer (cat. no. N7-00008-0A; Cytek Biosciences, China). All experiments were performed with three independent biological replicates. Annexin V binds to phosphatidylserine residues exposed on the outer cell membrane during early apoptosis, whereas PI penetrates cells with compromised membrane integrity and intercalates into DNA, labeling late apoptotic or necrotic populations. Based on this differential staining, cells were classified as viable (Annexin V⁻ /PI⁻), early apoptotic (Annexin V⁺ /PI⁻), or late apoptotic/necrotic (Annexin V⁺ /PI⁺).

Transmission electron microscopy (TEM)

Approximately 5 × 10⁶ cells were collected by centrifugation, washed three times with ice-cold PBS, and fixed in electron microscopy fixative. The samples were subsequently processed for transmission electron microscopy (TEM) by Wuhan Saville Biotechnology according to standard protocols. Representative ultrastructural images were acquired using a Hitachi HT7800 transmission electron microscope. Representative images from three independent experiments are shown.

Immunostaining and confocal microscopy

Glass coverslips were placed in 24-well plates, sterilized with 75% ethanol for 20 min, rinsed with sterile water, and air-dried. HeLa cells were seeded onto the coverslips and cultured to approximately 50% confluence. Cells were fixed with ice-cold methanol for 5 min, permeabilized, washed three times with PBS, and blocked with 5% bovine serum albumin (BSA) for 1 h at room temperature. Cells were then incubated overnight at 4 °C with the indicated primary antibodies, including anti-BRAP (Abcam, cat. no. ab181073; 1:100), anti-LMAN1 (HUABIO, cat. no. CSB-PA012991LA01HU; 1:200), anti-SEC16A (Proteintech, cat. no. 20025-1-AP; 1:200), anti-CKAP4 (Abcam, cat. no. ab302539; 1:200), anti-SEC24D (Proteintech, cat. no. 67409-1-Ig; 1:200), and anti-GM130 (ZENBIO, cat. no. R380777; 1:200). After washing with PBS, cells were incubated with the corresponding Alexa Fluor 488- or Alexa Fluor 594-conjugated secondary antibodies (Abcam; 1:200) for 1 h at room temperature in the dark. After mounting with an antifade mounting medium containing DAPI, fluorescence images were acquired using a laser scanning confocal microscope. Fluorescence intensity or puncta number was quantified using ImageJ software from at least five randomly selected microscopic fields for each experimental group.

Construction of yeast two-hybrid plasmids

Yeast two-hybrid plasmids were constructed using the Uniclone One Step Seamless Cloning Kit (cat. no. SC612, Beijing Genesand Biotech, Beijing, China) according to the manufacturer’s instructions. Briefly, plasmid backbones were linearized by restriction enzyme digestion and purified by agarose gel electrophoresis. Target DNA fragments were amplified by PCR using primers containing 5′ vector-homologous sequences and purified before assembly. Homologous recombination reactions were performed at 50 °C for 15 min, and the resulting constructs were transformed into chemically competent Escherichia coli cells. Positive colonies were selected on appropriate selective media, and recombinant plasmids were verified by Sanger sequencing.

To generate prey plasmids for interaction analysis with the previously constructed pGBKT7-BRAP-CTD bait plasmid [19], the pGADT7 vector was digested with EcoRI/XhoI for SEC23B, SEC24D, and TSG101, BglII/XhoI for SEC24B and SEC24C. To construct BRAP truncation plasmids, the pGBKT7 vector was digested with EcoRI and BamHI. All primer sequences used for plasmid construction are listed in Table 3.

Table 3.

Primers used for plasmids expressing the candidate genes

Primers	Sequences
SEC23B	Forward: 5’-GCCATGGAGGCCAGTGAATTCATGGCGACATACCTGGAGTTCA-3’ Reverse: 5’-ACGATTCATCTGCAGCTCGAGTTAACAGGCACTGGAGACAGCC-3’
SEC24B	Forward: 5’-CAAACCCAAAAAAAGAGATCTATGTCGGCCCCCGCCGGG-3’ Reverse: 5’-ACGATTCATCTGCAGCTCGAGTCACTTACAAATCTGCTGCTGAACA-3’
SEC24C	Forward: 5’-TCCACCCGGGTGGGCATCGATATGAACGTCAACCAGTCAGTTCC-3’ Reverse: 5’-ACGATTCATCTGCAGCTCGAGTTAGCTCAGTAGCTGCCGAATCT-3’
SEC24D	Forward: 5’-GCCATGGAGGCCAGTGAATTCATGAGTCAACAAGGTTACGTGGC-3’ Reverse: 5’-ACG ATTCATCTGCAGCTCGAGTTAATTAAGCAGCTGACAGATCTCCT-3’
TSG101	Forward: 5’-GCCATGGAGGCCAGTGAATTCATGGCGGTGTCGGAGAGC-3’ Reverse: 5’-ACGATTCATCTGCAGCTCGAGTCAGTAGAGGTCACTGAGACCGG-3’
BRAP	Forward:5’-CTGATCTCAGAGGAGGACCTGCATATGTTTGTGATTCAACCTTTCAGCC-3’ Reverse:5’-TGCGGCCGCTGCAGGTCGGGATCCCTGCTCACAGTCGTTTGTCCACCAG-3’

Yeast two-hybrid assay

Yeast two-hybrid assays were performed according to our previous study [19]. Full-length coding sequences of candidate proteins were cloned into the pGADT7 vector to generate GAL4 activation domain (AD) fusion constructs. The resulting plasmids were transformed into Saccharomyces cerevisiae strain Y187 and mated with Y2HGold cells carrying the pGBKT7-BRAP-CTD bait plasmid. Diploid yeast cells were selected on SD/-Leu/-Trp medium and incubated at 30 °C for 3 days, followed by transfer to SD/-Ade/-His/-Leu/-Trp medium supplemented with 40 μg/mL X-α-Gal for 3–5 days. Protein-protein interactions were considered positive when colonies grew on selective medium and developed blue coloration.

Co-immunoprecipitation (Co-IP)

HeLa cells were cultured in Dulbecco’s modified Eagle’s medium (DMEM) supplemented with 10% fetal bovine serum (FBS) at 37 °C in a humidified atmosphere containing 5% CO₂. Approximately 1 × 10⁷ cells were lysed in 500 μL of ice-cold IP lysis buffer (Beyotime, cat. no. P0013). After centrifugation at 12,000 × g for 20 min at 4 °C, 200 μL of the clarified lysate was incubated overnight at 4 °C with 1 μg of anti-SEC24D antibody (Proteintech, cat. no. 67409-1-Ig) or 1 μg of normal rabbit IgG (Santa Cruz Biotechnology, cat. no. sc-2025) together with 25 μL of Dynabeads™ Protein A (Invitrogen, cat. no. 10001D) under gentle rotation. The beads were washed three times with ice-cold NP-40 buffer as previously described [19]. Bound proteins were eluted by boiling the beads in 1× SDS loading buffer for 5 min and analyzed by western blotting using anti-SEC24D (HUABIO, cat. no. HA721902; 1:1000) and anti-BRAP (Abcam, cat. no. ab181073; 1:1000) antibodies.

Protein–protein docking analysis

Predicted three-dimensional structures of BRAP and SEC24D were obtained from the AlphaFold Protein Structure Database. Protein–protein docking was performed using the GRAMM-X docking server. The docking models were visualized and analyzed using PyMOL (version 2.5) [29]. Protein interaction interfaces were further analyzed using PDBePISA to evaluate the interface area and identify potential intermolecular interactions, including hydrogen bonds, salt bridges, and interacting residues [30].

Western blotting

Total cellular proteins were extracted using lysis buffer supplemented with protease inhibitor cocktail, phosphatase inhibitor cocktail, DTT, and PMSF. Protein concentrations were determined using the BCA Protein Assay Kit. Equal amounts of protein (20 μg) were separated by 10% SDS-PAGE and transferred onto PVDF membranes (Thermo Fisher Scientific, cat. no. 88520). Membranes were blocked with 5% non-fat milk for 1 h at room temperature and incubated overnight at 4 °C with the indicated primary antibodies. After washing three times with TBST, membranes were incubated with HRP-conjugated secondary antibodies for 1 h at room temperature. Protein bands were visualized using an enhanced chemiluminescence (ECL) detection system.

Primary antibodies included anti-β-actin (Santa Cruz Biotechnology, cat. no. sc-47778), anti-GAPDH (Proteintech, cat. no. 60004-1-Ig), anti-BRAP (Abcam, cat. no. ab181073), anti-Cleaved-caspase3 (Proteintech, cat. no. 19677-1), anti-N-cadherin (Proteintech, cat. no. 10366-1-AP), anti-E-cadherin (Bioss, cat. no. bs-10009R), anti-Caspase-3 (ZENBIO, cat. no. R23315), anti-SEC24D (Proteintech, cat. no. 67409-1-Ig), anti-ATF4 (Cell Signaling Technology, cat. no. 11815), anti-ATF6 (ZENBIO, cat. no. R26445), anti-PERK (ZENBIO, cat. no. R25331), anti-phospho-PERK (Thr982; ZENBIO, cat. no. 340846), anti-DDIT3/CHOP (ZENBIO, cat. no. 381679), anti-IRE1α (ZENBIO, cat. no. 680345), anti-phospho-IRE1α (Ser724; ZENBIO, cat. no. R26310), anti-CDK6(Proteintech, cat. no. 14052-1-AP), anti-CyclinD1(Proteintech, cat. no. 60186-1-Ig). Representative blots from three independent biological experiments are shown. Band intensities were quantified using ImageJ and normalized to GAPDH or β-actin.

Statistical analyses

Statistical analyses were performed using GraphPad Prism 7.0 (GraphPad Software, San Diego, CA, USA). Unless otherwise indicated, all in vitro experiments were performed independently at least three times, and data are presented as the mean ± standard deviation (SD). Comparisons between two groups were performed using the unpaired two-tailed Student’s t-test. For comparisons involving multiple groups, one-way or two-way analysis of variance (ANOVA) was performed as appropriate, followed by Tukey’s or Šídák’s multiple-comparisons test. For longitudinal xenograft experiments, tumor volume and body weight measurements obtained repeatedly from the same animals over time were analyzed using two-way repeated-measures ANOVA, with time as the within-subject factor and experimental group as the between-subject factor, followed by Šídák’s multiple-comparisons test. Associations between BRAP expression and clinicopathological characteristics were analyzed using the Pearson’s χ² test (or Fisher’s exact test when appropriate). A two-sided p < 0.05 was considered statistically significant.

Results

BRAP is preferentially upregulated in cervical adenocarcinoma and correlates with poor histological differentiation

To determine the protein expression pattern of BRAP in cervical cancer, immunohistochemical staining was performed to evaluate BRAP expression in normal endocervical tissues (obtained from cervical polyps, CP), cervical squamous cell carcinoma (SCC), and cervical adenocarcinoma (ADC). As shown in Fig. 1A, strong BRAP immunoreactivity was observed in normal endocervical glandular epithelial cells (red arrows) and ADC cells (blue arrows), whereas little or no BRAP staining was detected in SCC tissues. Quantitative analysis of the immunohistochemical staining demonstrated that BRAP expression was significantly higher in ADC tissues than in both normal endocervical tissues and SCC tissues (Fig. 1A). These observations were further validated by western blot analysis of tissue lysates. Compared with normal endocervical tissues and SCC tissues, ADC tissues exhibited significantly higher BRAP protein expression (Fig. 1B). BRAP protein expression was subsequently evaluated in cervical cancer cell lines. Western blot analysis showed that BRAP expression was highest in HeLa cells, intermediate in HT-3 cells, and lowest in SiHa cells. HaCat cells, included as a non-malignant epithelial reference cell line, exhibited minimal BRAP expression (Fig. 1C). Because HeLa and SiHa cells displayed the highest and lowest endogenous BRAP expression, respectively, and represent cervical adenocarcinoma- and squamous cell carcinoma-derived cell lines, these two cell lines were selected for subsequent functional studies.

Fig. 1.

BRAP expression is elevated in cervical adenocarcinoma. (A) Representative immunohistochemical staining of BRAP in cervical polyp (CP), cervical squamous cell carcinoma (SCC), and cervical adenocarcinoma (ADC) tissues. Scale bar = 50 μm. BRAP-positive staining was quantified as the percentage of the total tissue area using ImageJ. Representative images from four independent patients are shown (n = 4). (B) Representative western blot analysis of BRAP protein expression in CP, SCC, and ADC tissues, together with densitometric quantification (n = 3 tissue samples per group). (C) Representative western blot analysis of BRAP protein expression in HaCat, HT-3, HeLa, and SiHa cells, together with densitometric quantification (n = 3). Quantitative analyses are shown in the adjacent panels. Data are presented as the mean ± SD from three independent experiments unless otherwise indicated. *p < 0.05, **p < 0.01, ***p < 0.001,****p < 0.0001

The clinical relevance of BRAP expression was further assessed in an independent cohort of 61 cervical adenocarcinoma specimens by immunohistochemistry. According to the predefined IHC scoring criteria, patients were classified into BRAP-positive and BRAP-negative groups. As summarized in Table 4, BRAP expression showed no significant association with patient age (p = 0.797) or FIGO stage (p = 0.787). In contrast, the proportion of BRAP-positive cases was significantly higher in poorly differentiated tumors than in well- or moderately differentiated tumors (78.3% vs. 47.4%, p = 0.031). These findings suggest that elevated BRAP expression is associated with poor histological differentiation in cervical adenocarcinoma.

Table 4.

Association between BRAP protein expression and clinicopathological parameters in ADC patients

Clinicopathological features	BRAP		p-value
	Negative expression (%)	Positive expression (%)
Age
≤49 years(n = 31)	12(38.7)	19(61.3)
＞49 years (n = 30)	13(43.3)	17(56.7)	0.797
Pathological grades
Well and Moderate(n = 38)	20(52.6)	18(47.4)
Poor(n = 23)	5(21.7)	18(78.3)	0.031*
Clinical stages
I(n = 39)	17(43.6)	22(56.4)
II-III (n = 22)	8(36.4)	14(63.6)	0.787

BRAP promotes proliferation of cervical adenocarcinoma cells in vitro and in vivo

HeLa and SiHa cells were transfected with either C6orf89-specific siRNA (siC6orf89) or a non-targeting siRNA (siCtrl). Efficient BRAP knockdown was confirmed in both cell lines by qRT-PCR and western blot analysis, demonstrating a marked reduction in BRAP mRNA (Fig. 2A) and protein expression (Fig. 2B). We next investigated the effects of BRAP depletion on cell proliferation. Compared with the corresponding siCtrl-transfected cells, BRAP knockdown significantly reduced colony formation in HeLa cells but had no significant effect in SiHa cells (Fig. 2C). Likewise, compared with the corresponding siCtrl group, EdU incorporation was markedly decreased in BRAP-depleted HeLa cells, whereas no significant difference was observed in SiHa cells (Fig. 2D). Similarly, BRAP depletion significantly reduced cell viability in HeLa cells but not in SiHa cells compared with their respective siCtrl groups (Fig. 2E). Cell cycle analysis further demonstrated that, compared with the corresponding siCtrl group, BRAP knockdown induced G1-phase arrest in HeLa cells, as evidenced by an increased G1-phase population and a reduced S-phase population, whereas cell cycle distribution remained unchanged in SiHa cells (Fig. 2F). These results suggest that the proliferative effects of BRAP depletion are cell type dependent.

Fig. 2.

Effects of BRAP silencing on cell viability, proliferation, and cell-cycle progression in cervical cancer cells. (A) Quantitative real-time PCR (qRT-PCR) was performed to assess C6orf89 mRNA expression in HeLa (left) and SiHa (right) cells transfected with C6orf89-specific siRNA (siC6orf89) or non-targeting control siRNA (siCtrl) (n = 4). (B) Representative western blot analysis of BRAP protein expression in HeLa (left) and SiHa (right), together with densitometric quantification (n = 3). (C) Colony formation assays were performed to evaluate clonogenic growth. HeLa (left) and SiHa (right) cells were retransfected with C6orf89-specific siRNA or control siRNA every 3 days throughout the 14-day culture period. (D) EdU incorporation assay was performed to evaluate cell proliferation. Scale bar = 100 μm. (E) CCK-8 assay was performed to assess cell viability (n = 5). (F) Flow cytometry was performed to analyze cell-cycle distribution (n = 3). Quantitative analyses are shown in the adjacent panels. Data are presented as the mean ± SD. *p < 0.05, **p < 0.01, ***p < 0.001,****p < 0.0001

The effect of BRAP depletion on tumor growth was further evaluated in a xenograft model. Tumors derived from BRAP-knockdown HeLa cells grew significantly more slowly and exhibited smaller tumor volumes and lower final tumor weights than those derived from control cells (Fig. S1A–D). No significant difference in mouse body weight was observed between the two groups (Fig. S1E). Western blot analysis of xenograft tumor tissues confirmed BRAP knockdown and demonstrated significantly lower PCNA protein expression in BRAP-knockdown tumors than in control tumors (Fig. S1F), consistent with reduced tumor cell proliferative activity in vivo.

To further verify these findings, a stable BRAP-knockdown HeLa cell line was established using lentivirus-mediated shRNA (shC6orf89). Lentiviral transduction efficiency exceeded 90%, as indicated by GFP fluorescence (Fig. S2A). Western blot analysis further confirmed efficient and sustained BRAP knockdown (Fig. S2B). Consistent with the transient knockdown results, stable BRAP knockdown significantly suppressed xenograft tumor growth. Compared with the shCtrl group, mice implanted with shC6orf89 HeLa cells developed markedly smaller tumors (Fig. S2C–D), as reflected by reduced tumor volume throughout the observation period (Fig. S2E) and lower tumor weight at the experimental endpoint (Fig. S2F), whereas no significant difference in body weight was observed between the two groups (Fig. S2G). Western blot analysis of xenograft tumor tissues further confirmed sustained BRAP knockdown and significantly reduced PCNA protein expression in shC6orf89-derived tumors compared with shCtrl-derived tumors (Fig. S2H), further supporting impaired tumor cell proliferation in vivo.

Collectively, findings from both transient and stable BRAP knockdown models demonstrate that BRAP promotes the proliferation and tumorigenic growth of cervical adenocarcinoma-derived HeLa cells in vitro and in vivo.

BRAP inhibits apoptosis and promotes migration, invasion, and EMT in cervical adenocarcinoma cells

HeLa and SiHa cells were transfected with either C6orf89-specific siRNA (siC6orf89) or a non-targeting control siRNA (siCtrl). Compared with their respective siCtrl-transfected controls, BRAP knockdown significantly increased apoptosis in HeLa cells, as demonstrated by Annexin V-FITC/PI staining and flow cytometric analysis, whereas apoptosis remained unchanged in SiHa cells (Fig. 3A). Consistently, Western blot analysis showed increased Cleaved-caspase3 protein expression in HeLa cells, with no corresponding change observed in SiHa cells (Fig. 3B). Likewise, BRAP knockdown significantly reduced wound closure in HeLa cells at 48 h, whereas wound healing was unaffected in SiHa cells (Fig. 3C). Transwell assays further demonstrated that BRAP depletion significantly inhibited migration and invasion in HeLa cells, an effect not observed in SiHa cells (Fig. 3D).

Fig. 3.

BRAP suppresses apoptosis and promotes migration, invasion, and EMT in cervical adenocarcinoma cells. (A) Apoptosis was analyzed by Annexin V-FITC/propidium iodide (PI) staining followed by flow cytometry in HeLa (left) and SiHa (right) cells 48 h after transfection with C6orf89-specific siRNA (siC6orf89) or non-targeting control siRNA (siCtrl) (n = 3). (B) Representative western blot analysis of Caspase3 and Cleaved-caspase3 protein expression, together with densitometric quantification (n = 3). (C) Wound healing assays were performed to evaluate cell migration. Cells were pretreated with mitomycin C before scratch generation to inhibit cell proliferation. Scale bar = 100 μm. (D) Transwell migration and invasion assays were performed to assess the migratory and invasive capacities of HeLa and SiHa cells. Scale bar = 100 μm. (E) Representative western blot analysis of EMT markers (E-cadherin and N-cadherin), together with densitometric quantification (n = 3). (F) qRT-PCR was performed to analyze the mRNA expression of the EMT-associated transcription factors SLUG, TWIST1, ZEB1, and ZEB2 (n = 3). Quantitative analyses are shown in the adjacent panels. Data are presented as the mean ± SD. *p < 0.05, **p < 0.01, ***p < 0.001,****p < 0.0001

Given the critical role of epithelial–mesenchymal transition (EMT) in tumor metastasis [31], we next examined EMT-associated markers. Compared with their respective siCtrl-transfected controls, BRAP knockdown increased E-cadherin expression and decreased N-cadherin expression in HeLa cells, whereas neither marker was significantly altered in SiHa cells (Fig. 3E). In addition, BRAP depletion reduced the expression of the EMT-associated transcription factors SLUG, TWIST1, and ZEB1 in HeLa cells. By contrast, SiHa cells exhibited increased expression of SLUG and ZEB2 following BRAP knockdown (Fig. 3F), indicating distinct molecular responses to BRAP depletion between the two cell lines.

Collectively, these findings demonstrate that BRAP suppresses apoptosis while promoting migration, invasion, and EMT in cervical adenocarcinoma-derived HeLa cells, whereas these effects are not evident in SiHa cells.

BRAP overexpression rescues the phenotypic defects induced by BRAP knockdown

The specificity of the observed phenotypes was further validated through rescue experiments by re-expressing BRAP in BRAP-silencing HeLa cells. Stable BRAP knockdown cells were generated via lentiviral transduction of C6orf89-specific shRNA followed by puromycin selection, and subsequently reconstituted with a BRAP-expressing plasmid (pcDNA3.1-BRAP) or empty vector control. Successful BRAP re-expression was confirmed by western blotting (Fig. S3A). Compared with BRAP-knockdown cells, BRAP re-expression significantly restored cell viability (Fig. S3B), colony-forming ability (Fig. S3C), and EdU incorporation (S3. S3D). Consistently, BRAP re-expression also restored the migratory and invasive capacities of BRAP-deficient HeLa cells, as demonstrated by wound healing and Transwell assays (Fig. S3E–F).

At the molecular level, BRAP knockdown reduced Cyclin D1 and CDK6 expression while increasing Cleaved-caspase3 levels; these alterations were effectively reversed upon BRAP re-expression (Fig. S3G). In parallel, BRAP re-expression reversed the BRAP knockdown-induced increase in E-cadherin expression and decrease in N-cadherin expression, accompanied by restoration of SLUG, TWIST1, and ZEB1 expression. (Fig. S3G–H).

Collectively, BRAP re-expression fully rescues the proliferative, migratory, invasive, apoptotic, and EMT-related defects caused by BRAP depletion, confirming that these phenotypes are specifically attributable to BRAP loss.

BRAP silencing activates the PERK–ATF4–DDIT3/CHOP pathway to regulate EMT and apoptosis in HeLa cells

Transmission electron microscopy (TEM) revealed marked ultrastructural abnormalities in BRAP-depleted HeLa cells. As shown in Fig. 4A, siCtrl-transfected HeLa cells exhibited intact morphology and normal architecture of the endoplasmic reticulum (ER) and Golgi apparatus. In contrast, BRAP-knockdown HeLa cells displayed pronounced ER swelling and dilation, together with vacuolar degeneration, degranulation, and Golgi disorganization, indicative of ER stress. These morphological alterations were not observed in BRAP-depleted SiHa cells (Fig. S4A), further supporting a cell type-specific stress response.

Fig. 4.

BRAP silencing drives EMT and apoptosis via activation of the PERK–ATF4–CHOP pathway in HeLa cells. (A) Transmission electron microscopy (TEM) was performed to examine ultrastructural changes in the endoplasmic reticulum and Golgi apparatus in HeLa cells. Scale bars, 5 μm (left) and 2 μm (magnified image, right). (B) Representative western blot analysis of UPR-related proteins (GRP78, PERK, phospho-PERK, ATF4, DDIT3/CHOP, ATF6, IRE1α, and phospho-IRE1α), together with densitometric quantification (n = 3). (C) Representative western blot analysis of UPR-related proteins in the indicated groups (shCtrl, shC6orf89, vector, BRAP, and shC6orf89 + BRAP), together with densitometric quantification (n = 3). (D) Representative western blot analysis of UPR-related proteins (PERK, phospho-PERK, ATF4, and CHOP), EMT-related proteins (N-cadherin and E-cadherin), and apoptosis-related proteins (Caspase3 and Cleaved-caspase3) in BRAP-silenced cells treated with or without 4-phenylbutyric acid (4-PBA; 10 μM, 24 h), together with densitometric quantification (n = 3). (E) Representative western blot analysis of UPR-related proteins (PERK, phospho-PERK, ATF4, and CHOP), EMT-related proteins (N-cadherin and E-cadherin), and apoptosis-related proteins (Caspase3 and Cleaved-caspase3) in BRAP-silenced cells treated with or without GSK2606414 (1 μM, 24 h), together with densitometric quantification (n = 3). Quantitative analyses are shown in the adjacent panels. Data are presented as the mean ± SD. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001

Western blot analysis further demonstrated that BRAP knockdown selectively activated the UPR in HeLa cells. As shown in Fig. 4B, BRAP-depleted HeLa cells exhibited significantly increased expression of GRP78, phosphorylated PERK (p-PERK), ATF4, and DDIT3/CHOP compared with siCtrl-transfected HeLa cells, whereas ATF6 and phosphorylated IRE1α (p-IRE1α) remained unchanged, indicating preferential activation of the PERK branch of the UPR. In contrast, BRAP knockdown did not significantly alter these UPR markers in SiHa cells (Fig. S4B). Furthermore, BRAP re-expression significantly reversed the BRAP knockdown-induced activation of the PERK–ATF4–DDIT3/CHOP pathway, as evidenced by reduced levels of p-PERK, ATF4, and DDIT3/CHOP (Fig. 4C).

We next investigated whether ER stress is required for BRAP depletion–induced EMT in HeLa cells. Pharmacological alleviation of ER stress using the chemical chaperone 4-phenylbutyric acid (4-PBA) effectively suppressed the upregulation of p-PERK, ATF4, and DDIT3/CHOP induced by BRAP silencing in HeLa cells (Fig. 4D). In addition, 4-PBA reversed the BRAP knockdown-induced increase in E-cadherin expression and decrease in N-cadherin expression, and also attenuated the BRAP knockdown-induced increase in Cleaved-caspase 3 levels (Fig. 4D), indicating that the EMT alterations and apoptosis induced by BRAP depletion are mediated by ER stress-associated UPR activation. However, 4-PBA treatment did not significantly alter UPR markers, EMT markers, or Cleaved-caspase3 expression in BRAP-knockdown SiHa cells (Fig. S4C).

In addition, GSK2606414, a potent and selective PERK inhibitor, was used to inhibit PERK–eIF2α signaling [32]. Compared with untreated BRAP-knockdown cells, GSK2606414 significantly reduced p-PERK, ATF4, and DDIT3/CHOP expression (Fig. 4E). GSK2606414 treatment also reversed the BRAP knockdown-induced increase in E-cadherin expression and decrease in N-cadherin expression, while reducing Cleaved-caspase3 protein levels (Fig. 4E).

These findings demonstrate that BRAP depletion induces ER stress and preferentially activates the PERK–ATF4–DDIT3/CHOP branch of the UPR, thereby mediating the EMT alterations and apoptosis observed in HeLa cells.

BRAP regulates ER–Golgi trafficking through interaction with SEC24D

To explore the molecular mechanisms underlying BRAP-mediated ER stress regulation, RNA sequencing was performed in HeLa cells transfected with C6orf89-specific siRNA (siC6orf89) or non-targeting siRNA (siCtrl), followed by Gene Ontology (GO) enrichment analysis (Fig. 5A). Differentially expressed genes were significantly enriched in pathways related to metal ion homeostasis (including copper and zinc), cell cycle regulation, and transcriptional termination. At the cellular component level, enrichment of the SMN complex, transcription export complex, Sec61 translocon, and cytoplasmic ribonucleoprotein granules suggested a close association with protein translocation and RNA–protein processing. Molecular function analysis further highlighted SNARE binding, protein–RNA adaptor activity, metal ion binding, and transmembrane transporter activity. Collectively, these findings indicate that BRAP-associated transcriptional changes are enriched in biological processes related to protein transport and the secretory pathway.

Fig. 5.

BRAP regulates ER–Golgi trafficking through interaction with SEC24D. (A) Gene Ontology (GO) biological process enrichment analysis was performed using transcriptomic data from BRAP-silenced HeLa cells and control cells. (B) Yeast two-hybrid assays were performed to identify interactions between the C-terminal domain of BRAP and vesicle-associated proteins. (C) Confocal immunofluorescence microscopy was performed to examine the localization of the endoplasmic reticulum (ER; CKAP4) and the cis-Golgi (GM130) in control and BRAP-silenced cells (n = 6). Nuclei were counterstained with DAPI (blue). Scale bar = 25 μm. (D) Confocal immunofluorescence microscopy was performed to examine the localization of the ER exit site (ERES) markers SEC16A and LMAN1 in control and BRAP-silenced cells (n = 6). Nuclei were counterstained with DAPI (blue). Scale bar = 25 μm. (E) Yeast two-hybrid assays were performed to assess interactions between the C-terminal domain of BRAP and full-length SEC24D, SEC23B, SEC24B, SEC24C, or TSG101. Empty vectors (pGBKT7 and pGADT7) served as negative controls, whereas ATG5 served as a positive interaction control. (F) Co-immunoprecipitation assays were performed using anti-SEC24D or control IgG antibodies, followed by western blot analysis using anti-BRAP and anti-SEC24D antibodies. (G) Molecular docking analysis was performed to predict the interaction interface between BRAP and SEC24D. (H) Representative western blot analysis of SEC24D protein expression, together with densitometric quantification (n = 3). GAPDH served as the loading control. (I) Quantitative real-time PCR (qRT-PCR) was performed to assess SEC24D mRNA expression (n = 3). (J) Representative western blot analysis of SEC24D protein expression in HaCat, HT-3, HeLa, and SiHa cells (n = 3). (K) Representative western blot analysis of SEC24D protein expression in CP, SCC, and ADC tissues. Quantitative analyses are shown in the adjacent panels. Data are presented as the mean ± SD from three independent experiments unless otherwise indicated. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001

As summarized in Fig. 5B, our previous yeast two-hybrid (Y2H) screen identified multiple ER-to-Golgi trafficking-related proteins as candidate BRAP-interacting partners [19]. Based on this observation, we performed immunofluorescence analysis to investigate the role of BRAP in maintaining ER–Golgi organization and ER exit site (ERES) homeostasis. Compared with siCtrl-transfected HeLa cells, BRAP-depleted cells exhibited ER expansion, as indicated by CKAP4 staining, and Golgi fragmentation, as revealed by GM130 staining (Fig. 5C). Moreover, quantification of SEC16A-positive ERES and peripheral LMAN1 (ERGIC-53)-positive structures, well-established markers of ERES number [33] and function [34], respectively, revealed a significant reduction following BRAP knockdown (Fig. 5D). Re-expression of BRAP largely reversed these defects, restoring ER morphology, Golgi integrity, and ERES architecture (Fig. S5A–B), thereby reinforcing the role of BRAP in maintaining ER–Golgi homeostasis.

To identify the molecular mediator of these effects, candidate proteins identified in the Y2H screen were further examined. Validation Y2H assays demonstrated that SEC24D, but not SEC23B, SEC24B, SEC24C, or TSG101, interacted with the C-terminal domain of BRAP (Fig. 5E). ATG5, a previously validated BRAP-interacting protein [19], was included as a positive control to verify the reliability of the Y2H assay. Co-immunoprecipitation experiments further confirmed the interaction between endogenous SEC24D and BRAP in HeLa cells (Fig. 5F). Structural docking analysis using the GRAMM-X server further predicted a binding interface between BRAP and SEC24D (Fig. 5G).

We next examined whether BRAP regulates SEC24D expression. Compared with siCtrl-transfected HeLa cells, BRAP knockdown significantly reduced SEC24D protein expression (Fig. 5H), whereas SEC24D mRNA levels remained unchanged (Fig. 5I), suggesting that BRAP regulates SEC24D protein stability at the post-transcriptional level. Further analysis showed that SEC24D protein expression was markedly higher in HeLa cells than in SiHa cells and was significantly increased in cervical adenocarcinoma tissues compared with cervical squamous cell carcinoma tissues (Fig. 5J–K). These observations reveal that BRAP maintains SEC24D protein stability and thereby preserves ER–Golgi trafficking and ERES homeostasis in cervical adenocarcinoma cells.

BRAP depletion promotes autophagic degradation of SEC24D

Given that BRAP depletion reduced SEC24D protein expression without affecting its mRNA expression, we next investigated whether BRAP regulates SEC24D protein stability. Compared with siCtrl-transfected HeLa cells, cycloheximide (CHX) chase assays showed that SEC24D degradation was significantly accelerated in BRAP-knockdown HeLa cells (Fig. 6A), indicating reduced SEC24D protein stability.

Fig. 6.

BRAP silencing promotes autophagic degradation of SEC24D. (A) Representative western blot analysis of SEC24D protein stability following cycloheximide (CHX; 100 ng/mL) treatment at the indicated time points, together with densitometric quantification (n = 3). (B) Representative western blot analysis of SEC24D protein expression following chloroquine (CQ; 25 μM, 30 min) treatment, together with densitometric quantification (n = 3). (C) Representative western blot analysis of SEC24D protein expression following MG132 pretreatment (10 μM, 30 min), together with densitometric quantification (n = 3). (D) Representative western blot analysis of LC3 and p62 protein expression, together with densitometric quantification (n = 3). (E) Representative images and quantitative analysis of GFP-p62 fluorescence in HeLa cells. Cells were transduced with Ad-GFP-p62 adenovirus for 24 h, cultured in serum-free medium for an additional 24 h, and then treated with CQ (25 μM, 24 h). GFP-p62 fluorescence was examined by fluorescence microscopy. Scale bar = 100 μm. Nuclei were counterstained with DAPI (blue). (F) Representative images and quantitative analysis of mCherry-GFP-LC3B fluorescence in HeLa cells. Cells were transduced with Ad-mCherry-GFP-LC3B adenovirus for 24 h, cultured in serum-free medium for an additional 24 h, and then treated with CQ (25 μM, 24 h). GFP and mCherry fluorescence were examined by fluorescence microscopy. Scale bar = 100 μm. Nuclei were counterstained with DAPI (blue). Quantitative analyses are shown in the adjacent panels. Data are presented as the mean ± SD from three independent experiments. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001

To determine the degradation pathway responsible for SEC24D turnover, BRAP-knockdown HeLa cells were treated with chloroquine (CQ), an autophagy–lysosome inhibitor, or MG132, a 26S proteasome inhibitor. Compared with untreated BRAP-knockdown cells, CQ treatment partially restored SEC24D protein expression, whereas MG132 had no detectable effect, as shown in Fig. 6B–C, suggesting that SEC24D is predominantly degraded through the autophagy–lysosome pathway. Consistent with these findings, BRAP knockdown markedly enhanced autophagic activity in HeLa cells. Compared with the siCtrl group, LC3-II protein expression was significantly increased (Fig. 6D), indicating enhanced autophagosome formation. Autophagic flux was further assessed using GFP-p62 and mCherry-GFP-LC3B reporters. Autophagic flux was further assessed using GFP-p62 and mCherry-GFP-LC3B reporters. Consistently, BRAP-knockdown cells exhibited markedly fewer GFP-p62 puncta under both basal conditions and following CQ treatment (Fig. 6E), consistent with enhanced p62 degradation. Moreover, BRAP-knockdown cells displayed increased numbers of mCherry-positive puncta, and CQ treatment further enhanced puncta accumulation (Fig. 6F), confirming increased autophagic flux. Taken together, these results demonstrate that BRAP depletion enhances autophagic flux in HeLa cells, thereby promoting lysosome-dependent degradation of SEC24D.

SEC24D overexpression rescues BRAP knockdown-induced phenotypic defects by restoring ER–Golgi trafficking and suppressing UPR activation

To determine whether BRAP and SEC24D function within the same signaling axis, rescue experiments were performed by overexpressing SEC24D in BRAP-knockdown HeLa cells. HeLa cells were transfected with non-targeting siRNA (siCtrl) or C6orf89-specific siRNA (siC6orf89), followed by transfection with a SEC24D expression plasmid (pcDNA3.1-SEC24D) or the corresponding empty vector. Successful SEC24D overexpression was confirmed by western blot analysis (Fig. 7A).

Fig. 7.

SEC24D overexpression rescues the phenotypic defects induced by BRAP silencing. (A) Representative western blot analysis of BRAP and SEC24D protein expression in the indicated groups (siCtrl, siC6orf89, vector, SEC24D, and siC6orf89 + SEC24D), together with densitometric quantification (n = 3). (B) Cell viability was assessed using the CCK-8 assay (n = 5). (C) Colony formation assays were performed to evaluate clonogenic growth. (D) EdU incorporation assay was performed to evaluate cell proliferation. Scale bar = 100 μm. (E) Wound healing assays were performed to evaluate cell migration. Cells were pretreated with mitomycin C before scratch generation to inhibit cell proliferation. Scale bar = 100 μm. (F) Transwell migration and invasion assays were performed to assess the migratory and invasive capacities of the indicated groups. Scale bar = 100 μm. (G) Representative western blot analysis of cell cycle-related proteins (Cyclin D1 and CDK6), apoptosis-related proteins (Caspase3 and Cleaved-caspase3), and EMT-related proteins (E-cadherin and N-cadherin), together with densitometric quantification (n = 3). (H) qRT-PCR was performed to analyze the mRNA expression of the EMT-associated transcription factors SLUG, TWIST1, and ZEB1 (n = 3). Quantitative analyses are shown in the adjacent panels. Data are presented as the mean ± SD from three independent experiments unless otherwise indicated. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001

Compared with the siC6orf89 group, SEC24D overexpression significantly rescued the proliferative defects induced by BRAP knockdown. CCK-8 assays showed increased cell viability (Fig. 7B), accompanied by enhanced colony formation (Fig. 7C) and increased EdU incorporation (Fig. 7D). SEC24D overexpression also restored the migratory and invasive capacities of BRAP-knockdown HeLa cells, as demonstrated by wound healing and Transwell assays (Fig. 7E–F). BRAP knockdown reduced the expression of Cyclin D1 and CDK6 while increasing Cleaved-caspase 3 levels; these alterations were effectively reversed by SEC24D overexpression (Fig. 7G). Similarly, SEC24D overexpression reversed the BRAP knockdown-induced increase in E-cadherin expression and decrease in N-cadherin expression, accompanied by restoration of SLUG, TWIST1, and ZEB1 expression (Fig. 7G–H). Together, these findings indicate that SEC24D acts as a downstream effector of BRAP, mediating its effects on cell proliferation, migration, invasion, apoptosis, and EMT in HeLa cells.

We next investigated whether SEC24D restoration could rescue the defects in ER–Golgi trafficking and UPR activation caused by BRAP depletion. Compared with the siC6orf89 group, SEC24D overexpression effectively restored ER morphology and Golgi integrity (Fig. 8A), accompanied by recovery of SEC16A-positive ER exit sites (ERES) and peripheral LMAN1-positive structures (Fig. 8B), indicating restoration of ER–Golgi trafficking. Furthermore, SEC24D overexpression attenuated activation of the PERK–ATF4–DDIT3/CHOP pathway, as evidenced by reduced expression of phosphorylated PERK, ATF4, and DDIT3/CHOP (Fig. 8C). These findings demonstrate that restoration of SEC24D re-establishes ER–Golgi trafficking homeostasis and suppresses BRAP depletion-induced PERK-dependent UPR activation.

Fig. 8.

SEC24D restoration rescues BRAP silencing-induced endoplasmic reticulum stress. (A) Confocal immunofluorescence microscopy was performed to examine the localization of GM130 and CKAP4 in the indicated groups (siCtrl, siC6orf89, vector, SEC24D, and siC6orf89 + SEC24D) (n = 6). Nuclei were counterstained with DAPI (blue). Scale bar = 25 μm. (B) Confocal immunofluorescence microscopy was performed to examine the localization of the ERES markers SEC16A and LMAN1 in the indicated groups (n = 6). Nuclei were counterstained with DAPI (blue). Scale bar = 25 μm. (C) Representative western blot analysis of UPR-related proteins (PERK, phospho-PERK, ATF4, and DDIT3/CHOP), together with densitometric quantification (n = 3). Quantitative analyses are shown in the adjacent panels. Data are presented as the mean ± SD from three independent experiments unless otherwise indicated. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001

Discussion

Our current study identifies BRAP as a previously unrecognized regulator of endoplasmic reticulum (ER)–Golgi homeostasis in cervical adenocarcinoma. Although BRAP has previously been implicated in NF-κB signaling, inflammatory responses, EMT, and autophagy [18–21], its involvement in regulating the early secretory pathway during tumor progression has remained largely unexplored. Here, we demonstrate that BRAP is preferentially upregulated in cervical adenocarcinoma and promotes malignant progression by maintaining SEC24D stability, preserving ER-to-Golgi trafficking, and preventing excessive activation of PERK-dependent ER stress. These findings substantially expand the current understanding of BRAP biology and identify a previously unrecognized BRAP–SEC24D signaling axis that links secretory pathway homeostasis to cervical adenocarcinoma progression.

One of the most noteworthy findings of this study is the subtype-specific role of BRAP in cervical cancer. BRAP expression was markedly elevated in cervical adenocarcinoma tissues and HeLa cells but remained extremely low in cervical squamous cell carcinoma tissues and SiHa cells. Consistently, BRAP depletion profoundly impaired proliferation, migration, invasion, EMT, and survival only in HeLa cells, whereas little effect was observed in SiHa cells. These observations suggest that BRAP is not a universal regulator of cervical cancer but instead exerts lineage-dependent functions. Cervical adenocarcinoma originates from glandular epithelial cells, which are specialized for the synthesis and secretion of proteins, including mucins, digestive enzymes, and peptide hormones, and therefore require a highly developed secretory apparatus to sustain their physiological functions. Consistent with this concept, previous studies have shown that glandular epithelial cells exhibit enhanced secretory activity together with prominent ER and Golgi remodeling, and that secretory pathway remodeling is closely associated with the progression of glandular malignancies [35, 36]. These observations suggest that glandular-derived tumors may be particularly susceptible to perturbations in ER–Golgi trafficking. Therefore, disruption of the early secretory pathway would be expected to impose a greater burden on cervical adenocarcinoma cells than on squamous carcinoma cells, providing a plausible explanation for the selective vulnerability of HeLa cells observed in the present study.

Mechanistically, our study identifies SEC24D as a novel downstream effector of BRAP. SEC24D is a cargo-selective adaptor of the COPII coat complex and plays an indispensable role in ER export and maintenance of secretory pathway integrity [37, 38]. Previous studies have demonstrated that disruption of SEC24D impairs COPII vesicle formation, compromises ER export, and induces ER stress [39, 40]. In agreement with these observations, BRAP depletion markedly reduced SEC24D protein abundance without affecting its mRNA expression, accompanied by ER swelling, Golgi fragmentation, loss of ER exit sites, and activation of ER stress. Importantly, restoration of SEC24D largely rescued these abnormalities together with the impaired proliferation, migration, invasion, and EMT phenotypes, demonstrating that SEC24D represents a major functional downstream mediator of BRAP.

Our findings further establish impaired SEC24D-dependent ER export as the primary trigger of ER stress following BRAP depletion. Efficient COPII-mediated trafficking is essential for maintaining ER proteostasis by ensuring continuous export of newly synthesized proteins from the ER. Disruption of this process leads to protein retention within the ER lumen and activation of the UPR [41–43]. Consistent with this model, BRAP-deficient cells exhibited ER swelling, reduced ER exit site numbers, Golgi fragmentation, and selective activation of the PERK–ATF4–DDIT3/CHOP signaling pathway. Importantly, pharmacological attenuation of ER stress with 4-phenylbutyric acid (4-PBA) or inhibition of PERK signaling with GSK2606414 significantly reduced apoptosis and restored EMT-associated protein expression, demonstrating that excessive PERK activation is a key driver of the malignant phenotypes induced by BRAP depletion. Collectively, these findings identify defective SEC24D-dependent ER export as the mechanistic critical link between BRAP deficiency and persistent ER stress, underscoring the importance of secretory pathway integrity in maintaining cervical adenocarcinoma cell homeostasis.

An important mechanistic question raised by our findings is how BRAP maintains SEC24D stability. Our data demonstrate that BRAP regulates SEC24D predominantly through the autophagy–lysosome pathway rather than the ubiquitin–proteasome system. MG132 failed to restore SEC24D abundance following BRAP depletion, whereas chloroquine markedly rescued SEC24D protein expression. Consistently, BRAP deficiency enhanced autophagic flux, as evidenced by increased LC3-II accumulation, enhanced mCherry-GFP-LC3B puncta formation, and accelerated p62 degradation. These observations indicate that BRAP preserves SEC24D stability by limiting its autophagic degradation. This mechanism differs from the previously reported TECPR2-dependent regulation of SEC24D, in which TECPR2 selectively protects SEC24D from lysosomal degradation without affecting global autophagic activity [40]. In contrast, BRAP suppresses autophagic activity at the cellular level, thereby indirectly maintaining SEC24D abundance. Recent studies further demonstrated that site-specific O-GlcNAcylation dynamically regulates SEC24D protein interactions during cargo transport, highlighting that SEC24D is a dynamically regulated COPII adaptor rather than a constitutively stable structural component [44]. Building upon our previous finding that BRAP physically interacts with ATG5 and negatively regulates autophagy [19], the present study identifies BRAP as a previously unrecognized upstream regulator linking autophagy to COPII homeostasis. Future studies are warranted to determine how SEC24D is selectively recognized by the autophagy machinery and whether additional COPII regulatory mechanisms cooperate with BRAP to maintain ER homeostasis.

From a clinical perspective, BRAP expression was significantly associated with poor histological differentiation in cervical adenocarcinoma, suggesting that BRAP may contribute to tumor progression and aggressive tumor behavior. Nevertheless, the relatively limited sample size and the lack of survival data precluded a definitive assessment of its prognostic value. Future studies involving larger multicenter cohorts with long-term follow-up are warranted to determine whether BRAP could serve as a prognostic biomarker or a potential therapeutic target for cervical adenocarcinoma.

Several limitations of the present study should also be acknowledged. First, although BRAP was shown to physically interact with SEC24D, the precise binding domains and structural determinants underlying this interaction remain to be defined. Second, although our data demonstrate that BRAP depletion promotes autophagic degradation of SEC24D, the selective autophagy receptors responsible for this process remain unknown. Third, owing to the limited availability of cervical adenocarcinoma cell models, mechanistic studies were primarily performed in HeLa cells, and further validation using additional adenocarcinoma models, patient-derived organoids, or genetically engineered mouse models will strengthen the physiological relevance of the BRAP–SEC24D axis.

In conclusion, our study establishes BRAP as a central regulator of ER–Golgi homeostasis in cervical adenocarcinoma. By maintaining SEC24D stability and preserving COPII-dependent ER export, BRAP restrains excessive PERK-mediated ER stress, thereby promoting proliferation, survival, migration, invasion, and EMT. These findings not only expand the biological functions of BRAP but also identify the BRAP–SEC24D axis as a potential therapeutic vulnerability in cervical adenocarcinoma.

Electronic supplementary material

Below is the link to the electronic supplementary material.

Abbreviations

AD

Activation Domain

ADC

Cervical Adenocarcinoma

ANOVA

One-Way Analysis of Variance

BRAP

Bombesin Receptor- Activated Protein

BSA

Bovine Serum Albumin

CC

Cervical Cancer

CHX

Cycloheximide

Co-IP

Co-immunoprecipitation

COPII

Coat Protein II

CP

Cervical Polyps

CQ

Chloroquine

DMEM

Dulbecco’s Modified Eagle’s Medium

ECL

Enhanced Chemiluminescence

EDU

5-Ethynyl-2’-deoxyuridine

EMT

Epithelial–Mesenchymal Transition

ER

Endoplasmic Reticulum

ERES

Endoplasmic Reticulum Exit Site

FBS

Fetal Bovine Serum

FFPE

Formalin-Fixed, Paraffin-Embedded

FIGO

International Federation of Gynecology and Obstetrics

GO

Gene Ontology

HaCat

Immortalized Human Keratinocyte Cell Line

HeLa

Human Cervical Adenocarcinoma Cell Line

HPV

Human Papillomavirus

HT-3

Human Cervical Carcinoma Cell Line

IECC

International Endocervical Adenocarcinoma Criteria and Classification

IHC

Immunohistochemical

MOI

Multiplicity Of Infection

NF-κB

Nuclear Factor-kappa B

PBS

Phosphate Buffered Saline

PI

Propidium Iodide

PVDF

Polyvinylidene Fluoride

SCC

Cervical Squamous Cell Carcinoma

SD

Standard Deviation

SDS-PAGE

Sodium Dodecyl Sulfate-Polyacrylamide Gel Electrophoresis

SiHa

Human Cervical Squamous Cell Carcinoma Cell Line

TBST

Tris-Buffered Saline with Tween-20

TEM

Transmission Electron Microscopy

UPR

Unfolded Protein Response

Y2H

Yeast Two-Hybrid

4-PBA

4-Phenylbutyric Acid

Author contributions

H. Wang performed and analyzed most of the experiments; J. Zheng provided the human tissue samples; H.Wang, Y.Li, J.Zheng, J.Wang, J.Duan, Y.Wang performed some experiments; H. Wang designed the research and wrote the manuscript; Q.Yang, X.Qu, M.Liu reviewed and commented on the manuscript and corrected the language.

Funding

This work was primarily supported by by Guangdong Basic and Applied Basic Research Foundation, grant number 2023B1515230010 (to Mubiao Liu); Zhuhai People’s Hospital Introduced Talents Research Initiation Program, grant number KYQD2022002 (to Mubiao Liu); Natural Science Foundation of Hunan Province Grant 2025JJ50625, National Natural Science Foundation of China (NSFC) Grant 81570026 (to Xiangping Qu).

Data availability

The data supporting the findings of this study are included within the article. The raw data supporting the findings are available from the corresponding author, upon reasonable request.

Declarations

Ethics approval

All animal use was approved by the Institutional Animal Care and Research Advisory Committee of Central South University (Approval No. XMSB-2022-0003). All clinical studies were approved by the Institutional Ethics Committee of the School of Basic Medical Sciences of Central South University (Approval No. 2022-KT118).

Competing interests

The authors have declared that no conflict of interest exists.