Abstract

   Cervical cancer, primarily squamous cell carcinoma, is the most
   prevalent gynecologic malignancy. Organoids can mimic tumor development
   in vitro, but current Matrigel inaccurately replicates the
   tissue-specific microenvironment. This limitation compromises the
   accurate representation of tumor heterogeneity. We collected
   para-cancerous cervical tissues from patients diagnosed with cervical
   squamous cell carcinoma (CSCC) and prepared uterine cervix
   extracellular matrix (UCEM) hydrogels. Proteomic analysis of UCEM
   identified several tissue-specific signaling pathways including human
   papillomavirus, phosphatidylinositol 3-kinase–AKT, and extracellular
   matrix receptor. Secreted proteins like FLNA, MYH9, HSPA8, and EEF1A1
   were present, indicating UCEM successfully maintained cervical
   proteins. UCEM provided a tailored microenvironment for CSCC organoids,
   enabling formation and growth while preserving tumorigenic potential.
   RNA sequencing showed UCEM-organoids exhibited greater similarity to
   native CSCC and reflected tumor heterogeneity by exhibiting
   CSCC-associated signaling pathways including virus protein-cytokine,
   nuclear factor κB, tumor necrosis factor, and oncogenes EGR1, FPR1, and
   IFI6. Moreover, UCEM-organoids developed chemotherapy resistance. Our
   research provides insights into advanced organoid technology through
   native matrix hydrogels.
     __________________________________________________________________

   Cervical cancer organoids, integrating cervix-specific
   microenvironment, more closely mimic patients’ actual pathological
   state.

INTRODUCTION

   Cervical cancer is the most common gynecological malignancy and a
   leading cause of cancer-related death in women, particularly in
   countries with low development indices ([56]1, [57]2). Although human
   papillomavirus (HPV) has been identified as the primary cause of
   cervical cancer, and worldwide efforts are being made to eradicate the
   disease through HPV vaccination and early screening, a notable number
   of patients are still being diagnosed each year. Cervical squamous cell
   carcinoma (CSCC) is the predominant tissue type of cervical cancer,
   accounting for more than 70% ([58]3). Surgery, radiotherapy, and
   comprehensive chemotherapy offer limited success in curing patients and
   often come with significant side effects and risk of complications,
   making precise medical treatment elusive. Cancer cell lines and tumor
   tissue xenotransplantation models commonly used in research fail to
   accurately replicate the native pathological state of human CSCC in
   vitro due to the absence of the specific microenvironment of human
   cervical tissue. Therefore, reproducing tumor occurrence in vitro
   remains challenging ([59]4, [60]5). Consequently, these limitations
   hinder drug development trials and the realization of precision
   medicine.

   Organoids are three-dimensional microorganisms that closely resemble
   the structure of their source organs and can partially replicate the
   physiological and pathological characteristics of native tissues,
   making them a powerful pathological model for tumor-related studies
   ([61]6, [62]7). Currently, the culture of organoids heavily relies on
   Matrigel, a substance derived from the extract of basal membrane
   proteins present in EHS mouse osteosarcoma. Laminin is the major
   component of Matrigel ([63]8). Nevertheless, Matrigel has a number of
   limitations in accurately reflecting human pathology: (i) Due to the
   distinct microenvironments required for the growth of different tumor
   cells, Matrigel fails to effectively provide tissue-specific
   microenvironments for organoids from different tissues, and there are
   also species differences ([64]8, [65]9). (ii) In terms of biochemical
   composition, Matrigel lacks the high proportion of collagen fibers
   found in CSCC tumor tissues ([66]10). These factors collectively result
   in a deficiency in representing tumor heterogeneity and an inadequate
   reflection of individual differences. Notably, changes in extracellular
   matrix (ECM) properties have a profound effect on cell phenotype in
   vivo. Organoids constructed using ECM derived from native tissue
   exhibit a greater similarity to the real pathological and physiological
   state in vivo ([67]11, [68]12). For instance, Saheli et al. ([69]13)
   used liver-derived ECM to culture liver organoids, displaying higher
   cell viability and significant up-regulation of hepatocyte-specific
   transcriptional products and functions. The ECM not only provides
   structural support but also contains multiple cytokines and proteins
   from extracellular vesicles (EVs), which are critical for
   reconstructing the physiological structure and function of the source
   tissue. Collagen proteins, proteoglycans, and glycoproteins present in
   brain ECM (BEM) can influence neuronal polarization, migration, axon
   growth, and synaptic development. However, Matrigel lacks these
   directional induction effects ([70]14). In contrast, small intestinal
   organoids cultured in intestinal-derived ECM hydrogels are not only
   capable of forming small intestinal structures, such as crypts and
   villi, but also have a significantly higher proportion of crypt stem
   cells compared to those cultured in Matrigel ([71]15). Kim et al.
   ([72]16) also demonstrated that the ECM derived from the small
   intestine is better suited for the growth of small intestinal
   organoids. Furthermore, gastrointestinal organoids cultured in
   decellularized hydrogels of gastrointestinal tissues exhibit high
   viability, whereas significant cell death is observed in decellularized
   hydrogels derived from other tissues like skin, lymph, heart, and
   muscle. This result highlights the tissue-specific effects of the ECM
   microenvironment on organoid development ([73]17). There exists a
   process of mutual adaptation between tumor cells and the surrounding
   environment, which hinges on the direct contact between tumor cells and
   matrix components, as well as biochemical signaling factors in the
   microenvironment. The ECM in various tissues has tissue-specific
   structural components and signaling factors that regulate cell
   adhesion, migration, proliferation, and differentiation, leading to the
   guidance of cellular activity processes and development. Thus, in
   research, it is imperative to evaluate tumors as complete organs rather
   than merely studying them as individual tumor cells ([74]18, [75]19).
   Research models that disregard the tissue-specific biological
   microenvironment of tumors fail to accurately reflect tumor
   heterogeneity. All of these factors highlight the significance of
   tissue-specific ECM microenvironments for organoid culture and also
   suggest that Matrigel cannot serve as a universal substitute for the
   ECM. The use of ECM materials derived from native tissues to construct
   native tumor organoids proves to be more conducive to replicating the
   biological behavior of tumor cells and more closely resembling the real
   pathological and physiological states in vivo.

   In this study, we collected para-cancer cervical tissue from patients
   diagnosed with CSCC and used it to prepare uterine cervix ECM (UCEM)
   hydrogels. UCEM exhibited specific proteomic characteristics of
   cervical tissue, preserving the structural components of cervical ECM,
   as well as signaling pathways and secreted proteins associated with
   carcinogenesis. Our experimental findings demonstrated that
   UCEM-induced transcriptional characteristics of CSCC organoids closely
   resembled those of the patients’ native CSCC tissues, retaining a
   higher presence of oncogenes and signaling pathways linked to CSCC.
   This similarity with native CSCC proved advantageous in capturing the
   tumor heterogeneity of CSCC ([76]Fig. 1).

Fig. 1. The construction of CSCC organoids using UCEM hydrogels enables the
representation of tumor heterogeneity.

   [77]Fig. 1.
   [78]Open in a new tab

   UCEM is prepared through the decellularization technique, containing
   both the structural components (collagen, fibronectin, and laminin) and
   functional components (cytokines and secreted proteins involved in
   cellular activities) of the cervical extracellular matrix. This
   composition preserves the specific microenvironment of cervical tissue.
   On the basis of this foundation, CSCC organoids were constructed using
   UCEM. Consequently, UCEM influences the metabolism of tumor cells
   within CSCC organoids, leading to the promotion of CSCC oncogene
   expression. Moreover, UCEM is partially associated with the resistance
   of organoids to chemotherapy drugs. The in vitro drug response of
   UCEM-organoids is correlated with the clinical response observed in
   patients. In summary, organoids cultured within UCEM closely resemble
   the pathophysiological status of cervical cancer tissue in vivo, thus
   effectively reflecting the tumor heterogeneity in CSCC. EMT,
   epithelial-mesenchymal transition; RNA-seq, RNA sequencing.

RESULTS

Decellularization and gelation of para-cancer cervical tissue

   A hydrogel of human uterine cervical ECM was prepared using
   decellularization techniques. Following our previous reports, the
   cervical tissue was minced and decellularized, followed by
   sterilization, lyophilization, and grinding into a powder. Digestion
   with hydrochloric acid pepsin yielded a pre-gel. UCEM was obtained
   through neutralization with sodium hydroxide ([79]Fig. 2A). The
   synthesized UCEM exhibited high transparency and reduced impurities.
   After incubation at 37°C for 20 to 30 min, they solidified into
   hemispherical droplet-shaped structures, providing a reliable
   three-dimensional environment for cell growth. Analysis through
   hematoxylin and eosin (H&E) staining, Masson staining, and
   immunofluorescence demonstrated the removal of most cellular components
   (CCs) after decellularization while preserving crucial ECM components,
   such as type I and IV collagen fibers, laminin β1, and fibronectin
   ([80]Fig. 2B). Quantification of DNA content revealed a substantial
   decrease from 4048.34 ± 270.72 ng/mg in the cervical tissue before
   decellularization to 20.78 ± 4.98 ng/mg in decellularized tissue,
   indicating a significant reduction in DNA after the decellularization
   process ([81]Fig. 2C). Moreover, the dry weight of the decellularized
   cervix decreased significantly in comparison to the fresh native tissue
   ([82]Fig. 2D). Decellularization led to the removal of intact cells and
   nuclei, eliminating most of the CCs. Further investigation into the
   microstructure of UCEM demonstrated the formation of a dense network
   through cross-linking of collagen fibers ([83]Fig. 2E). At 37°C, UCEM
   exhibited a higher storage modulus (G′) than loss modulus(G″),
   indicating the successful formation of a stable ECM network with high
   viscoelasticity and mechanical strength (fig. S1). These findings
   indicate the successful removal of cell components while retaining key
   ECM structural components, resulting in a relatively intact ECM
   morphology.

Fig. 2. Preparation of the human cervix–derived extracellular matrix hydrogel
(UCEM) by decellularization.

   [84]Fig. 2.
   [85]Open in a new tab

   (A) Preparation process of human cervical derived ECM hydrogel (the
   arrow showed a squamous cell carcinoma of the cervix). (B) Comparison
   of cervical tissue before and after decellularization. Hematoxylin and
   eosin (H&E) staining showed that the acidophilic collagen fibers of the
   cervix after decellularization were stained pink. Masson staining
   showed collagen fibers dyed blue. Immunofluorescence staining showed
   collagen I, collagen IV, fibronectin, and laminin β1 in the cervix
   before and after decellularization. The blue fluorescence represents
   DAPI staining of the cell nuclei. Scale bars, 25 μm. (C) Changes of DNA
   content (in nanograms per milligram) in cervical tissues before and
   after decellularization. (D) Changes in dry weight (in grams) of
   cervical tissue before and after decellularization. (E) Scanning
   electron microscopy showed that collagen fibers were interconnected
   into a network in UCEM. ***P < 0.001.

UCEM preservation of the distinctive cervical ECM Microenvironment

   Mass spectrometry is the primary method for studying proteomics. To
   gain further insights into the proteomics of UCEM, we used mass
   spectrometry to identify and analyze the protein constituents of UCEM
   (data S1). In three biological replicates, a total of 1368 proteins
   were identified, surpassing the number found in Matrigel ([86]Fig. 3A
   and fig. S2A). In addition, UCEM exhibited a higher abundance of human
   cervical tissue–specific proteins compared to Matrigel (fig. S2B). The
   core matrisome analysis revealed the presence of various human ECM core
   matrisome and core matrisome–related proteins in UCEM, including
   collagen, glycoproteins, proteoglycans, ECM regulators, ECM-affiliated
   proteins, and secreted factors (fig. S3A). Functional classification of
   these proteins, based on the clusters of orthologous groups of proteins
   (COG)/eukaryotic orthologous groups (KOG) database, predominantly
   linked them to protein modification, signal transduction, and vesicle
   transport (fig. S4). Further analysis using Gene Ontology (GO)
   enrichment indicated that the proteins within UCEM played a dual role,
   serving as both the structural support of the ECM and participating in
   intracellular metabolism and cell-ECM interactions ([87]Fig. 3E). In
   the biological process (BP) group, protein functions were mainly
   associated with nutrition, energy metabolism, and nucleotide
   metabolism, including processes such as citrate metabolism,
   tricarboxylic acid metabolism, ribonucleoside triphosphate metabolism,
   and purine nucleoside triphosphate metabolism. The CC group
   predominantly concentrated on intracellular and EVs, as well as the
   ECM, exemplified by the cytoplasmic vesicle lumen, secretory granule
   lumen, ECM, and secretory vesicle. The molecular function (MF) group
   primarily engaged in metabolic complex composition and enzyme activity
   regulation, involving functions such as purine nucleoside binding,
   nucleoside phosphate binding, nucleotide binding, carbohydrate
   derivative binding, and nucleoside triphosphatase activity. Moreover,
   Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment analysis
   demonstrated significant enrichment of UCEM protein components in focal
   adhesion, ECM-receptor interaction, PI3K-AKT signaling pathway, and HPV
   ([88]Fig. 3B). These signaling pathways play crucial roles in cell
   adhesion, proliferation, motility, and morbidity. The focal adhesion
   pathway facilitates cytoskeletal reorganization, adhesion, and
   migration, which are vital for cell adhesion and localization ([89]20,
   [90]21). The interaction between ECM and receptors plays a crucial role
   in converting extracellular signals into intracellular BPs and
   regulating the invasion and metastasis of tumor cells ([91]22, [92]23).
   The PI3K-AKT signaling pathway is activated in a wide range of tumors
   and is involved in cell metabolism, protein synthesis, cytoskeletal
   reorganization, as well as the regulation of cell proliferation,
   survival, and differentiation processes ([93]24, [94]25). HPV is
   considered the primary cause of cervical cancer. UCEM preserved the HPV
   infection-related signaling pathway, essential for researching the
   pathogenesis and treatment of cervical cancer caused by HPV. The
   protein abundance analysis revealed that UCEM contained high levels of
   collagen (types I, IV, and VI) and basement membrane components (such
   as proteoglycans and glycoproteins) that closely resembled those found
   in human tissues ([95]Fig. 3C). In addition, a significant number of
   secreted and exosomal proteins exist ([96]Fig. 3D), playing pivotal
   roles in the proliferation, invasion, and metastasis of CSCC. Note that
   UCEM preserved essential structural components of the ECM as well as
   protein components crucial for cellular activities. This preservation
   is advantageous in unraveling the intricate interplay between tumors
   and the ECM. The protein components of UCEM were found to be
   distributed in groups through protein interaction analysis. For
   analysis, we selected five groups that exhibited a high degree of
   interaction. The central proteins identified, as shown in fig. S3B,
   were primarily ECM structural proteins such as VCL, ACTG1, HNRNPK, FN1,
   and HSP90AA1. Fibronectin 1 (FN1) is a key constituent of the ECM,
   contributing to cell adhesion and migration. It plays a role in
   improving the motility of cancer cells and is closely implicated in the
   invasion and metastasis of CSCC ([97]26). The up-regulation of HSP90AA1
   in cervical cancer tissues is specifically linked to the development of
   chemotherapy and radiotherapy resistance ([98]27). These findings
   suggest that UCEM not only contributes to the provision of ECM
   structural components necessary for cell growth but also retains cell
   factors and tissue-specific proteins associated with cellular
   activities such as energy metabolism and signal transduction from
   native tissues. Consequently, UCEM could preserve a cervix
   tissue-specific ECM microenvironment that is absent in other ECM
   sources like Matrigel.

Fig. 3. Mass spectrometry detection and bioinformatics analysis of UCEM.

   [99]Fig. 3.
   [100]Open in a new tab

   (A) The Venn diagram of UCEM proteins identified in three biological
   replicates. (B) Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway
   enrichment analysis of UCEM showed significant enrichment in HPV
   pathways related to the development of cervical cancer, in addition to
   cell adhesion and ECM-receptor interaction. (C) Protein abundance
   analysis revealed structural components of ECM (including collagen,
   proteoglycan, laminin, and fibronectin) in UCEM. (D) Protein abundance
   analysis revealed secretory and exosome proteins in UCEM. (E) UCEM
   protein component Gene Ontology (GO) enrichment analysis. BP,
   biological process; CC, cellular component; MF, molecular function.

Supporting the formation and growth of CSCC organoids with UCEM

   We collected a total of 52 cases of CSCC, and the general information
   of the patients is presented in table S1. This table contains recorded
   information regarding the patient’s age, HPV infection type, clinical
   stage (FIGO 2018), and tumor size. After digestion and filtration, CSCC
   cells were embedded and solidified in Matrigel and UCEM, and then added
   to the culture medium for CSCC organoids. CSCC organoids were observed
   in both UCEM and Matrigel after 7 to 10 days. Tumor cells from 31
   patients self-organized to form CSCC organoids in Matrigel (59.62%) and
   were able to form organoids in UCEM from 34 patients (65.38%). Among
   them, four were only formed in Matrigel, seven were only present in
   UCEM, and the rest were cultured in both mediums ([101]Fig. 4, A and
   B). The success rate of organoid culture in the two groups was analyzed
   using a chi-square test, which showed no statistically significant
   difference (χ^2 = 0.369, P > 0.05). This indicates that UCEM can
   achieve a similar success rate to Matrigel in the organoid culture of
   CSCC. We observed a lower efficiency of organoid formation in UCEM
   compared to Matrigel ([102]Fig. 4C). Then, we compared and analyzed the
   differences in protein components and content between UCEM and Matrigel
   (fig. S5). The results showed that the content of laminin in UCEM was
   significantly lower than that in Matrigel. Laminin may be a crucial
   factor affecting organoid formation efficiency. Laminin is a major
   component of the basement membrane, consisting of three distinct chains
   (α, β, and γ) that provide a structural scaffold for the ECM ([103]28).
   Different combinations of α, β, and γ chains form various laminin
   isoforms. We further analyzed the proteomics of UCEM to clarify the
   possible laminin isoforms present in UCEM. The laminin subunits
   detected in UCEM include LAMA2, LAMA3, LAMA4, LAMA5, LAMB1, LAMB2, and
   LAMC1. The possible laminin isoforms include (fig. S6, A to D)
   laminin2/211, laminin4/221, laminin6/311, laminin7/321, laminin8/411,
   laminin9/421, laminin10/511, and laminin11/521. Dysregulation of
   cell-laminin interactions is a key feature of many cancers ([104]29).
   Certain members of the laminin family, including laminin-521/332/111,
   play more complex roles in tumor survival, cancer cell migration, and
   tumor invasion ([105]30–[106]32). Laminin-521 is involved not only in
   maintaining embryonic stem cells and inducing pluripotent stem cells
   but also in cancer self-renewal and invasion ([107]32). Therefore, we
   supplemented UCEM with varying concentrations of laminin-521
   (rhLaminin-521, Gibco, A29248) and assessed the formation efficiency of
   CSCC organoids. As depicted in [108]Fig. 4C, the efficiency of organoid
   formation in UCEM increased with an elevation in laminin-521
   concentration. However, no further improvement in organoid formation
   efficiency was observed beyond a laminin-521 concentration of 0.1
   mg/ml. Nevertheless, the organoid formation efficiency in UCEM remained
   lower than that in Matrigel, with a statistically significant
   difference (P < 0.01) ([109]Fig. 4D). Laminin-332 can serve as a marker
   for early invasion of cervical cancer and facilitates HPV infection and
   replication ([110]33, [111]34). Laminin-111 is a laminin isoform with a
   high content in Matrigel ([112]9, [113]35) and may be an important
   factor affecting organoid formation. In addition, the expression of
   laminin-111 is increased by 3.8-fold in CSCC compared with normal
   cervical tissue ([114]36). We conducted additional experiments
   targeting laminin isoforms to further investigate the effects of
   laminin-332 and laminin-111 on CSCC organoid formation. The results
   showed that with the increase in laminin-332 and laminin-111
   concentrations, the formation efficiency of CSCC organoids was
   improved, reaching a peak at 0.04 and 0.1 mg/ml, respectively
   ([115]Fig. 4, E to H). These results indicate that UCEM can support the
   formation of CSCC organoids, while laminin can enhance their formation
   efficiency. The simultaneous addition of multiple laminins could
   potentially further enhance the formation efficiency of CSCC organoids,
   possibly reaching the level of Matrigel. Scanning electron microscopy
   further revealed the presence of collagen and other protein components
   surrounding the CSCC organoids. Notably, collagen exhibited a mesh-like
   distribution, while other proteins were dispersed (fig. S7).
   Furthermore, to examine the influence of tissue-specific
   microenvironments on organoid formation, we cultured CSCC organoids
   using ECM hydrogels derived from the human uterine body (UB), porcine
   cervix (PC), and porcine UB (PUB), and compared them to UCEM. The
   findings demonstrated that CSCC organoids could maintain a robust
   morphology in both UCEM and UB, whereas, in PC and PUB, they exhibited
   smaller sizes and struggled to maintain a stable sphere-like structure
   ([116]Fig. 4I). Comparative analysis of organoid formation efficiency
   revealed that UCEM exhibited the highest efficiency in CSCC organoid
   formation ([117]Fig. 4J). These results highlight the influence of the
   specific microenvironment within cervical tissue on the formation of
   CSCC organoids. We further compared the size and morphology of
   organoids cultured in Matrigel and UCEM obtained from the same patients
   ([118]Fig. 4L). Both CSCC organoids cultured in Matrigel and UCEM
   exhibited a dense spherical structure, with the majority ranging
   between 100 and 200 μm in diameter ([119]Fig. 4K). Notably, the
   organoids maintained a consistent size and morphology in both Matrigel
   and UCEM, as indicated by their circularity close to 1 ([120]Fig. 4N).
   A higher circularity value is advantageous for maintaining the
   stability of tumor organoids. Staining results for live/dead cells
   revealed that CSCC organoids retained high activity in both Matrigel
   and UCEM after approximately 1 week of culture, with most cells
   demonstrating robust survival. This suggests that CSCC cells can
   successfully adapt to the biological microenvironment of UCEM and
   sustain their activity ([121]Fig. 4, L and M). In addition, we
   conducted dynamic observations and comparisons of the growth process of
   organoids in both UCEM and Matrigel. The results demonstrated that,
   similar to Matrigel, CSCC organoids cultured in UCEM maintained a
   spherical structure and smooth morphology. Furthermore, the size of
   organoids in UCEM and Matrigel remained comparable throughout the
   growth process, with no statistically significant difference observed
   (P > 0.05; fig. S8A). Following the passage, organoids cultured in UCEM
   and Matrigel exhibited similar properties. Notably, cryopreserved CSCC
   organoids in UCEM could be successfully revived, regrown, and amplified
   (fig. S8B). These revived organoids retained a regular spherical
   structure and stable morphology, which is highly advantageous for
   long-term organoid culture and the establishment of a CSCC biobank. To
   investigate whether CSCC organoids cultured in UCEM and Matrigel
   maintained the structural and phenotypic characteristics of CSCC
   tissues, we performed H&E staining and immunohistochemical staining of
   multiple markers ([122]Fig. 5) ([123]37). Histological analysis showed
   that in both UCEM and Matrigel, the arrangement of cells in CSCC
   organoids was disordered, with loss of polarity, increased nuclear
   volume, deep staining, atypia, and disordered ratio of nuclei to
   cytoplasm. The basal cell–restricted marker TP63 and the cell
   proliferation marker KI67 were positive, indicating active cell
   division. The marker for persistent HPV infection, P16INK4α, was also
   positive, while the expression level of the cell differentiation marker
   KRT13 was low. These findings are consistent with the histological and
   immunophenotypic characteristics of CSCC and similar to those observed
   in native CSCC tissues. In conclusion, our experimental results suggest
   that UCEM can support the formation and growth of CSCC organoids and
   that the microenvironment specific to cervical tissue can influence the
   formation of these organoids. Furthermore, CSCC organoids can adapt to
   the ECM microenvironment provided by UCEM. UCEM enables a long-term
   culture of CSCC organoids without the need for Matrigel while
   preserving the structural and phenotypic characteristics of CSCC.

Fig. 4. CSCC organoids cultured in UCEM and Matrigel and comparison.

   [124]Fig. 4.
   [125]Open in a new tab

   (A) Comparison of the success rate of CSCC organoid culture between
   UCEM and Matrigel. (B) Culture of CSCC organoids in UCEM and Matrigel.
   M represents Matrigel, U represents UCEM, and HPV represents human
   papillomavirus. (C) CSCC organoids cultured in UCEM with the addition
   of different concentrations of laminin-521. Scale bars, 100 μm. (D)
   Effect of different concentrations of laminin-521 on the efficiency of
   organoid formation in UCEM. (E) CSCC organoids cultured in UCEM
   supplemented with different concentrations of laminin-332. Scale bars,
   100 μm. (F) Effect of different concentrations of laminin-332 on
   organoid formation efficiency in UCEM. (G) CSCC organoids cultured in
   UCEM supplemented with different concentrations of laminin-111. Scale
   bars, 100 μm. (H) Effect of different concentrations of laminin-111 on
   organoid formation efficiency in UCEM. (I) CSCC organoids cultured in
   ECM hydrogels from different tissues. Scale bars, 100 μm. (J)
   Comparison of organoid formation efficiency in ECM hydrogels from
   different tissues. UCEM, uterine cervix extracellular matrix; UB,
   uterine body; PC, porcine cervix; PUB, porcine uterine body. (K)
   Comparison of organoids diameters in UCEM and Matrigel; P > 0.05. (L)
   CSCC organoids cultured in UCEM and Matrigel and live/dead cell
   staining of organoids. Scale bars, 100 μm. PI, propidium iodide. (M)
   Comparison of the percentage of living cells in CSCC organoids cultured
   in UCEM and Matrigel; P > 0.05. (N) Comparison of circularity of CSCC
   organoids in UCEM and Matrigel (the closer the ratio of the short axis
   to the long axis is to 1, the more circular the morphology). n.s., not
   significant.*P < 0.05; **P < 0.01; ***P < 0.001.

Fig. 5. Structural and phenotypic characteristics of native tumor tissue and
CSCC organoids.

   [126]Fig. 5.
   [127]Open in a new tab

   H&E staining and immunohistochemical staining of CSCC organoids in UCEM
   and Matrigel showed that KI67, P63, and P16 were positive and KRT13 was
   weakly positive. Scale bars,100 μm.

Closer transcriptional resemblance of CSCC organoids in UCEM to native tissue

   To further investigate the transcriptomic characteristics of CSCC
   organoids cultured in UCEM, mRNA sequencing was performed on native
   CSCC tissues, as well as CSCC organoids cultured in Matrigel and UCEM
   from three patients (data S2). In three biological replicates, the mRNA
   coexpression in CSCC tissues, organoids cultured in Matrigel and UCEM
   exceeded 1 × 10^4 in each group ([128]Fig. 6A). This indicates that
   both organoid culture methods, Matrigel and UCEM, effectively preserve
   the transcriptomic characteristics of native CSCC tissues. We acquired
   transcriptomic data of normal cervical squamous epithelial tissue from
   healthy women through the Genotype-Tissue Expression (GTEx) database
   (GTEX-S32W-1526-SM-4AD6Z, GTEX-TSE9-2726-SM-4DXSQ,
   GTEX-S4UY-1426-SM-4AD6Y, GTEX-TSE9-2826-SM-4DXTF) and performed
   principal components analysis on normal cervical squamous epithelial
   tissue, CSCC tissue, and organoids in Matrigel and UCEM. As shown in
   [129]Fig. 6B, CSCC tissues and organoids were distinct from normal
   cervical squamous epithelium tissue, indicating significant differences
   between cervical cancer tissues and normal tissues. In tumor tissues,
   CSCC organoids cultured in Matrigel diverged from native CSCC tissues,
   while CSCC organoids cultured in UCEM showed closer similarity to
   native CSCC tissues, with some overlap. The heatmap of differentially
   expressed genes also revealed differences between CSCC organoids
   cultured in Matrigel and UCEM (fig. S9A), indicating that UCEM
   influenced the vital activities of CSCC organoids to some extent.
   Building upon this, we further analyzed the differentially expressed
   genes (fold change >2) between CSCC organoids cultured in Matrigel and
   UCEM. In comparison to Matrigel, 229 genes were significantly
   up-regulated and 70 genes were down-regulated in CSCC organoids
   cultured in UCEM (fig. S9B). GO enrichment analysis of the differential
   genes ([130]Fig. 6D) exhibited significant enrichment of BPs (GOBP)
   related to immune response, signal transduction, and cell adhesion. GO
   cellular component (GOCC) was notably concentrated in the extracellular
   region, which reflects the remodeling of the ECM by tumor cells and
   explains how tumor evolution can provide unique niches for cancer cells
   to engage in altered ECM remodeling. GO molecular function (GOMF) was
   primarily associated with chemokines, cytokine activity regulation,
   receptors, and carbohydrate binding. Transcription of pro-inflammatory
   genes indicates the recruitment of tumor-related immune cells in vivo.
   KEGG enrichment analysis ([131]Fig. 6E) demonstrated significant
   enrichment of signaling pathways involving viral protein-cytokine
   interaction and chemokine signaling. This could help elucidate the role
   of HPV infection and E6 and E7 virus-associated proteins in the
   development of cervical cancer. In addition, organoids cultured in UCEM
   showed significant enrichment in tumor-related signaling pathways such
   as nuclear factor κB (NF-κB) and tumor necrosis factor (TNF) signaling
   pathways. TNF-α activates NF-κB activity and regulates the expression
   of a series of tumor-associated genes ([132]38, [133]39). The NF-κB
   signaling pathway is normally transient and tightly regulated in normal
   cells. However, in cervical cancer, it becomes activated and plays a
   crucial role in tumor initiation, progression, and metastasis ([134]40,
   [135]41). Furthermore, we observed that several of the up-regulated
   genes were associated with cancer-promoting effects (fig. S9B).
   Building upon this, we selected well-known oncogenes implicated in
   cervical cancer and conducted a comparative analysis. The results
   indicated that the expression levels of various oncogenes involved in
   cervical cancer development and progression were higher in organoids
   cultured in UCEM compared to Matrigel, and they closely resembled those
   observed in CSCC-derived tissues ([136]Fig. 6C). Not only did these
   genes play a role in the proliferation, invasion, and metastasis of
   cervical cancer cells [NEDD9 ([137]42), EGR1 ([138]43), FPR1 ([139]44),
   AIF1 ([140]45), and EBI3 ([141]46)] but they are also involved in the
   activation of immune cells and tumor-related immune regulation in CSCC
   tissues [RGS1 ([142]47) and TLR10 ([143]48)]. Moreover, IFI6 expression
   was associated with the expression of HPV16 E7 oncoprotein ([144]49).
   The transcriptome sequencing results demonstrated that both UCEM and
   Matrigel-cultured CSCC organoids retained the transcriptomic
   characteristics of CSCC tissues similarly. However, note that
   UCEM-cultured organoids outperformed Matrigel-cultured organoids in
   preserving multiple oncogenes and signaling pathways associated with
   cervical cancer, making them more akin to patient-derived CSCC tissues.
   These findings also suggested that the tissue-specific microenvironment
   might stimulate the transcription of oncogenes in primary tumor
   tissues.

Fig. 6. Transcriptomic analysis of CSCC organoids cultured in UCEM and
Matrigel.

   [145]Fig. 6.
   [146]Open in a new tab

   (A) Venn diagram of transcriptomic sequencing of three biological
   replicates (CC1T, CC1M, and CC1U; CC2T, CC2M, and CC2U; CC3T, CC3M, and
   CC3U). (B) Principal components analysis (PCA) of normal cervical
   squamous epithelium tissue from healthy female individuals (NO),
   primary tumor tissue (T), CSCC organoids cultured in UCEM (U), and
   Matrigel (M). (C) Comparison of highly expressed genes in CSCC between
   organoids and native tumor tissues, using LSD for multiple comparisons.
   (D) GO enrichment analysis of differentially expressed genes in CSCC
   organoids cultured in UCEM and Matrigel. (E) KEGG pathway enrichment
   analysis of differentially expressed genes in CSCC organoids cultured
   in UCEM and Matrigel.

Chemotherapy resistance in UCEM-Cultured CSCC organoids

   In vitro drug trials were conducted to evaluate the response of CSCC
   organoids cultured in UCEM and Matrigel to chemotherapy drugs. The
   combination of paclitaxel and carboplatin (TC) is a commonly used
   chemotherapy regimen for CSCC. We compared the in vitro drug
   sensitivity test results with the clinical treatment response of a
   patient with advanced CSCC (2018 FIGO, stage IIB). CSCC organoids were
   cultured using patient-derived tumor tissue obtained from a biopsy and
   subjected to the in vitro drug sensitivity test. Following treatment
   with 1 μM paclitaxel combined with 10 μM carboplatin for 3 days,
   dissociation of CSCC organoids was observed ([147]Fig. 7B). This
   finding supports the effectiveness of the TC chemotherapy regimen for
   this patient. Furthermore, radiographic examinations revealed a
   significant reduction in tumor size after treatment with the TC regimen
   compared to before ([148]Fig. 7C). These results demonstrate the
   concordance between the treatment effect observed in our organoid model
   in vitro and the clinical response of the patient, which can guide the
   screening of chemotherapy drugs. In addition, the interaction between
   tumor cells and the specific microenvironment of surrounding tissues
   has a notable influence on tumor progression and the response to
   chemotherapy drugs. The use of UCEM to culture CSCC organoids allows
   for the establishment of a connection between the cervical
   tissue–specific microenvironment and cervical squamous cell cancer
   cells. Building upon this foundation, we compared the response to
   chemotherapy drugs in CSCC organoids cultured by UCEM and Matrigel
   under varying concentrations of carboplatin ([149]Fig. 7A). Live/dead
   staining of the organoids revealed the cytotoxic effect of carboplatin
   at different concentrations ([150]Fig. 7E). Following carboplatin
   treatment, the outer cells of the CSCC organoids perished first. As the
   concentration of carboplatin reached 100 μM, the growth of the
   organoids was significantly inhibited, resulting in a reduction in
   diameter. Consistently, there was a dose- and time-dependent decrease
   in organoid viability and proliferation in response to chemotherapy
   drugs ([151]Fig. 7D). Extended exposure to chemotherapy drugs,
   accompanied by an increase in concentration, led to the gradual
   development of resistance to carboplatin in CSCC organoids cultured by
   UCEM, consequently resulting in higher proliferation activity compared
   to Matrigel. Next, we conducted a comparative analysis between in vitro
   drug responses of tumor organoids from 10 patients with CSCC and the
   clinical responses observed in matched patients. The clinical response
   of patients was assessed using RECIST criteria (version 1.1) through
   the evaluation of imaging changes before and after chemotherapy
   ([152]50). The radiological presentation and in vitro drug sensitivity
   testing of patients resistant to the TC regimen were shown in fig. S10
   (A and B). To test the drug response, we cultured CSCC organoids using
   UCEM and Matrigel and exposed them to TC regimens in vitro. The drug
   response curves were then plotted (fig. S10, C to F). Overall, both
   UCEM and Matrigel cultured CSCC organoids exhibited a strong predictive
   capability for patients’ response to chemotherapy drugs ([153]Fig. 7F).
   Notably, the IC[50] (median inhibitory concentration) values of poor
   response patients (patients 6, 8, and 9) were significantly higher than
   those of good response patients (patients 1, 2, and 5) ([154]Fig. 7G
   and data S3). Moreover, we observed that, in poor response patients,
   CSCC organoids cultured using UCEM displayed higher IC[50] values
   compared to those cultured using Matrigel ([155]Fig. 7G), suggesting
   that CSCC organoids cultured using UCEM exhibited more pronounced drug
   resistance and have the potential for accurate identification of
   drug-resistant patients. This phenomenon can be ascribed to the
   influence of the specific microenvironment of cervical tissue on
   chemotherapy drugs, alongside factors like the heightened protective
   effect of collagen fiber deposition on tumor cells. In addition, it is
   related to tumor-associated malignant biological behaviors in CSCC
   organoids cultured by UCEM, such as the enrichment of the NF-κB
   signaling pathway. Therefore, CSCC organoids cultured by UCEM hold the
   potential for clinical screening of antitumor drugs and serve as
   valuable tools for investigating the role of tissue-specific
   microenvironment in the development of tumor drug resistance. Moreover,
   this suggests that a combined therapy approach targeting
   tissue-specific microenvironments may offer patients the possibility of
   achieving deeper remission.

Fig. 7. Drug assay of CSCC organoids cultured by UCEM and Matrigel.

   [156]Fig. 7.
   [157]Open in a new tab

   (A) Schematic diagram of CSCC organoids treated with different
   concentrations of carboplatin. (B) The TC chemotherapy regimen induces
   dissociation of CSCC organoids in vitro. Scale bars, 100 μm. (C) The
   radiographic examinations (computed tomography, CT) of patients with
   CSCC after treatment with the TC regimen. The yellow arrow indicates
   the uterine artery. The red lines indicate changes in tumor size at the
   same plane (uterine artery level) before and after treatment.
   Pre-treatment: 48.55 mm, post-treatment: 26.59 mm. (D) Changes of CSCC
   organoid viability after carboplatin treatment with different
   concentrations. (E) Live/dead staining of CSCC organoids after
   treatment with different concentrations of carboplatin. Scale bars, 100
   μm. (F) For patients’ clinical response, blue represents sensitivity to
   TC regimen treatment, and yellow indicates resistance. For drug
   sensitivity testing using CSCC organoids in vitro, green indicates
   sensitivity to the TC regimen, and red indicates resistance. (G)
   Comparison of semi-inhibitory concentration (Log[10]IC[50]) in drug
   sensitivity testing of CSCC organoids cultured in UCEM and Matrigel.

Preservation of CSCC tissue characteristics through xenotransplantation of
organoids in UCEM

   To determine whether CSCC organoids cultured in UCEM can regenerate
   tumor tissues in vivo, we implanted the CSCC organoids cultured in UCEM
   and Matrigel, obtained from the same patients, into the subcutaneous
   region of nude mice. Each group consisted of 10 nude mice. After 1.5
   months following implantation, the nude mice were euthanized and the
   subcutaneous tumors were excised for comparison. The results depicted
   in [158]Fig. 8A indicated that subcutaneous tumors formed in six
   (60.00%) nude mice in the UCEM group and seven (70.00%) nude mice in
   the Matrigel group, respectively. Notably, there was no statistically
   significant difference in the success rate of tumor formation between
   the two groups. Among these mice, two exclusively developed
   subcutaneous tumors in the Matrigel group, one solely formed
   subcutaneous tumors in the UCEM group, and five developed subcutaneous
   tumors in both groups ([159]Fig. 8, B and C). Blood vessels were
   observed surrounding the subcutaneous tumor tissues in both groups
   ([160]Fig. 8D and fig. S11B). We conducted a comparison of the
   co-occurrence of subcutaneous tumors in the two groups ([161]Fig. 8F
   and fig. S11A) and performed a statistical analysis using the paired
   sample t test. The UCEM group exhibited an average volume of (54.47 ±
   21.07) mm^3 with a mass of (52.32 ± 23.74) mg, while the Matrigel group
   displayed an average volume of (44.70 ± 31.87) mm^3 with a mass of
   (31.84 ± 16.91) mg. However, no statistically significant difference in
   the volume and mass of the subcutaneous tumors was observed between the
   two groups (P > 0.05; [162]Fig. 8, G and H). H&E staining and
   immunohistochemical staining revealed that the subcutaneous tumor
   tissues formed by organoids in both groups resembled CSCC tissues, as
   indicated by the expression of the cell proliferation marker KI67 and
   the HPV virus infection marker P16INK4a ([163]Fig. 8E). Furthermore,
   UCEM exhibited a higher presence of scattered tumor cells with positive
   expression of KI67 and P16INK4a, potentially attributed to the
   migration and proliferation of individual tumor cells in vivo. These
   observations served as confirmation of the proteomic characteristics of
   UCEM and the transcriptomic characteristics of CSCC organoids. The
   findings demonstrated that CSCC organoids cultured in UCEM maintained
   their tumorigenic properties in vivo and exhibited similarity to those
   cultured in Matrigel. UCEM partially replicated the migration behavior
   of tumor cells in the uterine cervical tissue–specific
   microenvironment. Notably, despite being a human-derived ECM material,
   UCEM did not diminish the tumorigenicity of CSCC organoids in
   immunodeficient mice.

Fig. 8. Detection of tumorigenic ability of CSCC organoids in vivo.

   [164]Fig. 8.
   [165]Open in a new tab

   (A) Comparison of tumor formation rates between CSCC organoids cultured
   in UCEM and Matrigel in nude mice. (B) Tumor formation in vivo of
   UCEM-organoids and Matrigel-organoids derived from the same patient
   inoculated into two nude mice with the same number. (C) Simultaneous
   subcutaneous tumor formation of CSCC organoids cultured in UCEM and
   Matrigel in nude mice. (D) Subcutaneous tumors and surrounding blood
   vessels. (E) H&E and immunohistochemical staining of subcutaneous
   tumors in both groups. Scale bars, 100 μm. (F) Subcutaneous tumors
   formed by CSCC organoids cultured in UCEM and Matrigel. (G) Comparison
   of the tumor weight of subcutaneous tumors developed by organoids in
   UCEM and Matrigel, the difference was not statistically significant, P
   > 0.05. (H) Comparison of the tumor volume of subcutaneous tumors
   developed by organoids in UCEM and Matrigel, the difference was not
   statistically significant, P > 0.05.

DISCUSSION

   The monolayer culture of CSCC tumor cells and xenotransplantation
   animal models is insufficient in replicating the authentic pathological
   state in humans, thereby limiting research on tumor pathogenesis, as
   well as the development of more effective therapies and drugs. The ECM
   is a complex structure composed of various proteins that play a crucial
   role in maintaining tissue function, and morphology and regulating cell
   phenotype, proliferation, differentiation, adhesion, motility, and
   signal transduction. The ECM is also a significant component of the
   tumor microenvironment. By providing cancer stem cell niches and
   regulating tumor cell growth, the ECM controls cell signaling essential
   to maintain tumor tissue ([166]51, [167]52). Tumor progression causes
   ECM remodeling, leading to the interaction between reconstructed ECM
   and tumor cells. This interaction activates cell migration and
   proliferation while inhibiting cell apoptosis, thereby maintaining
   tumor tissue homeostasis ([168]53). As a result, ECM materials not only
   serve as structural support for three-dimensional culture but also have
   significant biological activities. However, neither Matrigel, the most
   commonly used material, nor complex natural or synthetic chemicals can
   offer a tailored microenvironment for human cervical tissue, despite
   their applicability to organoid three-dimensional cultures. On one
   hand, the complex structure of the microenvironment and the wide
   variety of bioactive factors make it challenging to replicate
   artificially. On the other hand, the presence of numerous unexplored
   components also contributes to the difficulty. Previous studies have
   shown that supplementing the culture medium with decellularized
   hydrogel derived from porcine endometrium can enhance the proliferation
   of human endometrial organoids ([169]54). However, this method does not
   eliminate the potential impact of species variations on organoids. UCEM
   preserves not only the uterine cervical tissue–specific ECM
   microenvironment found in patients with CSCC but also eliminates
   potential species variations. UCEM-based CSCC organoids more accurately
   reflect the pathophysiological characteristics and tumor heterogeneity
   of patients with CSCC, having higher research value.

   In cancer organoid research, we aimed to address the challenges posed
   by species differences and the absence of a tissue-specific
   microenvironment in ECM materials. To overcome these challenges, we
   investigated the use of decellularized ECM hydrogel derived from the
   para-cancerous cervix as a scaffold for the growth of CSCC organoids
   and compared it to the commonly used Matrigel. Our experiment involved
   decellularizing the cervix of patients with CSCC to obtain the ECM. We
   verified the removal of cell components and the retention of ECM
   components, such as collagen and laminin, using H&E staining and Masson
   staining. This decellularization method is well-established, safe, and
   effective and enables the preservation of the ECM in its entirety. The
   solidified UCEM provides the necessary anchoring and support for
   organoid formation.

   Matrigel comprises primarily glycoproteins, constituting more than 95%
   of its composition, whereas collagen fibers and proteoglycans make up
   only 0.4 and 1%, respectively ([170]17). This starkly contrasts with
   human tissues, where collagen serves as the primary component
   ([171]55). The composition of UCEM includes a rich presence of type VI,
   I, and IV collagen fibers, aligning more closely with the structural
   characteristics of human tumor tissues. Collagen fibers also play a key
   role in the progression of cervical cancer. The elevated expression of
   type VI collagen fibers in cervical cancer tissues significantly
   correlates with tumor stage and prognosis and influences the extension,
   penetration, and invasion of tumor cells ([172]56). Moreover, the
   expression level of type I collagen fibers is considerably higher in
   cervical cancer tissues compared to normal tissues. COL1A1, a component
   of type I collagen, inhibits the apoptosis of cervical cancer cells and
   supports tumor survival ([173]57). Type IV collagen plays a critical
   role in tumor invasion and metastasis. As malignancy increases,
   collagen fibers gradually decline in the basement membrane surrounding
   cervical invasive carcinoma, and adjacent cancer tissues ([174]36).
   This alteration in collagen fiber composition creates favorable
   conditions for tumor invasion and metastasis. Considering that
   organoids represent a dynamic process involving the proliferation and
   differentiation of tumor stem cells, UCEM effectively replicates the
   dynamic changes of collagen fibers during the transition of tumor
   tissue from low-grade malignancy to invasion, metastasis, and
   progressive deterioration. In summary, an ECM lacking collagen fibers
   is inherently incomplete. Laminin is a critical cell adhesion molecule
   that plays a crucial role in determining the structure and
   physiological function of the basement membrane ([175]58).
   Laminin-cancer cell interactions play a critical role in tumor invasion
   and metastasis. Laminin promotes tumor cell proliferation and
   dissemination via diverse mechanisms ([176]28). Our experimental
   results also indicated that laminin enhances the formation efficiency
   of CSCC organoids. These ECM structural components not only contribute
   to the formation of the spatial structure but also play a role in the
   pathogenesis and progression of cervical cancer.

   EV proteins play a role in the BPs of tumorigenesis and progression.
   CD81, one of the markers for exosomal proteins ([177]59), is found in
   UCEM. Our results revealed that UCEM is enriched with EV proteins that
   facilitate intercellular communication. Most of these proteins were
   significantly overexpressed in cervical cancer and associated with
   malignant biological behaviors. The retention of EV proteins in UCEM
   benefits the reflection of tumor heterogeneity and is essential for
   studying the pathogenesis and treatment of cervical cancers. FLNA
   interacts with transmembrane proteins and signaling molecules to
   coordinate cell adhesion and migration, while also participating in DNA
   damage and repair. Moreover, FLNA is significantly overexpressed in
   cervical cancer tissues ([178]60, [179]61). MYH9 was found to be highly
   abundant in exosomes and exhibited abnormal expression in various tumor
   tissues. Endogenous overexpression of MYH9 has been shown to increase
   tumor cell proliferation and metastasis and induce
   epithelial-mesenchymal transition, thereby promoting cancer progression
   ([180]62, [181]63). The expression of EEF1A1 remains highly stable in
   the normal cervix, cervical intraepithelial neoplasia, and invasive
   carcinoma and is involved in protein synthesis ([182]64). Moesin (MSN)
   plays a crucial role in cell-cell and cell-ECM interactions,
   influencing cell adhesion, polarity, and migration, as well as tumor
   cell invasion and metastasis. In addition, high expression of MSN
   correlates strongly with cervical cancer malignancy and metastasis
   ([183]65, [184]66). PKM plays a pivotal role as a glycolytic enzyme
   that drives the conversion of glucose to lactic acid. The up-regulation
   of PKM expression in cervical cancer has a significant impact on the
   invasive, metastatic, and other malignant transformations of cervical
   cancer cells ([185]67, [186]68). Elevated levels of HSPA8 in cervical
   cancer significantly correlated with an unfavorable prognosis ([187]69,
   [188]70). ITGB1, a crucial integrin subunit, is essential for mediating
   cell-ECM interactions, thereby contributing to the formation of focal
   adhesions ([189]71). Furthermore, various other proteins present in
   EVs, such as ANXA2 ([190]72, [191]73), TLN1 ([192]74), ENO1 ([193]75),
   and TFRC ([194]76) exhibit substantial overexpression in cervical
   cancer. These proteins play a role in cell adhesion and movement,
   regulating tumor cell proliferation and metastasis, and are also linked
   to patient prognosis. In summary, UCEM not only preserves the
   structural components of the ECM but also maximizes the retention of
   the cervical tissue–specific ECM microenvironment, aiding in the
   restoration of the tumor cell niche in vivo. Preserving the retention
   of these bioactive proteins is advantageous for maintaining the
   heterogeneity of tumor tissue.

   The results of isolating and culturing organoids demonstrated that UCEM
   effectively supports the formation and growth of CSCC organoids. The
   size, morphology, and proportion of living cells in organoids cultured
   in UCEM were comparable to those in Matrigel. Furthermore, even after
   subculture, cryopreservation, and resuscitation, the CSCC organoids
   maintained stable morphology and structure in UCEM, indicating their
   suitability for long-term organoid culture and biobank construction. In
   addition, CSCC organoids cultured in both UCEM and Matrigel retained
   the structure and phenotypic characteristics of CSCC tissues, as well
   as demonstrated similar tumorigenicity in vivo. Consequently, UCEM
   emerges as a viable alternative to Matrigel for CSCC organoid culture.
   In cancer, signals induced by the ECM influence cell phenotype and
   function. Notably, at the transcriptional level, the expression levels
   of multiple oncogenes related to cervical cancer were higher in CSCC
   organoids cultured with UCEM compared to Matrigel. Furthermore, the
   transcriptional profile of UCEM-cultured organoids more closely
   resembled that of native CSCC tissues, potentially due to UCEM
   providing both structural support and a tissue-specific
   microenvironment. Consequently, organoids cultured in UCEM display
   improved fidelity to the true pathophysiological state of CSCC in
   patients, thus facilitating the reflection of tumor heterogeneity. Our
   experimental findings indicate that the tissue-specific
   microenvironment can enhance the transcriptional activity of oncogenes
   in primary tumor tissues. Transcriptomic, histological, and clinical
   data of ovarian cancer before and after chemotherapy indicate that
   platinum-based chemotherapy (carboplatin combined with paclitaxel) can
   induce tumor matrix changes in ovarian cancer patients ([195]77). In
   solid tumors, this treatment particularly affects genes encoding core
   matrix proteins (collagen, proteoglycans, and ECM glycoproteins) and
   ECM-related proteins ([196]78). In the experiment, we found that UCEM
   can increase the resistance of CSCC organoids to apoptosis induced by
   chemotherapy drugs. This indicates that the biochemical characteristics
   and bioactive signals of UCEM are key factors in cancer cell survival
   and chemotherapy resistance. Therefore, UCEM can partially optimize the
   pathological model of CSCC organoids, rendering them more similar to
   native CSCC tissues. This enhancement is particularly valuable for
   research into cervical cancer pathogenesis and treatment, aligning with
   the fundamental principles of precision medicine.

   In conclusion, this experiment demonstrates that UCEM, prepared using
   decellularization technology, retains both the structural components of
   the ECM and the protein components involved in cellular activities.
   Therefore, UCEM effectively maintains the specific microenvironment of
   the ECM in human cervical tissue. Our experiment confirmed that UCEM is
   capable of supporting the formation and growth of CSCC organoids.
   Analysis of the morphology, structure, growth process, and tumorigenic
   ability of CSCC organoids revealed that the performance of CSCC
   organoids cultured in UCEM was comparable to that of Matrigel in
   multiple aspects. Transcriptome sequencing revealed that, when compared
   with Matrigel, the expression level of cervical cancer oncogenes was
   higher in CSCC organoids cultured in UCEM and more closely resembled
   that of the native CSCC tissues. Therefore, our research suggests that
   UCEM could enhance the tumor heterogeneity of CSCC organoids by
   retaining a greater number of cervical cancer oncogenes, resulting in a
   model that closely resembles native CSCC tissues. This offers a
   promising platform for studying the pathogenesis and treatment of
   cervical cancer and provides creative ideas for the improvement of
   organoid technology.

MATERIALS AND METHODS

Preparation of cervical ECM hydrogel

   Para-cancerous cervical tissues from patients with CSCC were collected
   in multiple batches. A portion of these tissues was obtained
   immediately after isolating the uterine specimens. UCEM was prepared
   using previously reported methods ([197]79). The steps involved were as
   follows: (i) Decellularization: The cervix was cut and digested with
   0.02% pancreatin and 0.05% EDTA for 4 hours. It was then further
   digested in 3% Triton X-100 for 4 hours and stirred with 4% sodium
   deoxycholate for 4 hours to obtain white, porous cervical tissue. (ii)
   Disinfection and sterilization were carried out using ethanol and
   peracetic acid. (iii) The tissue was freeze-dried and ground into a
   powder. (iv) Digestion: A solution of pepsin hydrochloride was used to
   digest and prepare pre-gel solution (10 mg/ml). It was diluted with
   phosphate-buffered saline (PBS) before use and neutralized to a
   physiological pH of 7.35 to 7.45 using NaOH.

Characterization of cervical ECM hydrogels

   Histological analysis and immunofluorescence tests were conducted on
   the cervix before and after the decellularization process to confirm
   the presence of ECM components and the absence of CCs. The cervical
   tissues were fixed with a 4% paraformaldehyde solution and then
   embedded in paraffin. Thin sections of 4 μm were cut and stained using
   H&E and Masson methods. H&E staining was carried out following the
   standard procedure, which involved hematoxylin staining for 3 min and
   eosin staining for 1 min. In the Masson staining procedure, the
   sections were soaked in a potassium dichromate solution overnight,
   followed by staining with Weigert’s iron hematoxylin for 1 min. After
   rinsing with distilled water, the sections were stained with ponceau
   magenta for 6 min, rinsed again, immersed in a phosphomolybdic acid
   solution for 1 min, and stained with 2.5% aniline blue for 30 s. For
   immunofluorescence analysis, primary antibodies against collagen I
   (1:200; Abcam, ab34710), collagen IV (1:100; Abcam, ab6586),
   fibronectin (1:200; Abcam, ab6328), and lamininβ1 (1:100; Abcam,
   ab44941) were used. The primary antibodies were diluted according to
   the instructions and incubated overnight at 4°C. Afterward, the
   secondary antibody including donkey anti-rabbit (Abcam, ab150075), goat
   anti-rat (Abcam, ab150167), and goat anti-mouse (Abcam, ab150115) was
   applied and incubated at room temperature for 50 min. Subsequently, the
   samples were washed with PBS three times. The nucleus was stained with
   4′,6-diamidino-2-phenylindole (DAPI; Servicebio, G1012) and incubated
   at room temperature for 10 min. Imaging was performed using the Nikon
   DS-U3 system equipped with fluorescence microscopy (Nikon Eclipse C1,
   Japan).

DNA quantification and scanning electron microscopy to observe the
microstructure of UCEM

   Before decellularization, the cervical tissue and ECM were freeze-dried
   and ground into powder. Genomic DNA was extracted and quantified using
   the PureLink Genomic DNA Mini Kit (Invitrogen, K182001) following the
   manufacturer’s instructions. The DNA content in the cervical tissue and
   ECM was compared to confirm the successful removal of CCs. After
   solidification, the UCEM samples were fixed with an electron microscope
   fixing solution (Servicebio, G1102), rinsed three times with 0.1 M
   phosphate buffer (Servicebio, G0002), and then fixed with 1% osmium
   tetroxide (Ted Pella Inc., 18456) for 2 hours in a light-free
   environment. The samples underwent ethanol gradient dehydration,
   followed by conductive treatment after critical point drying. Last, the
   samples were observed and photographed under a scanning electron
   microscope (HITACHI, SU8100).

Oscillatory rheology

   A solution (10 mg/ml) of UCEM was quiescently incubated at 37°C for 30
   min. Subsequently, oscillatory rheology measurements were conducted
   using a rheometer (Anton Paar MCR 302e, Austria). The gap height was
   set to 1 mm. Amplitude (0.1 to 100 strain %) sweeps were performed on
   UCEM at a frequency of 1 Hz. The testing temperature was maintained at
   37°C.

Protein components and functional analysis of UCEM

   Mass spectrometry was used to detect and analyze the protein components
   in UCEM. The protein components were extracted from UCEM and then
   identified through liquid chromatography–mass spectrometry tandem
   analysis by hydrolyzing them with pancreatic enzymes. The identified
   proteins were subjected to systematic bioinformatics analysis using a
   database, which included protein annotation, functional classification,
   and enrichment analysis. The proteomic data of Matrigel are sourced
   from Suran Kim’s research ([198]17). Human matrisome data were obtained
   from ([199]https://sites.google.com/uic.edu/matrisome/home) ([200]80),
   while human cervical data were obtained from
   ([201]www.proteinatlas.org).

Culture of cervical squamous cell carcinoma organoids in UCEM and Matrigel

   This study was approved by the Ethics Committee of Zhujiang Hospital of
   Southern Medical University (approval number: 2023-KY-057-01). Written
   consent was obtained from patients before collecting samples of CSCC.
   Fresh tumor tissues, slightly firm in texture and pink in color, were
   collected from primary patients with CSCC. The tissues were obtained
   from biopsy or surgical specimens. The cervical cancer tissues were
   immediately placed in a basic culture medium (DMEM/F12, 1%
   penicillin/streptomycin), stored on ice, and promptly transported to
   the laboratory. To remove any adhering clots and mucus, the cancer
   tissue specimen was washed multiple times with PBS. Afterward, the
   cancer tissue specimen was cut into small granules using sterile
   scissors and digested with a solution of type IV collagenase (1 mg/ml;
   Gibco, catalog no. 17104019) on a shaker at 37°C for 1 hour. Digestion
   was terminated using a serum-free culture solution, and a cell
   suspension was filtered through a 100-μm filter. The cell suspension
   was then centrifuged to obtain cell precipitates, which were
   subsequently washed and centrifuged with DMEM/F12 for two to three
   times. Cell counting was conducted using blood cell counting plates.
   The cell precipitate was resuspended using growth factor–reduced
   Matrigel (Corning, catalog no. 356231) and UCEM (on ice). The cells
   were seeded into preheated 24-well plates at 25 μl per well and shaped
   into semispherical structures, followed by placement in a 37°C
   incubator for 20 to 30 min. After solidification, the CSCC organoids
   culture medium (BioGenous, K2169-CC) was added for group culture. The
   addition of Y-27632 (10 μM; Stemcell, catalog no. 72304) was used to
   enhance the cell survival rate. The growth of organoids was observed
   daily and recorded using an inverted microscope (Ningbo Shunyu, ICX41).

H&E staining and immunohistochemical staining of CSCC organoids

   The CSCC tissues were fixed in 4% paraformaldehyde overnight at 4°C,
   followed by paraffin embedding and sectioning. The organoids were
   obtained by dissolving Matrigel using Cell Recovery Solution (Corning,
   catalog no. 354253) and then fixed with 4% paraformaldehyde overnight
   at 4°C. UCEM was dissolved using IV collagenase (1 mg/ml; Gibco,
   catalog no. 17104019), and the resulting mixture was centrifuged to
   obtain an organoid pellet. The pellet was then fixed with 4%
   paraformaldehyde overnight at 4°C. CSCC organoids were pre-embedded in
   2% agar, paraffin-embedded, and sectioned. H&E staining, as well as
   immunohistochemical staining, was conducted on CSCC tissues and CSCC
   organoids cultured in UCEM and Matrigel. The tissues and organoids were
   incubated overnight at 4°C with the following antibodies: anti-P16INK4A
   (Abcam, catalog no. ab108349), KI67 (Abcam, catalog no. ab15580), TP63
   (Abcam, catalog no. ab735), and KRT13 (Abcam, catalog no. ab92551).
   After incubation, a secondary antibody (ZSGB-BIO, PV-6000) was applied
   at room temperature for 1 hour, followed by chromogenic visualization
   using DAB (ZSGB-BIO, ZLI-9017). Last, the CSCC tissues and organoids
   were observed and photographed using a Leica DM2500 microscope.

Comparison of success rate and formation efficiency of CSCC organoids

   (i) The success rate of organoid culture was compared in this study.
   CSCC tissue was obtained from 52 patients and CSCC organoids were
   cultured using Matrigel and UCEM. After 7 to 10 days, the formation of
   organoids was observed, and the success rate of culture was calculated
   using the formula: success rate of organoid culture = (number of
   patients successfully cultured/total number of patients) × 100%.
   Statistical analysis was performed using the chi-square test. (ii) The
   formation efficiency of organoids was compared in this study. The CSCC
   tissue from each patient was minced, digested, and filtered, and cell
   counting was conducted. Organoids were cultured using Matrigel and
   UCEM, respectively. The organoid formation efficiency was calculated
   using the formula: organoid formation efficiency (%) = (number of
   organoids generated/number of cells) × 100%. In a 24-well plate, each
   patient had three replicate wells per group, and the medium was changed
   every 2 to 3 days. After 7 to 10 days, the number of organoids with a
   diameter ≥80 μm in each well was manually counted. To minimize errors,
   the count was performed independently by three researchers.

Passage, cryopreservation, and resuscitation of CSCC organoids

   (i) Passage: Passage of cultured CSCC organoids can be initiated after
   approximately 10 days. Matrigel was dissolved on ice using Cell
   Recovery Solution (Corning, catalog no. 354253). UCEM was dissolved
   with IV collagenase (Gibco, catalog no. 17104019, 1 mg/ml). The
   organoids were pelleted by centrifugation. The organoids were
   dissociated through repeated pipetting and subcultured at a ratio of
   1:2 to 1:4. (ii) Cryopreservation: CSCC organoids cultured in Matrigel
   and UCEM were dissociated using the same method and then added to
   serum-free cryopreservation solution (Stemcell, catalog no. 07930) for
   cryopreservation. The organoids were stored at −80°C overnight and
   transferred to a liquid nitrogen tank for long-term storage after 24
   hours. (iii) Resuscitation: CSCC organoids frozen in liquid nitrogen
   were removed, quickly thawed in a 37°C water bath, centrifuged with
   culture medium, and then embedded in Matrigel or UCEM. After
   solidification, they were added to the organoid culture medium.

Live/dead staining of CSCC organoids

   The CSCC organoids cultured in Matrigel and UCEM were obtained
   following the methods described above. A live/dead cell detection
   solution was prepared using a centrifuge tube. First, 2 ml of PBS was
   added, followed by 1 μl of a 16 mM propidium iodide (PI) solution and 1
   μl of a 4 mM calcein AM solution. The solution was then mixed by
   agitation. The resulting working solution (2 μM calcein AM and 8 μM PI)
   was used directly for staining. The stained organoids were incubated in
   a 5% CO2 cell incubator at 37°C for 30 to 40 min. Last, the organoids
   were observed and photographed under a fluorescence microscope.

Transcriptome sequencing analysis

   mRNA transcriptome sequencing was conducted on CSCC tissues and
   organoids cultured in Matrigel and UCEM. The CSCC tissues and organoids
   originated from the same patient, and the experiment was performed with
   three biological replicates. The extraction and detection of RNA,
   library construction, quality inspection, and sequencing steps were
   performed by Novogene (Novogene, Beijing, China). The gene expression
   profiles of the organoids in Matrigel and UCEM were evaluated for their
   similarity to their corresponding source tissues.

Drug assay

   CSCC organoids were collected and dissociated into single cells,
   following the passage method described above. The cells were then
   counted. They were resuspended in UCEM and Matrigel separately and then
   seeded into the wells. After solidification, an organoid culture medium
   was added. After 3 days of culture, the medium was replaced with an
   organoid culture medium containing different concentrations of
   carboplatin (1, 10, and 100 μM). The viability of the organoids was
   assessed using CCK-8 (Dojindo, Lot. KP672) at 24, 48, and 72 hours
   after carboplatin treatment. The absorbance values (OD, optical
   density) were measured at 450 nm. Each concentration of carboplatin had
   three replicate wells, and a blank control was included. Patients with
   CSCC who underwent TC regimen chemotherapy (paclitaxel combined with
   carboplatin) were selected for our study. Pathological tissue samples
   were obtained through surgery or biopsy after obtaining informed
   consent from the patients. Drug sensitivity testing was performed for
   patients with successful organoid cultures. In combination therapy,
   paclitaxel (Sigma-Aldrich, catalog no. 580556) and carboplatin
   (Sigma-Aldrich, catalog no. 1096407) were mixed in a concentration
   ratio of 1:10. After 5 days of treatment with the TC protocol, the
   organoid activity was detected using CCK-8, and a drug response curve
   was plotted. In addition, the patient’s clinical response was evaluated
   through imaging after two to four cycles of chemotherapy.

Detection of tumorigenic ability of xenotransplantation in vivo

   The animal studies were conducted with the approval of the Ethics
   Committee of the Zhujiang Hospital of Southern Medical University
   (approval no. LAEC-2022-212). A total of twenty 4-week-old female nude
   mice of the BALB/c strain were procured and divided into two groups,
   with 10 mice in each group. The mice in each group were respectively
   inoculated with CSCC organoids which were grown in Matrigel or UCEM. A
   mixture solution was prepared by combining DMEM/F12 with UCEM or
   Matrigel in a 1:1 volume ratio. Approximately 1000 organoids with
   diameters <100 μm were suspended in 100 μl of the mixture solution.
   Following the group assignment, the organoids were implanted into the
   subcutaneous region of the abdominal side of the nude mice using a 1-ml
   syringe. Tumor volume (in cubic millimeters) was calculated as length ×
   width^2 × 0.5, with length representing the longest tumor diameter (in
   millimeters) and width indicating the longest transverse diameter
   perpendicular to the length (in millimeters). The growth of the tumors
   in the nude mice was monitored for 1.5 months, and the mice were
   euthanized before the tumor volume reached 1500 mm^3. Subsequently, the
   tumor tissues were excised and photographed for comparative analysis.

Statistical analysis

   The data were presented as means ± SD. The statistical analysis of the
   experimental data was performed using SPSS 21.0. GraphPad Prism 9.0 was
   used for statistical plotting, while ImageJ software was used for image
   analysis and quantitative assessment. The paired-samples t test was
   applied for paired data, while the two-sided Student’s t test was used
   for comparing two groups when the data followed a normal distribution.
   The chi-square test was used for counting data, and one-way analysis of
   variance (ANOVA) was performed for group comparisons. A significance
   level of P < 0.05 was considered statistically significant; *P < 0.05,
   **P < 0.01, and ***P < 0.001.

Acknowledgments