The role of decellularized cell derived extracellular matrix in the establishment and culture of in vitro breast cancer tumor model

Decades of research have shown that two-dimensional cell culture studies are insufficient for preclinical cancer diagnosis and treatment, and that cancer cells in three-dimensional (3D) culture systems have better cell–cell and cell–matrix interactions, gene expression, heterogeneity, and structural complexity that more closely resemble in vivo tumors. Researchers are still optimizing 3D culturing settings for different cancers. Despite promising tumor spheroid research, tumor cell-only aggregates lack the tumor microenvironment and cannot model tumors. Here, MCF-7 breast cancer cell derived decellularized extracellular matrix (CD-dECMs) were obtained and converted into autologous, biologically active, biocompatible, and non-immunogenic hydrogels to be used as micro-environment in both organoid formation and culture. For the production of organoids, CD-dECM doping concentrations ranging from 0.1 mg ml−1 to 1.5 mg ml−1 were evaluated, and the lowest concentration was found to be the most effective. For organoid culture, 8 mg ml−1 CD-dECM, 4 mg ml−1 rat tendon collagen type I (Col I) (4 mg ml−1) and a 1:1 (v/v) mixture of these two were used and the most viable and the biggest organoids were discovered in CD-dECM/Col I (1:1) group. The results show that autologous CD-dECM can replace hydrogels in tumor organoid generation and culture at low and high concentrations, respectively.


Introduction
For many years in cell-based research, researchers have employed two-dimensional (2D) cell culture methods to examine biophysical and biochemical cell responses.However, genuine replication of multicellular and three-dimensional (3D) structures is not conceivable with these approaches, as cell-cell and cell-ECM interactions are not provided in 2D cell culture techniques as they are in 3D cell culture techniques [1].In this context, 3D cell culture techniques have been created to better simulate the true microenvironment in vivo, and multiple studies have shown that cellular activities such as proliferation, metabolic activity, and cellular behavior are better provided in 3D culture environments than in 2D culture environments [2][3][4].These 3D culture techniques, which have yielded tangible results for the scientific community in the shape of spheroid and organoid, are still being investigated as trend research topic with the purpose of presenting more effective structures [5].
Spheroids are simple clusters of cells with a wide range of characteristics, such as tumor cells, stem cells, or a variety of mature or immature cells specific to different tissues.In spheroids, cells construct 3D structures without the use of scaffolding; they accomplish it by simply adhering to one another.They, however, lack the ability to self-assemble or renew, making them less evolved than organoids [6].In contrast to spheroids, organoids are complex cell clusters generated by organ specific adult stem or progenitor cells.They self-organize and grow into tiny replicas of parent organs, usually in the presence of an extracellular scaffolding material such as collagen or Matrigel.Organoids, also known as 3D ex vivo cellular cultures, emerge via self-organization or guided assembly in response to particular organogenesis signals.Organoids not only physically resemble the original tissue in terms of architecture, cellular organization, and composition, but they also retain the in vivo counterparts' genetic markers [7].Although the terms organoids, spheroids, and 3D cell cultures are frequently used interchangeably in the literature, it is worth noting that spheroids are low-complexity spherical cellular units that are frequently cultured as free-floating aggregates, whereas organoids are 3D progenitor cell-matrix structural units that resemble an organ in both structure and function [8].
Although tumor spheroids have great potential for preclinical chemosensitivity testing, several issues remain unresolved, including the fact that the ability to generate 3D tumor spheroids varies significantly between tumor cell lines, spheroids with regular morphologies cannot be generated, and the transfer of formed spheroids frequently remains challenging [9].Besides, in spheroid-based studies, the preparation of tumor cell aggregates alone is not sufficient for a real modeling, and peri-necrotic niche, peri-vascular niche and metastatic niche should also be provided [10].To date, dECM-based hydrogels have been extensively investigated in the literature in order to generate the complex environments stated herein [11].In a nutshell, extracellular matrix (ECM) hydrogels are water-swollen, non-cellular, fibrillary 3D networks made up of numerous macromolecules, primarily collagen type I, elastin, fibronectin, laminins, glycoproteins, proteoglycans, and glycosaminoglycans (GAGs) [12].The collagen as the main component of the ECM structure, provides the ECM ability to polymerize via self-assembly in the way of gelation promoted by electrostatic and hydrophobic interactions and affected by temperature, collagen concentration, pH, and the presence of other biomolecules or polymers in the microenvironment [13][14][15].These prepared biologically active, biocompatible, and non-immunogenic dECM-based hydrogels possess a number of cell growth factors that can enhance the growth, migration, proliferation, differentiation and angiogenesis of entrapped cells.The real-time interactions of these hydrogels and the trapped cells will enable better mimicry of tissues and organs in vitro [16].Additionally, the structure and mechanical properties of these dECM-based hydrogels can be modified via chemical crosslinking techniques such as gluteraldhyde, genipin, and carbodiimide, as well as physical crosslinking strategies such as freeze-drying cycles and the formation of interpenetrated networks [17].
In this presented study, dECMs of MCF-7 breast cancer cells (dCD-ECM) were obtained and then, the chemical composition and the decellularization efficiency was evaluated by FT-IR analysis, collagen staining, DAPI staining, sGAG quantification, residual genomic DNA quantification and SEM analysis.Afterward, the dCD-ECMs obtained in powder form was turned into hydrogel by using acidic pepsin solution in different concentrations.The resulting hydrogels were tested both for their usability in tumor organoid formation and as a niche in which tumor spheroids can be cultured.The conventional hanging drop technique was utilized to fabricate tumor organoids, and the effects of various concentration of dCD-ECM solutions on organoid formation were investigated.First, the organoid formation process was observed utilizing light microscope imaging.Then, throughout the organoid formation process, analyses such as determination of the apoptosisnecrosis, glucose consumption, and mitochondrial activity were conducted, and the optimal dCD-ECM concentration was identified.Finally, close-up images of the organoids formed utilizing optimum dCD-ECM were acquired using light and fluorescence microscopy, and morphological analyses such as circularity, roundness, solidity, and area measurements were performed on the organoids using an image processing software.On the other hand, dCD-ECM hydrogels at the optimal concentration, type I collagen hydrogels at the same concentration, and the hydrogels combination of both were examined for usability as tumor niche in tumor organoid culture.The preparation, characterization and in vitro applications of MCF-7 tumor cell-derived ECM are presented schematically in figure 1.

Preparation of CD-dECM gel
5 × 10 5 MCF-7 cells (passage 10-12) were seeded to 35 mm culture dishes (n = 6) and were cultured as described above till they reach confluency (approximately 48 h).Upon the confluency, the cells were washed once with Dulbecco's Phosphate Buffered Saline (D-PBS, Sigma Aldrich, Germany) gently and then were exposed to D-PBS containing 0.5% Triton X-100 and 20 mM NH 4 OH for 5 min at 37 • C. Later, the cellular remnants were discarded carefully and the precipitated dECM was treated with 100 µg ml −1 DNase I and 100 µg ml −1 RNase A for 1 h at 37 • C. After, the obtained dECM was collected in a falcon tube and centrifuged at 5000 rpm for 5 min.The supernatants were removed carefully and the collected dECM was washed several times with D-PBS and then placed to a freezer set to −80 • C in 1 ml of D-PBS.Then the dECM was lyophilized at −80 • C under 0.001 mbar vacuum by using a freeze-dryer (FreeZone, Labconco, USA) for overnight.
The freeze-dried dECM was turned into powder form by using a mortar and a pestle under liquid nitrogen.Then, 20 mg of dECM powder was added to acidic pepsin solution (1 mg ml −1 pepsin in 0.01 M HCl).The suspension was kept at room temperature for 4-5 d on a rotating shaker till the powder will completely be dissolved.Then, the solution was taken onto ice and the pH was adjusted to physiological level (pH: 7.45) by using 0.1 M NaOH.Afterwards, 100 µl of 10X phosphate buffered saline (PBS, Sigma Aldrich, Germany) was added for each 900 µl of the solution in order to make the solution compatible with physiological conditions.Lastly, the desired amount of pre-gel dECM solution was mixed with culture medium to reach the final concentrations.

Preparation of organoids via hanging drop
The conventional hanging drop culture method was used to obtain MCF-7 organoids.For this purpose, the splitted cells were counted with an automated cell counting slide (EVE Cell Counting Kit, EVS-050, NanoEntek, Korea).Afterwards, the cell suspension was prepared as 4 × 10 4 cells ml −1 and the hanging droplets were formed in 25 µl consisting of 1 × 10 3 cells on a lid of a 60 mm petri dish.Lastly, 6 ml of D-PBS was added to the petri dishes in order to prevent the droplets from drying.Lastly, the petri dishes were placed to a CO 2 incubator set to 37 • C and were cultured throughout 5 d.In parallel, reconstituted dECM solutions were added to the cell suspension with final concentrations of 0.1 mg ml −1 , 0.5 mg ml −1 , 1 mg ml −1 , and 1.5 mg ml −1 , and the droplets were prepared via these dECM doped cell suspensions.Thence, the effect of dECMs in different ratios upon organoid formation was investigated.

Characterizations of dECMs and evaluation of decellularization efficacy
The amount of residual genomic DNA remaining in the structure after decellularization was measured by PicoGreen DNA fluorescence staining (Quant-iT™ PicoGreen ® dsDNA Assay Kit; Invitrogen).Briefly, 10 mg of decellularized and dried cell derived decellularized ECM (CD-dECM) was weighed and kept in 1 mg ml −1 Proteinase K (Sigma, Germany) solution set to 60 • C, prepared in 100 mM ammonium acetate (Sigma, Germany) till they completely dissolve.The samples taken from the lysates obtained were processed according to the test kit contents and the final absorbance values were measured with the help of a fluorescent spectrophotometer (excitation: 485 nm, emission: 520 nm) (PerkinElmer, USA) (n = 3, 5 parallel).The obtained absorbance values were compared with a DNA standard curve.To determine the extent to which the ECM structure may be preserved during the decellularization process, sGAG analysis was performed on the lysates mentioned above.Following that, 40 µl of the resulting CD-dECM was suspended in 200 µl of DMMB reactive solution (40 mmol l −1 sodium chloride, 40 mmol l −1 glycine, 46 mmol l −1 DMMB, and 0.1 M HCl), and the mixture was promptly measured at 525 nm using a microplate reader (BioTek, UK) (n = 3, 5 parallel).Obtained absorbance values were converted into nanograms by comparing with the standard curve of 0-50 mg ml −1 bovine cartilage derivative chondroitin sulfate (CS) (Sigma Aldrich, Gillingham, UK) prepared in PBS.In order to verify the removal of nuclear content dECM structures were entrapped between two glass slides and were stained with 10% EverBrite™ hardset mounting medium with DAPI (4 ′ ,6-diamidino-2-phenylindole) (Biotium, Fremont, CA) for 30 min at room temperature.Then, the stained dECMs were exposed to several washing steps in order to remove excess amount of stain.Lastly, the dECMs were observed by a fluorescence microscope (Leica, Germany).The amount of total collagen in CD-dECMS was determined by Hydroxyproline Assay Kit (Biovision, USA).
Briefly, 100 µl of ultrapure water and 12N HCl (Honeywell, USA) were added to 10 mg of dry decellularized CD-dECMs and incubated at 120 • C for 3 h.Following the incubation, the samples were centrifuged at 15 000 rpm for 5 min and the 50 µl of supernatant was taken to a 96-well plate to be dried at room temperature for 12 h (n = 3, 5 parallel).The 100 µl of Chloramine T agent included in the kit was added to the dried samples and incubated for 5 min at room temperature.Then, 100 µl of DMAB reactant was added to each well and the 96-well plate was incubated at 60 • C for 90 min.At the end of the incubation period, the absorbance value of the solutions was read at 560 nm with the help of a microplate reader (BioTek, UK).The absorbance values obtained were converted to micrograms using a standard curve prepared according to the procedure contained in the kit prior to analysis.
FT-IR analysis was conducted not only to reveal the chemical composition of the dECMs but also evaluate the effect of the decellularization on the proteinbased structure of ECM.An attenuated total reflection mode connected Thermo Nicolet Nexus FTIR spectrometer was used to record the FTIR spectra of the materials between 4000 and 400 cm −1 (Thermo Fischer Scientific Inc.).In order to reveal the presence and the distribution of the collagen fibers in dECMs picrosirius red staining was performed by a staining kit (Polysciences, Germany).Briefly, freeze-dried samples were entrapped between two glass slides and were immersed into the Solution A found in the kit for 2 min and then washed with distilled water.Then, samples were immersed into the Solution B for 2 h and at the end of the incubation time samples were placed in Solution C for 2 min.Later than, the samples were subjected to dehydration process by immersing 70% ethanol solution for 1 min and directly under a polarized light microscope (Leica DM4 P, Germany).On the other hand, a SEM analysis was conducted to the obtained freeze-dried samples in order to investigate the ECM based structure.For this purpose, the freeze-dried samples kept in +4 • C till they will use were immersed in 70% (v/v) ethanol solution for 1 h.Then, in order to make the samples dehydrated, the samples were exposed to an increasing alcohol series (80%, 90%, and 100%) for 10 min each.Subsequently, air-dried samples were subjected to hexamethyldisilazane (Sigma Aldrich, Germany).The samples were lastly coated with gold-palladium before imaging a SEM (Carl Zeiss Evo 50, Germany).

Gelation kinetics of hydrogels
The gelation kinetics of the hydrogels was determined via turbidimetric spectrophotometric analysis.For this purposes, CD-dECM was prepared at a ratio of 2 mg ml −1 , 4 mg ml −1 and 8 mg ml −1 respectively.Later than, three kind of hydrogels were prepared to be used as 3D culture milieu: CD-dECM (8 mg ml −1 ), rat tendon collagen type I (4 mg ml −1 ) (Neuromics, USA), and the mixture of them (1:1 (v/v)).All pregel solutions were prepared on ice prior to the analysis, and a 100 µl of each solution was placed in a 96-well plate (n = 3) to be read at 405 nm.The absorbance was measured in every 3 min throughout 90 min at 37 • C by using a microplate reader (BioTek, UK).The obtained raw values and the normalized absorbance values were depicted as a graph.The normalized absorbance (NA) value was obtained by using the following formula as described in a previous research [18], NA stands for normalized absorbance, A stands for absorbance at a specific time, A0 represents the beginning absorbance, and Amax represents the maximum absorbance.The resulting graphs were used to calculate kinetic parameters such as lag time (t lag ), time to half gelation (t 1/2 ), and gelation rate (S).The t lag was defined as the intercept of the linear region of the gelation curve with 0% absorbance, the t 1/2 as the time 50% absorbance, and the S as the slope of the linear region of the gelation curve [19].

Organoid characterizations Morphological analysis of 3D structures:
The organoid formation process was visualized by brightfield imaging (Leica DM IL Led, Netherlands).The sphericity index, roundness, solidity, diameter and the area of the resulting organoids were calculated via an automated image analysis algorithm (ImageJ, National Institutes of Health, Bethesda, Maryland, USA).Briefly, the obtained grayscale files were converted to binary images.Subsequently the automated edge detection and the area measurement were conducted as described in a previous study [20].All calculations were performed taking into account the methodology detailed in a previously published study [20].On the other hand, the cells were labeled with CellTracker TM Green Dye (CMFDA, Thermo Fisher Scientific, USA) according to the manufacturer instructions in order to be observed under fluorescence microscopy.
Live and dead assay: Acridine orange/Propidium iodide (AO/PI) (Sigma Aldrich, Germany) dual staining was used to determine the change in cell viability during the organoid formation in droplets.For this purpose, 25 µg ml −1 AO and 25 µg ml −1 PI stock solutions were prepared and mixed in a ratio of 1:1 (v/v) in a falcon.Then, a 10 µl of the resulting staining solution was feeded carefully to each droplet via a conventional 31G insulin syringe (BD Biosciences, USA).Later on, the stained cells were observed immediately under a fluorescence microscopy with FITC and Rhodamine filter (Leica DM IL Led, Netherlands).Samples were analyzed in triplicate.

Determination of apoptotic and necrotic cells in organoids:
Organoids were dissociated on each analysis date and the obtained single-cell suspensions were treated with AO/PI mixture in order to estimate the percentage of apoptotic and necrotic cells throughout the organoid formation process.Samples were analyzed in triplicate.Apoptosis and necrosis were determined by counting cells that had the following characteristics: (1) uniform green nucleus with organized structure, intact plasma membrane, and orange or green cytoplasm, viable cells; Cell proliferation by alamar blue assay: Cell proliferation was determined with Alamar Blue Assay Kit.For this purpose, the droplets were collected with culture medium from the lid of the petri dishes on each analysis day and transferred to a well plate in 1 ml culture medium.Subsequently, 100 µl of Alamar Blue test solution was added to the wells and mixed well.Later, the well plate was covered with aluminum foil and incubated at 37 • C for 2.5 h.At the end of the incubation period, 200 µl of test solution was taken to a 96 well-plate (n = 10 and 5 parallel) and was read at 570/600 nm by using a spectrophotometer (Multiscansky, ThermoFisher, Germany).The obtained absorbance value was depicted as a graph.
Glucose uptake assay: Glucose uptake capacity of the cells in the droplets throughout the organoid formation was determined by a conventional glucometer (ACON Biotech, China).The glucose consumption was calculated by the concentration difference between the culture medium before being formed droplet and after the organoid formation.Briefly, 10 µl of culture medium was exemplified from the droplets and was directly read in the glucometer.Samples were analyzed in triplicate.The calculated glucose consumptions were depicted as a graph.

Organoid culture in different hydrogel niches
Organoids were transferred into a 200 µl of different types of pre-gel solutions (CD-dECM (8 mg ml −1 )), rat tendon collagen type I (4 mg ml −1 ) (Neuromics, USA), and the mixture of them (1:1 (v/v)) placed into a 48-well plate and then immediately transferred to the cell culture incubator for 90 min without being disturbed to allow the pre-gels polymerize.Upon the completion of self-polymerization, 200 µl of culture medium pre-heated to 37 • C was added gently on top of the gel.The culture medium was replenished carefully in every 24 h throughout 7 d.

Monitoring organoid growth and metabolic activity
The obtained organoids were cultured for 7 d after being confined to the gel.On the 1st, 3rd and 7th days of the culture, organoids were visualized with the help of a light microscope and the changes in organoid size was calculated from the images obtained with the help of an automated image analysis algorithm (ImageJ, National Institutes of Health, Bethesda, Maryland, USA).On the other hand, the organoids were dissociated on the same analysis days via 5 units ml −1 collagenase type I (Sigma Aldrich, Germany) exposure prepared in PBS.Then, the disaggregated cells were transferred to a 12-well plate to be cultured for 12 h.Subsequently, the cell viability was evaluated via alamar blue assay as described above.

Statistical analysis and data availability
Multiple T-test was performed to determine the statistical differences between the groups by using GraphPad Prism software version 6 (GraphPad Software, La Jolla, CA).Statistical significance determined using the Holm-Sidak method, with p = 0.05.Different p values and their level of significance are as follows: p < 0.05 = * , p < 0.01 = * * , p < 0.001 = * * * , p < 0.0001 = * * * * .The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.

Results and discussion
In order to reveal the chemical composition of the obtained CD-dECM, the FTIR spectrum was recorded and given as figure 2(A).The relevant transmittance peaks in the FTIR spectrum were determined and interpreted.The peaks centered at 3290 and 2925 cm −1 were attributed to the amide A band of organic collagen due to the N-H stretching vibration and the amide B band due to the asymmetric stretching of the CH 2 stretching vibration, respectively.The region between 1700 and 1200 cm −1 is mainly dominated by protein adsorption where Amide I, Amide II and Amide III peaks were present.Amide I band sensitive to secondary structure of the protein due to the stretching vibration of carbonyl group (-C=O) of peptides was observed at 1647 cm −1 .Amide II band associated with C-N stretching and N-H in plane bending from amide linkages, including wagging vibrations of CH 2 groups on glycine backbone and proline side-chains was observed at 1543 cm −1 .The signals in the region between 1400 and 1450 cm −1 were attributed to various amino acid side chains and some lipids while the signals between 1300 and 900 cm −1 were associated with the phosphate groups stem from the residual nucleic acid, DNA, and RNA after decellularization.In particular, the peaks centered at 1293 cm −1 and 1077 cm −1 were associated with the asymmetric and symmetric phosphodiester vibrations of nucleic acids while the peak observed at 1151 cm −1 were linked with the hydrogen and non-hydrogen bonds of the C-O stretching vibrations.The chemical structure of CD-dECM was corroborated as the ECM via the existence of the several functional groups in the structure such as organic collagen amide A, amide B, amide I, amide II, and amide III with the help of previous literature findings [21][22][23].
DAPI staining was used for observing the decellularization efficiency and the related images were given in figure 2(B).The primary criteria for effective decellularization is the absence of DAPI-visible residual nuclear material [24].As a result of DAPI staining, no clear blue stained cell nuclei were found in the ECM, which indicates that the cells were successfully removed from the structure as a result of the applied decellularization protocol.Picrosirius Red staining was performed to show the presence of collagen inside the CD-dECM and the staining result was given as figure 2(C).Picrosirius Red is a histological procedure for determining the spatial distribution of collagen fibers, particularly type I and III collagen fibers [25].Positive Picrosirius Red staining was revealed the presence and distribution of collagen across the CD-dECM structure.Both the decellularization efficiency shown by DAPI staining and the determination of the presence of collagen in the structure demonstrated by Picrosirius Red staining were also confirmed by colorimetric analyses.
Moreover, sGAG, collagen and DNA quantification analyses were performed in order to identify the biochemical composition of CD-dECM, and the results were given as figure 2(D).The sGAG quantity was calculated as 1124.15± 75 µg while the collagen content was calculated as 367.98 ± 17 µg in 10 mg dry CD-dECM powder.On the other hand, the residual genomic DNA content was calculated as 194.10 ± 24 ng per the same amount of CD-dECM.
The GAGs are a kind of negatively charged linear carbohydrate with a repeating disaccharide unit [26].The GAGs are found mostly in cell membranes and within the ECM, where they function as molecular co-receptors in cell signaling for cell-cell and cell-ECM interactions that are critical for cell survival and differentiation [27].GAGs are divided into four categories based on the structure and sulfation degree of the repeating disaccharide: heparan sulfate, CS, keratan sulfate, and hyaluronic acid [28].The approach employed in this study was centered on the identification of sulfated ones.It has been documented in the literature that sGAG synthesis varies from cell to cell, and there has been no data upon sGAG accumulation neither in MCF-7 cell culture nor in cell-derived ECMs obtained from them.However, to give an example from a recent report examining dECMs from different cell types, the total sGAG amounts of human bone marrow mesenchymal stem cell, human synovial mesenchymal stem cell and human chondrocytes were reported as 206.8 ± 44, 73.8 ± 9.5, and 112.4 ± 43.6, respectively [29].The data obtained within the scope of this study were found similar to the total sGAG quantities per mg dECM presented in this previous report.On the other hand, it is well known that the amount of sGAG varies not only with the cell type employed, but also with the desired decellularization approach, particularly in decellularization-based investigations [30].Because the quantity of sGAG that may be kept in the final structure depends on the type of detergent or chemical substance utilized, as well as the postprocessing methods performed on the ECM upon decellularization [31].
Collagen is the primary component of the ECM structure as a glycoprotein, and the collagen family encompasses a range of distinct collagen subtypes; for example, type I and type III are fibrillary collagens found mostly in connective tissue and bones, whereas type IV is a basement membrane collagen [32].The main structure of collagen is defined by the tripeptide sequence glycin-X-Y, where X is typically proline and Y is frequently hydroxyproline [33].The collagen, which exists in both insoluble and soluble forms, can be quantified by using biochemical colorimetric techniques on the basis of its hydroxyproline content of its soluble form, as it was accomplished in this study.While this approach provides an estimate of the overall quantity of collagen, it should be noted that it does not account for the insoluble form of collagen.Similar to sGAG, the amount of collagen has not been reported in the literature before, neither in MCF-7 cell line culture nor in cell-derived ECMs obtained from them, so no comparison can be performed with any previous literature data.Since the ultimate goal of this study is not to investigate the effectiveness of a methodology that is frequently preferred in the literature for decellularization, it was only focused on the characterization of the final CD-dECM as well as its application in organoid engineering studies.
In decellularization-based studies, the following criteria have proposed for evaluating the success of removal of the genomic components: the dECM must contain (1) fewer than 50 ng double-stranded DNA (dsDNA) per mg ECM dry weight, (2) less than 200 bp DNA fragment length, and (3) no visible nuclear material by DAPI staining [26,27].The fact that the DNA content is substantially below 500 ng, which is the highest limit for 10 mg tissue weight within the scope of this study, demonstrates the efficiency of the applied decellularization method.
To determine the influence of CD-dECM on the development of organoids, droplets with CD-dECM concentrations ranging from 0.1 mg ml −1 to 1.5 mg ml −1 were formed and the effects of CD-dECM on the organoid formation process were observed in these suspended droplets and the relevant findings were given as figure 3.As seen in the turbidimetric gelation kinetics analysis findings in the further stages of the study, CD-dECM does not show any gelation at concentrations of 2 mg ml −1 and below.Therefore, the main purpose here is to obtain ECM-loaded organoids containing a certain amount of ECM proteins without any expectation of gelation and to focus on the reproducibility of organoid structures with a matrix.When the organoid formation process was evaluated, it was seen that the cells in the w/o CD-dECM group on day 1 gather together in full contact and form cell aggregates.Although a cell aggregate formation was observed in the group with the lowest concentration of CD-dECM (0.1 mg ml −1 ), similar to the control group, it was found that the cells were more loosely distributed in the droplet and were not in full contact in some regions.As the CD-dECM concentration was increased, it was observed that the CD-dECM structure was enveloped the cells from the outside like a capsule and the cells were observed like surrounded by a thick layer of ECM proteins.As CD-dECM concentration increases, cell-cell contacts decrease and the majority of cells prefer to adhere to hydrophobic proteins in the ECM structure, such as collagen, laminin, and fibronectin, rather than to adhere to each other and form organoids.In other words, the organoid structure was lost as the amount of CD-dECM employed rises, yet capsule forms containing cell-rich ECM can be generated.
Whether the addition of CD-dECM has an apoptotic or necrotic effect on cells during the organoid formation process was determined by AO/PI double staining.The staining results and the calculated apoptotic and necrotic indexes were given in figure 4.
In the AO/PI fluorescence microscopy images presented mutually with the light microscope images, very few red stained cells highlighting necrosis and orange stained cells highlighting apoptosis were found.This means that the doped CD-dECM and the amount of doping do not have any apoptotic and necrotic effects on cells.These small amounts of apoptotic and necrotic cells were attributed to the cells' own natural aging process.Although CD-dECM, a completely biologically derived and autologous biomaterial, is not expected to trigger cellular events such as apoptosis and necrosis, the possible effects of parameters such as concentration gradient, pH balance, osmotic balance and ion concentration of the dECM solution resulting from the production process were also tested.Any mature organoid consists of three main regions: an outer region consisting of cells with normal and intact nuclei with asynchronous active proliferation and active metabolism, a middle region that is stable in terms of proliferation and metabolic activity, and an interior region consisting of senescent and apoptotic or necrotic cells with fragmented nuclei [34,35].The leading causes of this apoptosis and necrosis seen in the middle and inner regions of a organoid are the limited access of the cells located in this region to the culture medium nutrients and gases, similarly the inability to properly remove cellular wastes from the environment, and the inevitable changes in pH and ion concentration due to the previous limitations [35,36].A similar situation was detected on the 7th day of culture in mature organoids without CD-dECM produced within the scope of this study, and the apoptotic index was calculated as approximately 15% in these organoids, while the necrotic index was determined as approximately 25%.Apoptosis and necrosis values of all organoids produced by doping at different rates of CD-dECM were calculated as approximately 5%-10% for each, and no statistically significant difference was observed between the groups.However, both apoptotic and necrotic percentages of organoids without CD-dECM were found to be statistically significantly different compared to those containing CD-dECM (p < 0.0001).Apoptosis and necrosis values of approximately 5%-10% seen in CD-dECM doped organoids were found to be compatible with other 3D culture findings with MCF-7 cells in the literature [37][38][39].Also, consistently with previous literature findings, it was thought that the significant necrosis and moderately high apoptosis observed in organoids without CD-dECM are the result of both the challenges encountered in mass transfer and diffusion restrictions as a result of the organoid's nature and the high cell-cell contact [40,41].Finally, the usual and low apoptosis and necrosis values observed in CD-dECM-containing groups were thought to be due to the relatively uniform distribution of nutrients and gases, thanks to the 3D irregular structure evident in microscopy findings.
After examining the effects of CD-dECM doping at different rates on the organoid formation process, cell proliferations in organoids were examined during the 7 d culture period and the relevant findings were presented in figure 5(A).The close proximity of the absorbance values on the first day of culture and the low standard deviations of the groups indicate that each of the created hanging drops acquired equivalent cell seeding.By the 3rd day of culture, the highest absorbance value was observed in the group without CD-dECM (control group) (p < 0.0001), while the absorbance values of the other groups containing CD-dECM were observed to be extremely close to each other, regardless of the concentration.On the 5th day of culture, the group without CD-dECM continued to outperform the other groups in terms of absorbance value (p < 0.0001), while the group with the lowest concentration of CD-dECM doping distinguished itself from the groups doped with CD-dECM at higher concentrations (p < 0.01).Interestingly, by the 7th day of culture, a sharp decrease was observed in the absorbance value of the group without CD-dECM and the highest absorbance value was detected in the group containing 0.1 mg ml −1 CD-dECM.This group also showed significantly higher cell viability than the other groups containing CD-dECM at different concentrations (p < 0.01).The possible reason for the loss of absorbance seen in the group without CD-dECM is the uncontrolled growth of the organoid and the inadequacy of the distribution of nutrients and gases due to mass transfer in the residual organoids [42].Tumors inside the body are complex and dynamic tissues comprised of many biological structures such as cancer cells, stromal cells, ECM, blood and lymph vessels [43].Physical stimuli in the tumor microenvironment such as hydrostatic pressure, shear stress, compression, and tension have profound effects on tumor cells, the primary component of tumors [43].As for the studies that tried to produce tumor conjugate under in vitro conditions, it has been reported that the lower the viscosity, shear stress and mechanical stress of tumor cells under environmental conditions, the better they exhibit proliferation [44][45][46].In the light of these literature findings, it was assumed that CD-dECM doping increased the viscosity and mechanical stresses in the cells' environment and hence the proliferation rate reduced as the amount of CD-dECM rose, in this study.
Glucose consumptions were also evaluated in parallel with the cell proliferation in the organoids, and the findings were presented in figure 5(B).While the amount of glucose uptake on the 1st day of culture was quite similar in all groups, by the 3rd day of the culture, the glucose uptake in the group without CD-dECM was found to be significantly higher than in all groups (p < 0.05).While a similar situation was observed on the 5th day of the culture, the glucose uptake of the group containing 0.1 mg ml −1 CD-dECM was found to be significantly different than the groups with higher concentrations (p < 0.05).By the 7th day of the culture, the glucose uptake value of the group without CD-dECM showed a sudden decrease and was found to be lower than all other groups (p < 0.0001).On the other hand, glucose uptake value of the group containing 0.1 mg ml −1 CD-dECM was recorded as the highest value among all groups (p < 0.0001).All findings regarding glucose consumption were found to be directly compatible with data on cell proliferation within the scope of the study, confirming the viability of the cells.
In later stages of this research, the morphologies of CD-dECM doped and undoped organoids were assessed afterwards in the study, and the results are shown in figure 6. Examining the organoid size growth in the later days of culture, it was determined that the groups lacking CD-dECM established larger organoids faster than those possessing it.Another striking finding was that the cell-rich capsule-like structures generated at the start of culture in CD-dECM groups were never destroyed during culture.While this was clearly observed especially at high concentrations such as 1 mg ml −1 and 1.5 mg ml −1 , it turned out that at lower CD-dECM concentrations, the organoid structure may change due to cell proliferation during culture.On the 7th day of culture, the morphological images of the organoid structure without CD-dECM and the organoid structure with 0.1 mg ml −1 CD-dECM were presented comparatively in figure 6(A).While the structure of organoids without CD-dECM showed a more spherical and well-defined structure, organoids containing 0.1 mg ml −1 CD-dECM were smaller in size and had a complete 3D cellular organization, with gaps in places.The rationale for this variation in cellular organizations in organoid structures is that the group lacking CD-dECM had entirely intercellular contacts, whereas the group with CD-dECM possessed both cell-cell and cell-ECM protein interactions.In organoids consisting of only cells, a significant increase in size was encountered as a result of excessive cell proliferation due to most probably the loss of contact inhibition.However, this uncontrolled cell proliferation was also resulted in cell death, and the images of dead cells pouring from the organoids were observed apparently in figure 6(A) (left).Organoids were also formed from cells marked with CellTracker Green, and organoids obtained with the help of fluorescent microscopy during 7 d of culture were analyzed and the relevant images are shared in figure 6(B).Fluorescence microscopy findings were also confirmed the light microscopy, and the parameters such as 3D organizational difference, organoid size difference, and organoid circularity difference between groups with and without CD-dECM were clearly visible with fluorescent staining.
In hanging drop technique, the substances added to the droplets have a direct impact on the rate of organoid formation, the quality of the organoids, their lifetime, and their durability [47].In the absence of a standardized scale for assessing the morphology of organoids in relation to their functionality, the researchers assessed the circularity and compactness of the organoids by visually reviewing the micrographs of the organoids [48].In this study, compactness parameter was evaluated with features from weak to strong such as dispersed, loose aggregate, tight aggregate with no clear border, compact clear border with come loose cells on periphery and tight aggregate-fully remodeled outline, respectively while the circularity parameter was evaluated by defining features such as dispersed, irregular with major concave contour, irregular with minor concave contour, elongated with no concave contour and circular, in the same order.In this presented study, CD-dECM doped organoids were evaluated as 'loose aggregate' in terms of compactness as well as 'irregular with minor concave contour' in terms of circularity, while those without CD-dECM were evaluated as 'tight aggregate-fully remodeled outline' and 'circular' in terms of compactness and circularity, respectively.Although organoids produced entirely from cells, not doped with CD-dECM, seem to be superior to others in terms of compactness and circularity, many studies in the literature have reported the negative effects of collagen and ECM-like molecules doped into the hanging droplet on organoid morphology, especially when used at increasing concentrations [49][50][51][52].Nevertheless, when assessed beyond morphological criteria, these structures were classified as organoids due to their enrichment with their own ECM and the irregular polarization and morphology of these organoids, resulting from their high protein content in the ECM, were found consistent with the previous literature findings [53,54].
One of the analyses to assess organoid quality by morphometric analysis is to use Image J to calculate variables such as roundness, circularity, organoid area and stiffness from the generated micrographs.The changes in the organoid structure in the presence and absence of CD-dECM during 7 d of culture were examined and presented in figure 6(C).The roundness of a organoid is a measure of the circularity of its projected area, ranging from 0 to 1; the closer to 1, the more circular the organoid [20].The roundness values of organoids containing CD-dECM were calculated as 0.870, 0.721 and 0.884 on the 1st, 4th and 7th days of culture, respectively, while the roundness values of organoids without CD-dECM were calculated as 0.928, 0.981 and 0.935, respectively.When the results were analyzed, it became clear that organoids grown in the absence of CD-dECM had roundness values very close to 1 and were therefore acceptable as round, while those grown in the presence of CD-dECM had roundness values that were significantly lower to begin with and could only approach 0.90 over the same course of culture.In terms of sphericity index values, tumor aggregates may be categorized as spherical (SI ⩾ 0.90) or non-spherical (SI ⩽ 0.90), with the latter classified into ellipsoidal, '8'-shaped, and irregular shapes [55].In this case, since sphericity indexes in both groups were below 0.90 in this study, the produced organoids were considered irregular.However, it was concluded that the reason for the irregularity of both groups is completely different from each other.While it was thought that the sphericity index decreased over time due to apoptosis or necrosis-induced cell death in the group without CD-dECM, it was found remarkable that the sphericity index increased over time in the group containing CD-dECM.As for solidity, it indicates the regularity of the surface of the organoids, and tumor organoids are deemed regular if their solidity values are greater than 0.90 and there are no surface edges [56].The solidity values of organoids containing CD-dECM were calculated as 0.539, 0.595, and 0.719 on the 1st, 4th, and 7th days of culture, respectively, while the solidity values of organoids without CD-dECM were calculated as 0.627, 0.880, and 0.849, respectively.Finally, the calculation of the organoid area yields a clue about the organoid growth indirectly [57].The calculated area values of organoids containing CD-dECM were noted as 0.181, 0.211, and 0.310 on the 1st, 4th, and 7th days of culture, respectively, while the values of organoids without CD-dECM were calculated as 0.181, 0.308, and 0.667, respectively.As was previously mentioned, organoids devoid of CD-dECM grew about six fold in size within 7 d due to the absence of contact inhibition and maximal cellcell interaction.The size of organoids generated in the presence of CD-dECM increases nearly thrice during the same culture phase.
Literature reviews reveal that additives introduced into the hanging drop during the organoid formation process often increase the viscosity, and that the organoid morphology is degraded or many polygamous organoids are formed depending on the increased viscosity.This is because enhanced viscosity nullifies gravitational forces that would ordinarily let cells to settle on the droplet's tip [58].In a study testing the ability of HepG2 cells to produce organoids in culture medium containing varying quantities of Matrigel (0.5 mg ml −1 to 4 mg ml −1 ) and methylcellulose (0.5% to 2%), organoid formation was reported to be decreased as the concentrations of both additions increased [59].In another recent work where Matrigel was employed from 0.3 mg ml −1 to 9 mg ml −1 , it was inferred that high concentration of ECM reduces the advancement of aggregation to produce organoids which inhibits ECM condensation throughout aggregation via gelation [60].In another study in which collagen type I and MethoCel were employed as media additives to enhance organoid formation, it was found that as the collagen concentration increases, organoid structures such as circularity and compactness deteriorate [48].In light of these examples from the literature, the results of this study are comparable, and as the concentration of CD-dECM doped as matrix increases, structures resembling capsules filled with cells and surrounded by ECM, distant from the actual organoid structure, were obtained due most probably to the increased viscosity.
The gelation kinetics of the hydrogels were revealed utilizing a turbidimetric assay based on the rise in turbidity encountered during the self-assembly organization of collagen, which was employed also in this study and whose gelation kinetic was well known from previous studies.To begin, turbidimetric analyses of CD-dECM hydrogels obtained within the scope of the study were performed at various concentrations, and the optimum and lowest concentration necessary for spontaneous gelation was identified, as shown graphically in figure 7.There was no turbidimetric change detected in CD-dECM at 2 mg ml −1 , and it was assumed that the molecules in the dECM were not sufficiently structured for cross-linking.A sigmoidal curve indicating turbidimetric gelation and cross-linking was found in the hydrogels when the dECM concentration was doubled, to 4 mg ml −1 [61].As the dECM concentration was raised to 8 mg ml −1 , however, a significantly more prominent sigmoidal graph occurred, and the maximum optical density increased 2.5 times when compared to 4 mg ml −1 .Due to the extensive culture period required and the necessity to conduct simultaneous culture studies in tens of petri dishes, obtaining CD-dECM is often quite challenging [62].For example, within the scope of this study, a 1-week culture was conducted using approximately 14-15 T75 flasks with an initial seeding density of 5 × 10 5 to yield 10 mg dry powder CD-dECM.As the gelation was observed at a concentration of 8 mg ml −1 at this stage of the investigation, it was deemed unnecessary to test higher concentrations, taking into account the bottlenecks such as long culture period and high consumable costs.
On the other hand, identical turbidimetric studies were undertaken on the following hydrogels intended for use as 3D culture milieu: CD-dECM (8 mg ml −1 ), rat tendon collagen type I (4 mg ml −1 ) (Neuromics, USA), and their combination (1:1 (v/v)) and the findings were illustrated in figure 8 and table 1.
All of the samples had a sigmoidal profile and gelled following a lag period, denoted by t lag .The t lag values for collagen based hydrogel, CD-dECM and the combination of both were calculated as 7.45 ± 0.55, 29.15 ± 0.85 and 15.32 ± 0.71, respectively.It was found noteworthy that the CD-dECM groups had a longer lag phase compared to collagen.The gelation rate of CD-dECM was also found to be approximately 3 times lower than the groups containing collagen.The low gelation rate in the CD-dECM group was found consistent with the longer lag time.On the other hand, in the group with collagen doped CD-dECM structure, it was found that both the gelation speed and half of the time to complete the gelation were similar to those in the group containing pure collagen.This is an indication of how collagen doping contributes to the gelation process of the CD-dECM-based hydrogel [63].In fact, the decrease in the lag time, which was calculated as 29.15 ± 0.85 in  In another part of the study, the use of hydrogels derived from CD-dECM in organoid culture was demonstrated both alone and in combination with collagen type I by using microscopic examination, cell proliferation rate and volumetric outgrowth calculation and the relevant findings were given as figure 9. First of all, it was seen that all groups successfully allowed organoid culture and the volumetric growth of MCF-7 organoids was found consistent with the existing literature [64].However, organoids cultivated in CD-dECM were found to be prominent in organoid growth compared to other groups, as determined by microscopic examination, volumetric outgrowth calculation, and cell proliferation findings.CD-dECM hydrogels were found flexible enough to allow cells in the proliferative zone on the outside of the organoid to move through the matrix while being sturdy enough to keep the organoid fixed in the culture.On the other hand, it was thought that oxygen and nutrients are dispersed evenly enough throughout the matrix to support cell proliferation.This may be due to the rich matrix membrane proteins, hormones and soluble growth factors content of atolog CD-dECM.Although collagen type I is the main component of ECM structure, CD-dECM derived hydrogel, which is rich in other non-collagen components, may have come to the fore in organoid culture.Collagen Type I has been successfully demonstrated in the culture of MCF-7 organoids so far, but its physical and mechanical properties such as matrix stiffness should be well reviewed and it is a material that may need enrichment in this direction [65][66][67].As a result, although their mechanical properties were not investigated in this study, CD-dECM hydrogels may be ideal candidates for integrating collagen type I hydrogels.
In the literature, natural tissue conjugates such as organ or tissue decellularized matrices or collagen, gelatin, Matrigel, alginate have been suggested as ideal structures for scaffold based 3D culture [68][69][70][71][72][73][74][75][76].The autologous cell derived extracellular matrix developed within the scope of this study can be also a good resource that would be used as an alternative to these materials or in combination with them.In addition to the promising findings revealed in this study, there are shortcomings and parts that need to be improved: (1) The content of CD-dECM obtained (matrix membrane proteins, soluble growth factors, hormones, other biomolecules) may vary from batch to batch, cell type to cell type, and even the passage number of the cells used.This may cause inconsistencies in experimental findings.(2) CD-dECM acquisition can be a timeconsuming, expensive, and labor-intensive process.(For example: it has been noted in this study that approximately 1.35 ± 0.25 mg of dry powder of CD-dECM can be produced from 1 T75 flask).(3) The hydrogels and their combinations obtained within the scope of the study were not evaluated mechanically.The physical and mechanical compatibility of the obtained structures to the tumor microenvironment needs to be addressed in future studies.(4) Within the scope of this study, only basic cellular functions were evaluated and advanced analyzes (cell specific protein secretion, gene expression, etc) are needed.

Conclusion
CD-dECM from MCF-7 breast cancer cells was obtained and characterized.In the next stage, these obtained matrices were prepared at varying concentrations and tested in organoid preparation and organoid culture in an effort to more accurately mimic the tumor microenvironment.The results indicate that produced CD-dECM can be used successfully in organoid production at low concentrations and in organoid culture at high concentrations.These autologous, biologically active, biocompatible, and non-immunogenic hydrogels can be both proposed as an alternative source of hydrogels and used in conjunction with the hydrogels currently being proposed.Moreover, these hydrogels can be degraded by different enzymes during subsequent stages.The organoids enclosed within the hydrogels can then be retrieved and utilized for various downstream analyses, including flow cytometry, molecular analyses, and phenotypic characterizations.Lastly, the in vitro methodologies showcased in this study is believed to present encouraging strategies for personalized medicine and preclinical research.

Figure 1 .
Figure 1.The schematic representation of the preparation, characterizations and in vitro applications of MCF-7 tumor cell derived ECM.
(2) bright green or dense orange areas of chromatin condensation in the nucleus, apoptosis; and (3) dark orange or red intact nucleus, necrosis.The following equations were used to quantify apoptotic and necrotic cells: Apoptotic Cells (%) = Total number of apoptotic cells Total counted cells × 100 (1) Necrotic Cells (%) = Total number of necrotic cells Total counted cells × 100.(2)

Figure 3 .
Figure 3. Examination of the effects of the addition of different concentrations of CD-dECM on organoid formation during the 7-day culture period by light microscopy.The group without CD-dECM defines spheroids, and the group containing CD-dECM defines organoids.As the rate of CD-dECM increases, a capsule-like structure surrounds the cells from the outside; this is represented by a blue arrow.In contrast to spheroids formed solely from cellular interactions, organoid structures with irregular morphology and polarization are considered to arise from the participation of ECM structure in intercellular interactions; cellular organizations are indicated by the red arrow.

Figure 5 .
Figure 5. (A) MTT analysis results, and (B) glucose uptake results of organoids containing different concentrations of CD-dECM during 7-day of culture.

Figure 9 .
Figure 9. Optic microscope images of organoid embedded into different types of hydrogels during 7 d cell culture period.