A single-chain derivative of an integrin-activating antibody potentiates organoid growth in Matrigel and collagen hydrogels

Main

Organoids are self-organizing, miniature versions of epithelial tissues that can be expanded over long time periods. Since the 2009 report of a 3D technique to culture adult stem cell-derived intestinal organoids¹, protocols have been published for the isolation of organoids representing epithelial tissues derived from all three germ layers¹. However, efforts to apply organoids in areas such as drug screening and clinical transplantation have been hindered by limitations of the culture conditions, which typically include xenogeneic, chemically undefined or non-standardized components.

One essential element of these culture conditions is a growth factor cocktail mimicking the stem cell niche. This typically has to be optimized for each individual tissue, yet always contains three key ingredients: potent activators of the Wnt pathway (for example, Wnt3A or its surrogate, plus R-spondins), a tyrosine kinase receptor activator (for example, EGF) and inhibitors of TGFβ and/or BMP (for example, noggin).

A second critical component for proper organoid formation is a cell culture substrate consisting of extracellular matrix (ECM) proteins, which in vivo are presented by the basement membrane² (Fig. 1a). This thin layer of a specialized ECM forms the supportive structure for epithelia and includes various laminins (Lns), collagen type IV and, to a lesser degree, arginine–glycine–aspartic acid (RGD) motif-containing molecules such as fibronectin. Lns are heterotrimeric proteins assembled from one of five α chains, one of three β chains and one of three γ chains. In higher organisms, these isoforms show varying degrees of cell and tissue specificity. Collagen IV is a heterotrimeric molecule composed of three α-type chains, which form polygonal networks through several types of intermolecular interactions. Both collagen type IV and Lns are produced in vivo by epithelial cells and form independent networks that are connected by nidogen and the heparan sulfate proteoglycan perlecan. A highly concentrated blend of these proteins, approximating the composition of the basement membrane, is produced by the mouse Engelbreth–Holm–Swarm tumor after inoculation into the abdomen of mice³. Its major component, embryonic α1β1γ1 Ln (Ln-111), has essential roles in the polarization and differentiation of epithelial cells⁴. The ECM protein collagen type IV is another major component. Engelbreth–Holm–Swarm tumor extracts, collected from mice, are marketed under the names Matrigel and basement membrane extract (BME). Both are enriched for ECM and, especially Matrigel, have been used extensively for culturing epithelial cells¹, including 3D structures from mammary epithelium⁴. Matrigel has also enabled the establishment of adult stem cell-based organoid technology⁵. The use of Matrigel is hampered by its cost, batch-to-batch variability, contamination by mouse proteins and ethical considerations on the use of experimental animals. Clinical-grade Matrigel has not been available.

**Fig. 1: Schematic of integrin functionality in epithelial cells in vivo and in vitro.**

The contact that epithelial cells make in vivo with ECM proteins (or in vitro with Matrigel) is mediated by αβ1 integrins⁶. The specificity of the interaction with individual ECM components is determined by the various α chains, while the resulting intracellular signals are transduced by the β1 chain. Interactions with Lns proceed via α3β1 and α6β1 (ref. ⁷). The GFOGER motif in collagen allows adhesion in conjunction with α1β1 and α2β1 integrins⁶. A third class of ECM proteins, which includes fibronectin, interacts with α5β1 integrin through the short RGD sequence.

Various studies have described synthetic hydrogels (for example, polyisocyanide (PIC)⁸ or polyethyleneglycol (PEG))⁹, functionalized with integrin-binding motifs taken from ECM proteins. The individual chains of heterotrimeric laminins act cooperatively in binding to integrins with high affinity, and as a consequence, no simple binding motifs have been defined. Therefore, recombinant full-length Ln-111 was coupled to a PIC gel¹⁰. The in vitro synthesis of trimeric laminins in mammalian cells is, however, complex and expensive. By contrast, several synthetic PEG gels have been described in which the GFOGER peptide, alone or in conjunction with the RGD adhesion motif, maintains gastrointestinal organoids, albeit less efficiently than the Matrigel benchmark¹¹. Similar results were obtained with ductal pancreas organoids¹².

To improve the growth of human intestinal organoids in hydrogels, we were inspired by allosteric antibodies activating the integrin β1 chain operating on leukocytes, first described approximately three decades ago¹³. Like epithelial cells, leukocytes express several β1 integrins, which mediate interactions with ECM components as well as other cells. Leukocyte adhesion is a highly regulated and essential process in the immune response and inflammation. The human integrin β1-specific TS2/16 antibody¹⁴ strongly enhances the adhesion by α2β1, α5β1 and α6β1 integrins with their respective ECM ligands, as well as of α4β1 to endothelial VCAM protein. A similar TS2/16-induced increase in cell-substrate adhesiveness has been observed for non-hematopoietic cells: for example, neuronal cells in defined stages of chicken embryo development were induced to bind to Ln in vivo in the presence of the β1-activating antibody TASC¹⁵.

The non-constitutive nature of integrin interactions implies that ligand binding requires structural changes in the integrins themselves. This was first shown for the αVβ3 and αIIbβ3 integrins¹⁶. Electron microscopy (EM) studies subsequently revealed that the β1 integrins adopt similar conformational changes¹⁷. Integrins are known to exist in three formational states¹⁶: two low-affinity states, bent–closed and extended–closed, and one high-affinity state characterized by an extended conformation with an open ligand binding domain (headpiece). A variety of β1 subunit-specific antibodies have been used to induce or stabilize these conformational states and measure their affinity (Supplementary Fig. 1). For example, the anti-integrin ß1 blocking antibody AIIB2 allosterically stabilizes the bent, low-affinity conformation of the β1 integrin chain¹⁸. By contrast, TS2/16, binds to closed headpieces in the absence of ligand¹⁷ and stabilizes the open, high-affinity headpiece conformation in the presence of ligand^14,17. This well-described ability of TS2/16 to enhance leukocyte adhesion inspired us to explore its properties in epithelial organoid cultures. We designed a single-chain variable fragment (scFv) of the original TS2/16 antibody that similarly activates β1 integrins with the additional advantage of being expressed in bacteria at high yields. The potency of this designed integrin activator in organoid culture is shown across various human epithelial organoids in 3D and 2D in Matrigel and collagen I hydrogels.

Results

Integrin β1 is indispensable for organoid growth

Adult stem cell-derived intestinal organoids, showing a basal-out phenotype in the presence of external Matrigel, invert their polarity to basal-in when cultured in the absence of Matrigel¹⁹. We tested the role of integrin β1, dominantly expressed in various organoid models (Extended Data Fig. 1 and Supplementary Fig. 2), on polarity and associated physiology. The rat anti-integrin β1 antibody AIIB2 has previously been shown to allosterically block the recognition of β1 integrins for all three major integrin ligand classes (Ln, collagen and RGD-class proteins)¹⁸. This antibody enabled us to comprehensively test the role of the β1 integrin component in basal–apical polarization of diverse organoid types. Figure 1b–f summarizes the strong effects of adding the AIIB2 antibody to human colon organoids. The ‘basal-out’ phenotype of these organoids, cultured in the presence of BME, inverted within 12 h to a basal-in phenotype following exposure to AIIB2 (Fig. 1b–f). This condition resulted in an immediate abrogation of growth, measured as a strong decrease in the percentage of cells expressing the nuclear proliferation marker Ki67 (Fig. 1d), in visual growth kinetics (Fig. 1e) and in a quantitative cellular ATP assay (Fig. 1f). These observations implied that integrin-mediated signals normally maintain growth and are blocked by AIIB2. Further confirmation for cell cycle exit in response to AIIB2 was seen during a time-course experiment on 2D BME-coated transwells loaded with a human ileal organoid line, stably expressing the fluorescent ubiquitination based cell cycle indicator (FUCCI) reporter(Extended Data Fig. 2a,b). A quantitative PCR analysis, performed on 3D BME-grown duodenal organoids, indicated that AIIB2-arrested cells differentiate into enterocytes and goblet cells (Extended Data Fig. 3).

hIgG and scFv versions of mTS2/16 activate cell adhesiveness through integrin β1

We next examined the ability of an activating antibody to functionally activate the integrins at the epithelial-organoid cell surface⁶. We constructed a humanized IgG version and single-chain (sc) derivative of the mouse TS2/16 antibody, the latter for ease of production in bacteria (Fig. 2a). We determined the hybridoma-derived sequences of the variable regions of the heavy (V_H) and light (V_L) chains of the mouse Ab TS2/16 (ref. ¹⁴) and used human codon-optimized synthetic DNA (Extended Data Fig. 4) to clone these into huIgG1 H- and қL-chain vectors. The humanized IgG was subsequently purified using standard chromatography purification (Fig. 2b). For the sc version, the combined V regions of the H and L chains were linked in two orientations (Fig. 2a) and connected via a 3×(GGGGS) linker. Both protein versions (Extended Data Fig. 5) were cloned into a pET expression vector using Escherichia coli codon-optimized DNA. To obtain pure sc versions, a routine protocol for protein extraction, solubilization and refolding was implemented²⁰. On average, we obtained 1 mg pure scTS2/16 per 50 ml of bacterial suspension, sufficient for approximately 7 l of cell culture medium at a final concentration of 5 nM. sodium dodecyl sulfate polyacrylamide gel (SDS-PAGE) analysis of the purified scFv version of TS2/16 (V_H–V_L) is shown in Fig. 2b. We also developed a protocol to produce scTS2/16 in the human cell line HEK293 (Extended Data Fig. 6a–c). The scTS2/16 (V_H–V_L) bound the ectodomain of the β1 subunit in solution, analogous to its parental IgG version (Extended Data Fig. 7a,b). Functionally, both huIgG1 and scFv versions (V_H–V_L) induced adhesion at low nM concentrations of K562 (erythroleukemia) cells to fibronectin, a standard assay for integrin-mediated adhesion (Fig. 2c,d). We decided to focus on the scFv (V_H–V_L) version of TS2/16 (scTS2/16) for subsequent studies.

**Fig. 2: Classical IgG and scFv versions of the activating β1 integrin antibody TS2/16.**

scFv-TS2/16 enhances growth of intestinal organoids in BME

One of the main components of BME is the protein Ln-111. Four different integrins, α3β1, α6β1, α7β1 and α6β4, differentially bind to members of the Ln family of proteins²¹. Integrin α6β1 (ref. ²²) binds to Ln-111, the embryonic Ln present in BME. The organoid-expressed α3β1 and α6β4 integrins bind to Ln-332 and Ln-511/521 (Extended Data Fig. 8), but these are absent in BME. The specific role of α6β1 integrin in Ln-111 binding was confirmed by exogenous expression of integrin α6β1 in K562 cells. Ln-111 recognition acquired following integrin-α6 transfection was evident, but certainly not maximal. It could be strongly enhanced by adding the activating TS2/16 antibody²³.

We tested whether this phenomenon would translate to potentiated adhesion of colon organoid cells to coated BME. We used a series of concentrations of scTS2/16 in the low nM range. The adhesion in the absence of scTS2/16 (11 × 10⁵ relative light units (RLU)) reached its maximum (22 × 10⁵ RLU) at 2 nM (Fig. 3a). The AIIB2 blocking antibody inhibited the spontaneous adhesion of organoid cells, as a validation of this adhesion assay. As BME contains both Ln-111 and collagen IV, we tested the contributions of these components independently in organoid adhesion. Again, we detected maximum enhancement effects of scTS2/16 at 2 nM for the individual components. The inhibitory AIIB2 antibody completely inhibited the adhesion of organoids to both coats (Supplementary Fig. 3). To assess whether the integrin α6 subunit was also crucial in mediating growth of colon organoids in BME, we added the inhibitory integrin α6-specific GoH3 antibody²⁴ while regularly splitting colon organoid cultures. The results shown in Fig. 3b indicated a strong GoH3-induced reduction of growth. We then tested whether activation of the β1 integrins could enhance the growth of colonic organoids. Addition of scTS2/16 to standard 3D colon organoid culture in BME at 5 nM led to a 3-fold increase in colon organoid growth (0.9 × 10⁶ to 2.8 × 10⁶ RLU). This was observed after 10 days of culture by assessment of total genomic DNA, ATP-dependent luminescent light units (Fig. 3c) and bright-field images of the organoid cultures (Fig. 3d). An essential component of integrin adhesion is the presence of cations of manganese (Mn²⁺)^21,25. We supplemented 1 nM MnCl₂-containing, complete organoid medium with 1 mM MnCl_2. The latter serves as the benchmark for maximal integrin adhesion. We then directly compared the Mn²⁺ effects on adhesion and proliferation to 5 nM scTS2/16. While both induced an increase in adhesion, only scTS2/16 potentiated growth (Supplementary Fig. 4b). The proliferation-promoting effects of scTS2/16 are reflected in the increase of Ki67-positive nuclei from 20% in the absence of the scFv to 36% in the presence of the scFv (P = 4.1 × 10⁻⁴) (Supplementary Fig. 5).

**Fig. 3: scFv TS2/16 enhances growth of intestinal organoids in BME.**

Single-cell cloning from primary tissue or from organoids is generally highly inefficient^5,26. We tested the effect of scTS2/16 on the plating efficiency of such single cells and noted that the clonal outgrowth in the presence of scFv TS2/16 (at 5 nM) improved from 30 × 10³ RLU in the absence of scTS2 to 330 × 10³ RLU in the presence of scTS2 (Fig. 3e,f). By contrast, the outgrowth of organoids, even in the presence of scTS2/16, was completely blocked by adding the inhibitory integrin α6-specific GoH3, emphasizing the importance of the integrin α6 component. The positive effects of integrin β1-stimulatory antibodies on the outgrowth of organoids from single cells were much less strong for the two other β1-chain conformation-stabilizing antibodies, 9EG7 (ref. ²⁷) and 12G10 (ref. ²⁸) (number of clones 15, 35, 20 and 100 at 5 nM for the control, 12G10, 9EG7 and TS2/16, respectively (Supplementary Fig. 6). The TS2/16-stabilized high-affinity conformation of β₁ integrins on leukocytes enhances intercellular (VCAM-1)²⁹ and cell-to-ECM adhesion^17,30. Our data show that an scFv version of the antibody also improves the establishment and expansion of epithelial organoids by the enhancement of α6β1-mediated interaction with Ln-111/collagen IV as presented in the BME.

BME-established organoids from different regions of the gastrointestinal tract also rely on integrin β1

We then tested whether the observed integrin β1-regulated proliferation of intestinal organoids in BME also held for organoids derived from other regions of the human gastrointestinal tract: duodenum (N = 2), J2 jejunum (N = 2) and N39 ileum (N = 2), stomach organoids³¹ (N = 1) and ductal-type organoids from the pancreas³² (N = 2) and liver³³ (N = 2). The pertinent organoids were grown in BME supplemented with control medium or with 5 nM of either the inhibitory mAb AIIB2 or the stimulating scTS2/16 antibody. AIIB2-induced inhibition and scTS2/16 enhancement of growth (compared with the medium without antibody) were observed in all cases (fold increases in examples shown in Fig. 4 were 5-fold (stomach), 2.2-fold (duodenum), 1.6-fold (liver), 8.1-fold (pancreas), 1.5-fold (jejunum) and 1.4-fold (ileum)) (Fig. 4). To exclude major morphological changes resulting from the use of scFv, high-magnification confocal images of 3D BME-scTS2/16-grown jejunal organoids are presented in Supplementary Fig. 7. To assess whether scTS2/16-stimulated growth caused shifts in the transcriptome of the cells, bulk sequence experiments were performed on control and Ab-stimulated 3D BME cultures of P26N colon, Duo-72 duodenum and J2 jejunum organoids (Extended Data Fig. 9). The statistical analysis revealed no significant changes. We concluded that growth of gastrointestinal tract organoids depends on β1-class integrins and is strongly enhanced by the addition of scTS2/16 to their medium.

**Fig. 4: BME growth of gastrointestinal tract organoids is mediated by β1 integrins.**

scTS2/16 antibody potentiates growth of established and primary isolated colon organoids in collagen I hydrogel

Based on the scFv-generated improvement of gastrointestinal organoids in BME, we tested the hypothesis that organoid growth in an alternative hydrogel would also benefit from the addition of scTS2/16. Collagen recognition by integrins was shown to be highly sensitive to TS2/16 (ref. ³⁴) and has been explored extensively for in vivo and in vitro applications^35,36,37. In a first attempt to enhance growth by scTS2/16 in a collagen type I hydrogel, we focused on colon organoids. A commercially available preformulated isotonic form of collagen I with neutral pH forms a 3D hydrogel by simply warming to 37 °C (PureCol EZ Gel). In total, 4 α-integrin subunits (α1, α2, α10 and α11) harbor a collagen-binding I-domain insert and share the β1 subunit. Of these, we observed the expression of α2β1 in BME-grown colon organoids, both at the RNA level (Extended Data Fig. 1) and by immunofluorescence staining (Fig. 5a). In accordance with these findings, the scTS2/16 amplified adhesion of colon organoid cells toward PureCol EZ Gel (uninduced 7.5 × 10⁵ to 22.5 × 10⁵ when scTS2/16-induced) (Fig. 5b). By contrast, the inhibitory AIIB2 antibody blocked the uninduced and induced components of this adhesion (Fig. 5b). P26N colon organoids showed a ±3-fold increase of their growth in collagen (Supplementary Fig. 8b). Also in these colon organoids, no major morphological changes were observed in confocal images (Supplementary Fig. 8c).

**Fig. 5: scTS2/16 improves integrin α2β1-mediated growth of primary organoids in PureCol EZ Gel.**

We then created null alleles of the gene for the integrin α2 subunit (ITGA2) by CRISPR to confirm the relevance of the collagen-recognizing (α2β1) integrin for colon organoid growth in collagen hydrogels (Fig. 5c). As predicted, ITGA2 gene mutation blocked organoid growth in collagen I, even in the presence of scTS2/16. While BME enabled isolation of the ITGA2^−/− cells, growth was slightly reduced compared with the parental organoids, indicating that BME growth of organoids is at least in part controlled by collagen IV (Fig. 5d).

We then tested fresh colon biopsies for organoid outgrowth in collagen I hydrogel, both in the presence and absence of scTS2/16. Colon tissue from donors (N = 3) was enzymatically processed and epithelial cell suspensions were embedded in collagen I. The addition of 5 nM of scTS2/16-containing medium yielded substantial more organoids (average of 25 clones; n = 4) compared with standard colon medium (average of 6 clones; n = 4) (Fig. 5e). Both BME-established organoid lines, as well as those from primary biopsies, showed difficulty expanding in collagen hydrogel after repeated splitting. A ‘niche-inspired’ medium was recently designed for intestinal organoids³⁸. We tested the contribution of the growth factors IGF1 and FGF2 (taken from this protocol) for application in collagen hydrogel (of note, the relevant receptors are expressed at the mRNA level). Each of these contributed to improved growth. In addition, CHIR and forskolin further enhanced the culture. We compared growth (percentage Ki67) and apical–basal polarity (ezrin and integrin β1, respectively) of an ileum organoid line in three hydrogel conditions: BME, collagen and collagen^scTS2/16. At passage 1, organoids growing in the absence of scTS2/16 lost their apical–basal polarity as well as the proliferation marker Ki67 (Supplementary Fig. 9). We then tested the same culture conditions growing an ileum organoid line harboring fluorescent reporters for all three secretory cell types, enabling monitoring of ongoing differentiation. These N39 triple-reporter organoids could be passaged at least six times in collagen (Supplementary Fig. 10d), but appeared fully dependent on it for the addition of scTS2/16. Cells from passages P0 and P2 showed differentiation toward all three secretory cell types when subjected to differentiation medium. Supplementary Fig. 10a presents fluorescent confocal images of the reporters while Supplementary Fig. 10b provides a quantitative analysis of these reporters based on flow cytometry.

When ITGA2-expressing human liver bile duct organoids (Extended Data Fig. 1) were stimulated by scTS2/16, increased adhesion toward collagen I ensued. Guided by these results, we tested the outgrowth of cholangiocyte organoids in collagen I hydrogels, directly from primary liver biopsies. This mirrored the adhesion results. There were consistently more organoids in the scTS2/16 medium compared with the control medium from 3 to 13 (ctr versus scFv) and from 13 to 22 (ctr versus scFv), respectively (n = 2) (Supplementary Fig. 11). These proved to be epCAM^high, as expected for this cell type. We subsequently quantified the scTS2/16 effects on the newly derived collagen-based cholangiocyte organoids (Fig. 5f). A substantial twofold increase in number of organoids (ctr 15 to 35) was seen for ctr and scTS2/16, respectively.

Based on the principles established in the 3D cultures, we tested whether collagen-based culturing of organoids in 2D would similarly be enhanced by scTS2/16. Such cultures have been described previously, in which intestinal cells were cultured on a monolayer of myofibroblasts³⁹. Other protocols allow short-term human organoid growth⁴⁰ or apply a more complex collagen sandwich technique in combination with the use of small molecules (CHIR and valproic acid) to maintain growth⁴¹.

Here we tested scTS2/16-supported growth of P26N colon, Duo-72 duodenal and J2-jejunal organoids in transwells. These were coated with the same collagen used in 3D. Growth was compared with that in BME-coated inserts under otherwise identical conditions. Results are shown in Supplementary Fig. 12 (P26N), Supplementary Fig. 13 (duodenum) and Supplementary Fig. 14 (jejunum). scTS2/16-supplemented P26N colon cultures produced 2.8-fold to 4.5-fold more tissue, on BME and collagen, respectively. For the Duo-72 duodenum cultures, improvements in response to scTS2/16 were measured both on BME (amplification ±2.4-fold) and collagen (amplification ±2.4-fold). For the jejunum line (Supplementary Fig. 14), these amplification factors were ±7.5 (BME) and ±4.75 (PureCol EZ Gel). Similarly to the 3D cultures, we tested the stability of the transcriptomes of these 2D cultures in the presence of scTS2/16. Bulk RNA sequencing confirmed the stability of these in vitro intestinal cultures (Extended Data Fig. 10).

Given these positive short-term 2D results, we then tested whether these effects would be maintained in long-term culture. We first cultured Duo-72 duodenum organoids in collagen-coated transwell inserts in the presence and absence of scTS2/16 (Fig. 6a). We passaged both types of cultures four times, restarting with a fixed number of single cells at each round. The experiment was quantified by scoring the number of cells in inserts at the end of each passage. By extrapolating these values over four passages, we found that approximately 15 times more cells were produced in the presence of scTS2/16. For the second experiment (Fig. 6b), we cultured J2 jejunum organoids in collagen-coated inserts for six passages in the presence of scTS2/16. We then tested the competency of these cells to differentiate to the major specialized intestinal cell types. Cells, sequentially for 2 weeks in patterning⁴² and 2 weeks in IL22⁺ maturation⁴² media (both scTS2/16⁺), had retained the capacity to differentiate toward enterocytes, goblet cells, Paneth cells and enteroendocrine cells (Fig. 6b). The results were found to be indistinguishable from the same cells grown under the same conditions but on a BME coat.

**Fig. 6: Long-term 2D growth and differentiation of intestinal organoids.**

Discussion

Here we exploit the functional importance of integrin β1 in epithelial organoid cultures derived from the human small intestine, colon, stomach, pancreas and liver. We describe an sc integrin-activating antibody, scTS2/16, that is easily produced and strongly enhances organoid outgrowth. We observe potent growth-blocking effects of the inhibitory integrin β1 subunit-specific antibody AIIB2, while the activating integrin β1 subunit antibody TS2/16 enhances organoid growth. The agonistic scTS2/16 binds the human β1 subunit with high affinity. Moreover, scTS2/16 is highly potent in supporting the establishment and growth of organoids in the standard BME hydrogel, simply by addition to the culture medium as a soluble monomer. We observe a crucial role for the integrin α6 component in BME-dependent growth. We currently believe that organoid growth in BME results from the parallel interaction of α2β1 with collagen IV and α6β1 with Ln-111, both natural ECM components present in BME. We also find that scTS2/16 supports the growth of organoids in hydrogels of purified collagen I. CRISPR-mediated ITGA2-gene knockout showed the α2β1 integrin to be responsible for this effect. Our 2D protocol reveals that scTS2/16 enables potent, long-term culturing of intestinal tissue on collagen coats. The tissue generated under these conditions remains competent to differentiate into secretory Paneth cells and cells participating in the absorption of nutrients (enterocytes) and the production of mucus (goblet cells) and hormones (enteroendocrine cells).

The TS2/16 antibody reinforces adhesion and growth in organoids most probably by providing a counterforce to contractility of the cytoskeleton. A strong indication of this is the inhibition of collagen contraction by TS2/16 (ref. ⁴³), identical to the effect of Rho-associated coiled-coil containing protein kinase (ROCK) inhibitors⁴⁴. Involvement of ROCK inhibition in growth was shown in primary keratinocytes, which are conditionally transformed following prolonged exposure to small-molecule inhibitors of the ROCK kinase⁴⁵. Functional evidence for the initiating role of integrin–ligand signaling in these effects was nicely shown by deleting ITGB1 in preimplantation epiblasts⁴⁶. The versatility of scTS2/16 arises most probably from the fact that the integrin α chains of the ECM proteins Ln and collagen I or collagen IV share the β1 signal transduction chain.

Several earlier studies describe well-defined hydrogels for gastrointestinal organoids. Generally, these studies have used organoids established in Matrigel as starting material. The first two involved synthetic PEG gel^11,12, functionalized with ligand mimetic peptides PHSRN-RGD (Fn) and GFOGER (Col), combined with a mix of ECM-sequestering peptides. In one of the studies¹¹, growth of human small intestinal organoids was successful, although the gel did not match the Matrigel benchmark of the organoid emergence rate and overall proliferation. The collagen mimetic GFOGER peptide clearly dominated in driving proliferation over the PHSRN-RGD peptides. This is in line with our collagen results and with fibrinogen hydrogel findings by others⁴⁷. The second study describes growth of normal mouse ductal pancreas organoids, and mouse and human adenocarcinoma organoids¹². In these three organoid types, growth was again largely dependent on the GFOGER peptides. Notably, growth was reduced in the presence of antagonists of laminin–α3β1 or α6β1 interactions in adenocarcinoma-type organoids. It is unclear whether this Ln-driven growth is a characteristic of the adenocarcinoma status, is specific for mouse species or is a result of the PEG-bound ECM-binding peptides retaining cell-secreted laminins. It thus remains unclear whether normal human pancreas organoids would also grow in this gel. An obvious possibility is to supplement such systems with scTS2/16, thereby improving integrin interaction with both ligand mimetics and cell-secreted ECM proteins. A PIC hydrogel, decorated with high concentrations of Matrigel isolated laminin–entactin, or recombinantly produced laminin-111, successfully maintained human liver organoid cultures¹⁰. It appears unlikely that this PIC-recombinant Ln-111 hydrogel would provide an affordable hydrogel. One previous study has described that human colon organoids can be grown in collagen I directly from primary tissue³⁹. We predict that this approach will benefit from the addition of scTS2/16.

The Yersinia protein invasin is known to activate α3β1, α4β1, α5β1 and α6β1 integrins, but not the collagen receptors α1β1 and α2β1. We have shown recently that an invasin coat replaces the use of Matrigel in 2D organoid sheet cultures⁴⁸. This approach is complementary to the current study: invasin acts fully independently of a hydrogel yet does not work in solution, unlike scTS2/16. Moreover, it replaces the ECM support provided by laminins but not by collagen.

The observation that the activation of β1 integrins by scTS2/16 in leukocytes extends to epithelial organoid cells opens a range of applications. The availability of this reagent, which allows organoids to grow in collagen-only hydrogels, may avoid the need for BME or Matrigel. Moreover, its production in E. coli or in the HEK293 line allows the generation of clinical-grade batches. Clinical-grade collagen preparations are already widely available for tissue repair³⁷ and cosmetic applications. Mouse and human colon organoids have been shown to be transplantable in mice^49,50,51. To date, two ASC-based early-phase organoid clinical trials have been registered: the transplantation of autologous colon organoids as treatment of ulcerative colitis (clinical trial UMIN000030117, Japanese Registry) and the transplantation of salivary gland organoids for treatment of radiation-induced xerostomia (dry mouth syndrome) (clinical trial NCT04593589). The use of scTS2/16 would improve expansion of such transplantable human organoids. Moreover, an scTS2/16-induced high-affinity state of the transplanted organoid cells may instruct these to engraft more efficiently.

Methods

Immunofluorescence analysis of apical–basal polarity and proliferation marker Ki67

P26N human colon organoids were grown for 2 days in BME ± the rat anti-integrin β1 antibody AIIB2 (Developmental Studies Hybridoma Bank (DSHB) catalog number AIIB2, Research Resource Identifier (RRID): AB_528306) at a concentration of 1 μg ml⁻¹. Organoids were collected using cell recovery solution (Thermo Fisher) and fixed in 4% formaldehyde (pH 7.4) at 4 °C for 1 h. Hereafter, organoids were washed with PBS 0.1% Tween (PBST) (2% BSA). Three-dimensional imaging of organoids was performed essentially as described previously⁵². In short, organoids were incubated overnight at 4 °C with primary antibodies binding ZO-I (Thermo Fisher Scientific catalog number 61-7300, RRID: AB_2533938), ITGB1 gene product (R and D Systems catalog number MAB1778, RRID: AB_357909) and Ki67 (Millipore catalog number AB9260, RRID: AB_2142366) or ITGA2-encoded gene (Abcam catalog number ab30487, RRID: AB_775702). Following extensive washing with PBST, binding of primary reagents was visualized by overnight incubation in PBST–2% BSA containing donkey anti-rabbit Alexa 568 (Thermo Fisher Scientific catalog number A10042, RRID: AB_2534017) and goat anti-mouse Alexa488 (Thermo Fisher Scientific catalog number A-11070, RRID: AB_2534114). Phalloidin Atto 747N (Sigma-Aldrich) was used for direct visualization of F-actin. DNA was stained with DAPI D1306 (Thermo Fisher). Organoids were optically cleared in a glycerol–fructose clearing solution before imaging. Organoids were imaged on a Leica SP8 confocal microscope. Analysis was done with ImageJ software.

TS2/16 hIgG1 production

We used a simplified workflow to sequence the variable regions of the mouse TS2/16 hybridoma (American Type Culture Collection (ATCC) HB-243) as described⁵³. A human codon-optimized DNA sequence encoding the V_H and V_L domains of antibody TS2/16 was synthesized by GeneArt (Thermo Fisher) and cloned into pFUSEss-CHIg-hG1 and pFUSE2ss-CLIg-hK, respectively (Invivogen). Expression vectors were co-transfected into Expi293F human suspension cells using ExpiFectamine (Thermo Fisher). IgG protein was purified by affinity chromatography using CaptivAHF protein-A affinity resin (Repligen), followed by size-exclusion chromatography using a Superdex 200 Increase 10/300 GL column (Cytiva).

TS2/16 scFv production in E. coli

E. coli codon-optimized DNA coding sequences of the TS2/16 scFvs (Supplementary Fig. 3) were synthesized by GeneArt (Thermo Fisher) and cloned into a pET21 vector harboring a C-terminal 6x-HIS tag (Novagen) using standard molecular biology techniques. Protein expression induction, using 1 mM IPTG (Sigma), was done in BL21(DE3) cells starting with an optical density (OD)⁶⁰⁰ of 0.6 in standard lysogeny broth (LB) medium. Production was at 28 °C for approximately 18 h. Bacterial pellets obtained were resuspended in lysis buffer (50 mM Tris pH 7.5, 200 mM NaCl, 10% glycerol, 0.4% Triton ×100) and then sonicated (5 cycles; 10 s on, 20 s off). The insoluble fraction, obtained by centrifugation at 20,000 × g for 15 min, was guanidine–hydrochloride solubilized and refolded as described previously²⁰. The partially purified and refolded scFv was further purified using Ni-NTA (Qiagen), followed by size-exclusion chromatography using a Superdex S200 increase 10/300 GL size-exclusion column (Cytiva) in HBS (10 mM HEPES, pH 7.2, 150 mM NaCl). An analysis of the quality of protein purification was performed running samples, under reducing and non-reducing conditions, on a 4–15% TGX precast SDS-PAGE (Biorad) gel. Proteins were visualized using InstantBlue Coomassie Protein stain (Abcam).

scFv TS2/16 production in Expi293 cells

The coding sequence of the TS2/16 scFv (V_H–V_L orientation) containing the artificial human kappa L-chain signal sequence, and C-terminal tags (8×His tag), HRV 3C site (LEVLFQ/GP), Fc-tag (huIgG1) and a stop codon, was cloned into the pcDNA3.1 mammalian expression vector (Extended Data Fig. 4). Expression was performed by transient transfection of Expi293 cells using the ExpiFectamine transfection kit (Gibco) according to the manufacturer’s instruction. After 4 days, the TS2/16 scFv-Fc was captured from the conditioned media with CaptivA⁻HF protein-A affinity resin (Repligen, CA-PRI-0100) and eluted with IgG elution buffer (Thermo Scientific) containing 300 mM NaCl and neutralized with 100 mM Tris pH 8.0 to reach eventually pH 7.0. The sample buffer was exchanged to 1 × HBS (20 mM HEPES pH 7.3, 300 mM NaCl) using Amicon centrifugal filters (MWCO 30 kDa) and treated with PreScission protease (Cytiva) overnight at 4 °C. Next, the Fc-tag was captured with CaptivA⁻HF protein-A affinity resin and the PreSission protease with Glutathione Sepharose 4C (Cytiva). Finally, the TS2/16 scFv was purified on size-exclusion chromatography with Superdex 75 Increase 10/300 GL column (Cytivia) on the ÄKTA pure chromatography system (Cytivia) in 1 × HBS running buffer (20 mM HEPES pH 7.3, 300 mM NaCl). The TS2/16 scFv was then concentrated using Amicon centrifugal filters (Millipore, MWCO 10 kDa) and sterilized by filtration over a 0.22-μm centrifugal filter (Merck). Protein concentration was determined via NanoDrop A280 reading using the respective protein molecular weight and extinction coefficient. Size-exclusion chromatography and SDS-PAGE analysis for final quality control are shown in (Extended Data Fig. 4).

Integrin β1 ectodomain production and immunoprecipitation

We used the procedure reported previously²⁰ to recombinantly produce the extracellular domain of β1 subunit in the absence of any α-subunit. Deletion of the ~30 amino acid specific-determining loop in the I-like domain is essential for stable secretion of this ectodomain. The DNA coding sequence was synthesized by GeneArt (Thermo Fisher) and cloned into vector pcDNA3.1 (Addgene). Expression vector was transfected into Expi293F human suspension cells using ExpiFectamine (Thermo Fisher). Purification of the C-terminally Flag epitope tagged was performed by absorption of secreted protein to M2-agarose beads (Sigma), and elution by excess 3 × Flag peptides (Sigma). Eluted proteins were subsequently subjected to size-exclusion chromatography using a Superdex 200 Increase 10/300 GL column (Cytiva). For immunoprecipitation experiments of the β1 ectodomain, we used ProteinG (Sigma) or M2-agarose (Sigma). All washes and antibody incubations were in standard PBS. The anti-integrin β4 antibody (Abcam catalog number ab110167, RRID: AB_10866385), the anti-β1 antibody (R and D Systems catalog number MAB1778, RRID: AB_357909) and the anti-HUTS4 (Millipore catalog number MAB2079Z, RRID: AB_11214024), M2 anti-Flag (Sigma-Aldrich catalog number F3165, RRID: AB_259529), anti-Flag HRP (Abcam catalog number ab49763, RRID: AB_869428) and anti-6 × HIS HRP (R and D Systems catalog number MAB050H, RRID: AB_357354) antibodies were used. We used 4–15% SDS-PAGE gel (Biorad) under reducing conditions, and PVDF membrane (Immobilon) for blotting. Enhanced chemiluminescence (ECL) (GE HealthCare) was used in the study.

Adhesion assays of K562 cells

Erythroleukemia cell line K562 cells (Thermo Fisher Scientific) were cultured in DMEM (Gibco) with 100 U ml⁻¹ penicillin–streptomycin and 5% FBS (Sigma-Aldrich). Bovine fibronectin (Thermo Fisher) was diluted to 20 μg ml⁻¹ in PBS and incubated overnight at 4 °C in a flat-bottom 96-well plate (Greiner). Subsequently, wells were blocked with 2% BSA in PBS for 30 min at room temperature. K562 cells were resuspended in DMEM/F12 advanced medium (Gibco). Wells were loaded with equal amounts of cells, in the absence of antibodies, or presence of 5 nM of TS2/16 in scFv or IgG format. Following incubation of cells for 90 min at 37 °C, unbound cells were removed by three washes with DMEM-F12 advanced medium. The number of adherent cells was analyzed by bright-field microscopic images and a luciferase-based ATP assay (CellTiter-Glo Promega).

FACS staining integrins

Colon (P26N) and duodenum (Duo-72) organoids were enzymatically processed to single cells and incubated with integrin component-specific antibodies, each directly conjugated to a different fluorophore. Stained cells were forward scatter (FSC) and side scatter (SSC) gated to remove debris and to focus the integrin analysis on live cells. Analysis of antibody staining was performed on 10⁴ cells per analysis and compared with the same unstained cells. Stainings were performed by using combinations of the directly fluorochrome-conjugated antibodies listed below:

(1)

Catalog number ab 303014, Biolegend, RRID: AB_493580, APC/cyanine 7 anti-human CD29 antibody Ts2/16 integrin β1
(2)

Catalog number ab 313619, Biolegend, RRID: AB_2128022, Pacific blue anti-human/mouse CD49f antibody GoH3 integrin α6
(3)

Catalog number ab 328009, Biolegend, RRID: AB_893368, PE anti-human CD49e antibody SAM-1 integrin α5
(4)

Catalog number ab 343808, Biolegend, RRID: AB_10641282, APC anti-human CD49c antibody ASC-1 integrin α3
(5)

Catalog number ab 328311, Biolegend, RRID: AB_2566271, PE/cyanine 7 anti-human CD49a antibody integrin α1
(6)

Catalog number ab 30486, Abcam, RRID: AB_775704,FITC anti-integrin α2 antibody (AK7)

Mn²⁺ growth assay

Ileum N39 (ref. ⁴²) organoids cultured in BME were released from the matrix by incubation with dispase (10 U ml⁻¹) at 37 °C (ref. ⁵). The organoids were then washed with cold DMEM/F12 medium and centrifuged (5 min, 4 °C, 300 × g) to remove residual dispase. The pelleted organoids were treated with TrypLE (no rho kinase inhibitor) at 37 °C to generate cell clusters. This enzymatic reaction was stopped by adding cold DMEM/F12, followed by centrifugation to pellet the cell clusters. The resulting cell clusters were resuspended in 2 mg ml⁻¹ PureCol EZ Gel or BME and plated as small collagen droplets (10–20 µl). The plate was incubated at 37 °C for 1 h to allow for complete collagen polymerization. After polymerization, the collagen droplets were overlaid with complete expansion medium supplemented with either 1 mM Mn²⁺ or 5 nM scTS2/16. On the first day of the cultures, 10 µM of rho kinase inhibitor Y-27632 was supplemented to the culture media as well. After day 1, the ileum organoids were maintained in expansion medium supplemented with either 1 mM Mn²⁺ or 5 nM scTS2/16, with the medium refreshed every 2–3 days. On day 7, organoid cultures were imaged using bright-field microscopy followed by quantification of viable organoids using CellTiter-Glo.

N39 ileum FUCCI reporter organoid line

The FUCCI sequence comprising mCherry-Cdt1-T2A-Geminin-hmAzami-Green was PCR amplified from a construct provided as a gift by G. Kops (Hubrecht Institute). With the use of a Gibson Assembly (NEBuilder HiFi DNA Assembly), it was cloned into a p2Tol-based vector harboring puromycin resistance (Addgene category catalog number 714859). Reporter organoids were generated by co-electroporation of 5 µg of the FUCCI expression construct together with 5 µg mT2TP transposase mediating the tol2-dependent random integration into the cell’s genome. Stable transfectants were isolated based on puromycin resistance and expression of both FUCCI fluorescent signals.

qPCR for intestinal differentiation markers

RNA was extracted using a QIAwave RNA Minikit (catalog number 74534) according to the manufacturer’s protocol. cDNA was synthesized using the Applied Biosystems High-Capacity cDNA Reverse Transcription Kit. Real-time PCR was performed on the Biorad CFX96 Real Time PCR System using iQ SYBR Green Supermix 1708880.

The primer sets used were the following:

GAPDH	FWD	GAPDH_FWD	AATGAAGGGGTCATTGATGG
GAPDH	REV	GAPDH_RV	AAGGTGAAGGTCGGAGTCAA
B-actin	FWD	B-actin_FWD	CATGTACGTTGCTATCCAGGC
B-actin	REV	B-actin_REV	CTCCTTAATGTCACGCACGAT
ANPEP	FWD	ANPEP_FWD	GCTGTTTGACGCCATCTCCTAC
ANPEP	REV	ANPEP_RV	GTTCTGGTAGGCAAAGGTGTGG
CYP2C9	FWD	CYP2C9_FWD	CAGAGACGACAAGCACAACCCT
CYP2C9	REV	CYP2C9_RV	ATGTGGCTCCTGTCTTGCATGC
CYP2C19	FWD	CYP2C19_FWD	GATCAAAATGGAGAAGGAAAAGCA
CYP2C19	REV	CYP2C19_RV	TCAGCTGCAGTGATTACCAAGT
TFF3	FWD	TFF3_FWD	TCCAGCTCTGCTGAGGAGTACG
TFF3	REV	TFF3_REV	ATCCTGGAGTCAAAGCAGCAGC

ITGA2 CRISPR–Cas9 base editing

For single-guide RNA (sgRNA) plasmid preparation, the following sequence was ordered containing a universal priming part and the sgRNA (underlined): TTGTGTAATACTGATTCCCCGGTGTTTCGTCCTTTCCACAAG. The additional bold cytosine residue is included to obtain sufficient transcription from the U6 promoter. Side-directed mutagenesis in SpCas9-vector 47511 was done using Q5 high-fidelity polymerase with the mentioned primer and the following universal 5′/phos/- GTTTTAGAGCTAGAAATAGCAAGTTAAAATAAGGC primer. Following PCR cleanup, T4-ligase and Dpn1 were used to relegate and generate the plasmid; this was done according to the manufacturer’s protocol (NEB). DNA was transformed into DH5α bacterial cells, and plasmids were verified using Sanger sequencing (Macrogen Europe BV) with sequence primer GGGCAGGAAGAGGGCCTAT.

Organoid electroporation

Organoid electroporation was performed as described previously⁵⁴ with some modifications. Wild-type organoids were maintained in expansion medium. Two days before electroporation, 1.25% (vol/vol) DMSO was added to the organoid medium. Electroporation was done with 10⁶ cells per electroporation with BTXpress solution containing 2.5 μg ITGA2 gene directed sgRNA plasmid, 5 μg plasmid DNA encoding Piggybac transposase, 5 μg plasmid DNA encoding Piggybac hygromycin cassette⁵⁵ and 7.5 μg C>T base editor_eGFP plasmid (Addgene plasmid number 112100). Hygromycin 100 μg μl⁻¹ (Invivogen) selection was started 5 days post-transfection. Then, 14 days after selection, surviving hygromycin selection clones were individually picked and RT-PCR obtained DNA, sequenced with the following primers: ITGA2_fwd GCACCTGCCACCTTCTCCATCA; ITGA2_rev GAAAAGAGGAAAGAAGCAGATT. DNA was obtained using a ZYMO Research DNA isolation kit. During clonal outgrowth and selection, scTS2/16 was always present in the media.

Western blot

Proteins were run on a 4–15% SDS-PAGE gel (Biorad) under reducing and non-reducing conditions. Marker proteins were obtained from Thermo Scientific (26616, Pager ruler). SDS-PAGE gel was transferred to a PVDF membrane (Thermo Scientific reference 88518). Transfer efficiency was controlled with Ponceau staining. For blocking, 5% non-fat milk protein in PBS was used. For direct HIS-tagged protein staining, a mouse anti-HIS tag HRP conjugated antibody (R and D Systems catalog number MAB050H, RRID: AB_357354) was used. Indirect staining for the HIS-tag was performed using anti-HIS (Abcam catalog number S2917, RRID: AB_10641949) in combination with anti-mouse-HRP (R and D Systems catalog number MAB050H, RRID: AB_357354). FLAG-tagged proteins were directly visualized using HRP-conjugated M2 anti-FLAG (Abcam, RRID: AB_869428) or indirectly by combining M2 anti-flag (Sigma-Aldrich catalog number F3165, RRID: AB_259529) and anti-mouse-HRP (Abcam catalog number ab6808, RRID: AB_955441). Chemiluminescent substrate ECL (GE-Healthcare) was used.

Bulk RNA sequencing

RNA was extracted using a QIAwave RNA Minikit (catalog number 74534) according to the manufacturer’s protocol. RNA integrity was measured using the Agilent RNA 6000 Nano kit with the Agilent 2100 Bioanalyzer, and RNA concentrations were determined using the Qubit RNA HS Assay Kit (Thermo Fisher Scientific). RNA integrity number values of RNA samples were typically 9.5–10.0 and never <9.0. RNA libraries were prepared with the TruSeq Stranded messenger RNA polyA kit and single-end (1 × 50 base pairs) sequenced on an Illumina NextSeq 2000. Reads were mapped to the human GRCh37 genome assembly. Differential gene expression analysis was performed using the DESeq2 package in RStudio (v.2022.02.2). Data visualization was performed using the packages ggplot2 and pheatmap in RStudio or manually plotted using GraphPad Prism (v.8.2.0). The coefficient of determination (R²) was determined using either the ggplot2 package in RStudio or GraphPad Prism.

BME and PureCol EZ Gel-based 3D organoid culturing

Organoids grown in BME (Cultrex) were passaged by releasing them first from the matrix by incubation for 20 min on ice in cell recovery solution (Thermo Fisher) supplemented with 10 μM of the Rho-kinase inhibitor Y-27632 (Abmole)⁵. Subsequent DMEM/F12-washed organoids were either mechanically sheared or TrypLE-digested to single cells. Following additional washing with DMEM/F12, organoids were resuspended either in fresh BME or, alternatively, in PureCol EZ Gel (Advanced Biomatrix). For complete gelation of the BME or PureCol EZ Gel gel, plates were incubated at 37 °C for 30 min or 60 min, respectively. Hydrogels were covered with relevant organoid medium and incubated at 37 °C and 5% CO₂. Passaging of organoids growing in collagen was achieved by incubation at 37 °C for 30 min maximally, with collagenase from Clostridium histolyticum (Sigma-Aldrich C0130) diluted in DMEM/F12 to a concentration of 1 mg ml⁻¹. For complete removal of the proteolytic enzyme, organoids were washed twice with an excessive volume of DMEM/F12 medium before re-entry into PureCol EZ Gel. The β1-specific inhibitory AIIB2 antibody was added to organoid cultures at 30 nM. The ITGA6-inhibitory antibody G0H3 was added to organoid cultures as 1:1,000 diluted ascites. The Fab versions of SNAKA51, 9EG7 and 12G10 antibodies were provided by J. Li, Harvard Medical School, Boston Children’s Hospital. Culture images were taken with an EVOS cell imaging microscope.

BME- and PureCol EZ Gel-based 2D organoid culturing

Transwell inserts (Thincerts, 12-well, pore size 0.4 μm, Greiner Bio-One 665641) were coated overnight with 2% dilutions of BME (ref. ⁵) or PureCol EZ Gel in DMEM-F12 medium at 4 °C. Hereafter, inserts were washed twice with DMEM-F12 before the cells were added. Inserts were loaded with approximately ±50,000 cells, prepared as a TrypLE-dissociated single-cell suspension. Following digestion, cells were 40-μm-filtered and washed twice in DMEM-F12 medium in the continuous presence of rho-kinase inhibitor Y-27632 (Abmole). Cells were resuspended in 250 µl relevant culture medium supplemented with rho-kinase inhibitor Y-27632 (overnight) and optionally long term with 5 nM scTS2/16. Subsequently, 1 ml of identical medium was added to the lower compartment. Passaging of the cultures was started by removing media from both compartments, briefly washing with DMEM-F12 and then filling both compartments with TrypLE.

Organoid differentiation media

Patterning medium⁴² is intestinal medium lacking the p38 inhibitor nicotinamide and PGE2, but has 3 μM CHIR 99021 (Stemgent 04-0004-base) and 5 nM scTS2/16.

Maturation medium⁴² is patterning medium lacking CHIR 99021 but having recombinant human IL22 (200-22 Peprotech) at a concentration of 2 ng ml⁻¹. Wnt-Surrogate was present at a 30× lower final concentration (±16 pM).

Organoid differentiation in 2D

For immunofluoresence staining, to allow differentiation of organoids, the intestinal expansion medium in the bottom compartment was replaced for 14 days by ‘patterning’ medium and successively by ‘maturation’ medium⁴². The upper compartment was exposed to air during this period. Hereafter, the medium was removed, tissue 4% PFA-fixed for 20 min, washed twice with excess PBS and permeabilized by incubating in PBS–0.1% Triton X-100–2% BSA for 1 h. Finally, tissue was washed with PBS 2% BSA. For visualization of enterocytes, we stained for the marker ApoA1 by using rabbit antibody PA5-88109 (Thermo Fisher) (RRID: AB_2804657). Goblet cells were visualized by mouse anti-muc2 antibody 11197 (Abcam) (RRID: AB_297837); for enteroendocrine cells, rabbit antibody 23342-1-AP (Proteintech) detecting chromogranin A (RRID: AB_2879259); and for lysozyme secreted by Paneth cells, rabbit antibody GTX72913 (Gene Tex) (RRID: AB_374812). Alexa-conjugated secondary antibodies (1:500) used were donkey anti-mouse Alexa488 (1:500) Thermo Fisher A21202 (RRID: AB_141607) and donkey anti-rabbit Alexa 568 Thermo Fisher A10042 (RRID: AB_2534017). Actin was stained using Phalloidin-Atto 647 from Sigma. DNA was stained with DAPI D1306 (Thermo Fisher). All reagents were diluted in PBS–2% BSA.

PureCol EZ Gel polymerization conditions

Ileum N39 (ref. ⁴²) organoids cultured in BME were first released from the matrix protein using dispase (10 U ml⁻¹)³⁸. The organoids were incubated with dispase at 37 °C for 30 min, after which they were washed twice with cold DMEM/F12 and collected by centrifugation (5 min, 4 °C, 300 × g). The washed organoids were then treated with TrypLE (without rho kinase inhibitor) at 37 °C for 5 min to dissociate the organoids into clusters of cells. To stop enzymatic activity, the cell suspension was washed again with cold DMEM/F12 and centrifuged (5 min, 4 °C, 300 g). The resulting cell pellet was resuspended in 1 ml of one of the following conditions: DMEM/F12, complete expansion medium (EM), EM + Rho kinase inhibitor Y-27632 or EM + Rho kinase inhibitor Y-27632 + scTS2/16. The cell suspension was herein incubated for 30 min (room temperature), allowing conditioning before embedding in hydrogel. After this incubation, cells were centrifuged (5 min, 4 °C, 300 × g) and cell pellets resuspended in 2 mg ml⁻¹ collagen EZcol hydrogel. The suspension was plated in 10–20-µl domes which were then incubated at 37 °C for 1 h to allow polymerization. Hereafter, cultures were covered with EM supplemented with rho kinase inhibitor Y-27632 and 5 nM scTS2/16. The next day, Y-27632 was removed from the medium, and cultures were maintained in EM with 5 nM scTS2/16 to analyze the effect of diffusion of scFv in collagen hydrogels. After 5 days of culture, the number of outgrown organoids was quantified using CellTiter-Glo viability assay (Promega).

Long-term 3D PureCol EZ Gel (Fujii medium)

Ileum (N39) organoids were cultured in expansion medium consisting of DMEM/F12 supplemented with 100 U ml⁻¹ P/S, 10 mM HEPES, 1× Glutamax, 2% B-27 with VitA, 3% (v/v) recombinant R-spondin-3, 2% (v/v) recombinant noggin, 1 mM N-acetylcysteine, 0.5 nM Wnt surrogate, 0.1 µg ml⁻¹ hIGF-1 (Peprotech), 0.05 µg ml⁻¹ hFGF2 (Peprotech), 10 nM gastrin, 0.5 µM A83-01, 0.05 µg ml⁻¹ hEGF, 3 µM CHIR 99021 and 10 µM forskolin (Tocris)⁴². For collagen cultures, ileum organoids grown in BME were first released from the matrix by incubation with dispase (10 U ml⁻¹ for 30 min at 37 °C). Hereafter, organoids were washed twice with cold DMEM/F12 and collected by centrifugation at 300 × g for 5 min at 4 °C. Organoids were then treated with TrypLE (no rho kinase inhibitor) at 37 °C for 5 min and further disrupted mechanically to dissociate organoids into small clumps of cells (avoiding complete single-cell dissociation). Organoids were washed again with cold DMEM/F12, centrifuged under the same conditions and resuspended in 2 mg ml⁻¹ PureCol EZ Gel. The matrix–cell suspension was seeded in 10–20-µl droplets onto a culture plate, which was subsequently incubated at 37 °C for 1 h to allow polymerization. Hereafter, the droplets were covered with the expansion medium supplemented with 10 μM Y-27632 rho-kinase inhibitor (initial 12 h) and 5 nM scTS2/16. The following day, this medium was refreshed, with the expansion medium lacking rho kinase inhibitor but continuing with 5 nM scTS2/16. This medium was refreshed every 2–3 days. For passaging of organoids, the collagen domes were incubated with 1 mg ml⁻¹ collagenase from Clostridium histolyticum (Sigma-Aldrich) for 30 min at 37 °C, and digestion monitored using bright-field microscopy (small clumps of cells). Organoids were collected in a 15-ml tube and washed twice (centrifugation for 5 min at 300 × g) with an excess volume of cold DMEM/F12. The clumps of cells were washed twice and resuspended in fresh PureCol EZ Gel (2 mg ml⁻¹).

3D collagen differentiation protocol

The ileum N39 triple reporter line, with fluorescently marked specific intestinal cell types (goblet cells: Muc2-GFP; enteroendocrine (EEC) cells: CHGA-iRFP; and Paneth cells: DEFA5-dsRED), was described⁴². It was propagated in 3D hydrogel. At both passage 0 (P0) and passage 2 (P2), the organoids were induced to differentiate. Then, 7 days following passaging, the expansion medium was replaced with differentiation medium in all 3 hydrogel conditions: (1) collagen, (2) collagen + scTS2/16 and (3) BME. The differentiation (patterning) medium was prepared as described previously⁴². The differentiation process was monitored daily using EVOS fluorescence microscopy to track the emergence of the differently marked mature epithelial types. After 5 days, mature epithelial cells were observed in the collagen- and BME-cultured organoids. At this point, organoids in all hydrogel conditions were fixed to maintain consistency in imaging and downstream analyses. For quantitative assessment of differentiated secretory cells, we used flow cytometry (BD Fortessa). A total of 5,000 DAPI-negative single cells per hydrogel condition were analyzed to compare differentiation efficiencies to the expansion culture. Organoids were fixed in 4% formaldehyde (pH 7.4) and subsequently analyzed using SP8 confocal microscopy.

Organoid lines

The human small-intestinal duodenum line Duo-72 was described⁵⁶.
Jejunum J2 organoids were provided by M. Estes, Baylor College of Medicine. Organoids were described⁵⁷.
Ileum N39 organoids were described⁵⁸.
The ileum N39 triple reporter was described⁴²
Primary colon biopsies (Fig. 5e, N = 4) were derived from the Utrecht Portal for Organoid Technology. Patients (University Medical Center Utrecht (UMCU)) provided informed consent for the generation of such organoids based on protocol UPORT Cancer Biobank 21-042 and was in accordance with the Declaration of Helsinki and Dutch law. This study complies with all relevant ethical regulations regarding research involving human participants.
P26N colon organoids were described⁴⁷.
Primary liver human ductal organoids (Fig. 5f, N = 3) were derived from UMC biopsy codes 057932, 061530 and 061500. The study was approved by the UMCU ethical committee (TCBio 21-042) and was in accordance with the Declaration of Helsinki and Dutch law. This study complies with all relevant ethical regulations regarding research involving human participants.
The ductal pancreas line, code HUB-08-A2-007, was obtained from the biobank of UMCU (Toetsingscommisie Biobanken TCBio (Biobank research Ethic Committee)) approved by protocol 12-093 HUB-Cancer and informed donor consent.
Established human liver and pancreas ductal lines were described earlier⁵⁹.

All patients participating in this study signed informed consent forms approved by the responsible authority. In all cases, patients can withdraw their consent at any time, leading to the prompt disposal of their tissue and any derived material. Future distribution of organoids to an academic party will have to be authorized by the Medical Ethics Committee of the UMCU (TCBio).

Statistics

All effects measured between two groups, in response to agonistic (TS2/16) or antagonistic (AIIB2 or G0H3) antibodies, such as CellTiterGLO units—number of cells—or fold-change differences in gene expression levels in qPCR, were tested for significance using one-tailed t-tests. P values and the number and type of replicates (designated n) are provided in the relevant legends of the main figures, extended data figures and supplementary figures. Significance of results from bulk sequence comparisons (Extended Data Figs. 9 and 10) were analyzed using a Wald test for negative binomial distribution in the expression between scFv and control samples. Bonferroni correction was applied to adjust for multiple testing. The number of independent identical experiments is designated N in all legends.

Organoid expansion medium

Organoid expansion medium for each organoid type included the following components: a ‘+’ indicates that the component is present in the medium for the corresponding organoid.

Reagent	Source	Concentration	Intestinal	Stomach	Liver ductal	Pancreas ductal
F12 advanced	Gibco	Not applicable	+	+	+	+
HEPES	Gibco	10 nM	+	+	+	+
Glutamax	Gibco	1:100	+	+	+	+
N-acetylcysteine	Sigma-Aldrich	1.25 mM	+	+	+	+
Nicotinamide	Sigma-Aldrich	10 µM	+		+	+
B27	Sigma-Aldrich	1:50	+	+		+
p38-inhibitor SB202190	Sigma-Aldrich	10 µM	+
A83-01 inhibitor TGFβ type I receptor	Tocris	0.5 µM	+	+	+	+
PGE2	Tocris	1 µM	+			+
Gastrin	Tocris	10 nM		+	+	+
Forskolin	Tocris	10 µM			+
EGF	Peprotech	50 ng ml⁻¹	+	+		+
hFGF10	Peprotech	0.01 µg ml⁻¹		+	+	+
hHGF	Peprotech	25 ng ml⁻¹			+
NGS Wnt Surrogate-Fc	Item number: N001 IpA (Immunoprecise)	0.5 nM	+	+		+
R-spondin-3-Fc	Item number: R001 IpA (Immunoprecise)	3%	+	+		+
Noggin-Fc	Item number: N002 IpA (Immunoprecise)	2%	+	+		+
Penicillin– streptomycin	Gibco	100 U ml⁻¹	+	+	+	+
Primocin	Invivogen	50 µg ml⁻¹	+	+	+	+

Reporting summary

Further information on research design is available in the Nature Portfolio Reporting Summary linked to this article.

Data availability

Bulk RNA sequencing data are available at the European Genome–Phenome Archive under code EGAD50000001606. Other data supporting the findings of this study are available at the European Genome–Phenome Archive under code EGAS50000001113. RNA sequencing data presented in Extended Data Figs. 9 and 10 are available via GitHub at https://github.com/Hubrecht-Clevers/Integrin_B1_TS216. Source data are provided with this paper.

References

Clevers, H. Modeling development and disease with organoids. Cell 165, 1586–1597 (2016).

Article
CAS
PubMed

Google Scholar
Bissell, M. J., Hall, H. G. & Parry, G. How does the extracellular matrix direct gene expression?. J. Theor. Biol. 99, 31–68 (1982).

Article
CAS
PubMed

Google Scholar
Orkin, R. W. et al. A murine tumor producing a matrix of basement membrane. J. Exp. Med. 145, 204–220 (1977).

Article
CAS
PubMed

Google Scholar
Li, M. L. et al. Influence of a reconstituted basement membrane and its components on casein gene expression and secretion in mouse mammary epithelial cells. Proc. Natl Acad. Sci. USA 84, 136–140 (1987).

Article
CAS
PubMed
PubMed Central

Google Scholar
Sato, T. et al. Single Lgr5 stem cells build crypt–villus structures in vitro without a mesenchymal niche. Nature 459, 262–265 (2009).

Article
CAS
PubMed

Google Scholar
Bachmann, M., Kukkurainen, S., Hytonen, V. P. & Wehrle-Haller, B. Cell adhesion by integrins. Physiol. Rev. 99, 1655–1699 (2019).

Article
CAS
PubMed

Google Scholar
Pang, X. et al. Targeting integrin pathways: mechanisms and advances in therapy. Signal Transduct. Target. Ther. 8, 1 (2023).

Article
CAS
PubMed
PubMed Central

Google Scholar
Zhang, Y. et al. Polyisocyanide hydrogels as a tunable platform for mammary gland organoid formation. Adv. Sci. 7, 2001797 (2020).

Article
CAS

Google Scholar
Cruz-Acuna, R. et al. PEG-4MAL hydrogels for human organoid generation, culture, and in vivo delivery. Nat. Protoc. 13, 2102–2119 (2018).

Article
CAS
PubMed
PubMed Central

Google Scholar
Ye, S. et al. A chemically defined hydrogel for human liver organoid culture. Adv. Funct. Mater. 30, 2000893 (2020).

Article
CAS
PubMed
PubMed Central

Google Scholar
Hernandez-Gordillo, V. et al. Fully synthetic matrices for in vitro culture of primary human intestinal enteroids and endometrial organoids. Biomaterials 254, 120125 (2020).

Article
CAS
PubMed
PubMed Central

Google Scholar
Below, C. R. et al. A microenvironment-inspired synthetic three-dimensional model for pancreatic ductal adenocarcinoma organoids. Nat. Mater. 21, 110–119 (2022).

Article
CAS
PubMed

Google Scholar
Byron, A. et al. Anti-integrin monoclonal antibodies. J. Cell Sci. 122, 4009–4011 (2009).

Article
CAS
PubMed

Google Scholar
Hemler, M. E. et al. Glycoproteins of 210,000 and 130,000 m.w. on activated T cells: cell distribution and antigenic relation to components on resting cells and T cell lines. J. Immunol. 132, 3011–3018 (1984).

Article
CAS
PubMed

Google Scholar
Neugebauer, K. M. & Reichardt, L. F. Cell-surface regulation of β1-integrin activity on developing retinal neurons. Nature 350, 68–71 (1991).

Article
CAS
PubMed
PubMed Central

Google Scholar
Takagi, J., Petre, B. M., Walz, T. & Springer, T. A. Global conformational rearrangements in integrin extracellular domains in outside-in and inside-out signaling. Cell 110, 599–511 (2002).

Article
CAS
PubMed

Google Scholar
Su, Y. et al. Relating conformation to function in integrin α₅β₁. Proc. Natl Acad. Sci. USA 113, E3872–E3881 (2016).

Article
CAS
PubMed
PubMed Central

Google Scholar
Hall, D. E. et al. The α1/β1 and α6/β1 integrin heterodimers mediate cell attachment to distinct sites on laminin. J. Cell Biol. 110, 2175–2184 (1990).

Article
CAS
PubMed

Google Scholar
Co, J. Y., et al. Controlling epithelial polarity: a human enteroid model for host–pathogen interactions. Cell Rep. 26, 2509–2520.e4 (2019).

Article
CAS
PubMed
PubMed Central

Google Scholar
Arimori, T., et al. Fv-clasp: an artificially designed small antibody fragment with improved production compatibility, stability, and crystallizability. Structure 25, 1611–1622.e4 (2017).

Article
CAS
PubMed

Google Scholar
Nishiuchi, R. et al. Ligand-binding specificities of laminin-binding integrins: a comprehensive survey of laminin–integrin interactions using recombinant α3β1, α6β1, α7β1 and α6β4 integrins. Matrix Biol. 25, 189–197 (2006).

Article
CAS
PubMed

Google Scholar
Sonnenberg, A. et al. Integrin recognition of different cell-binding fragments of laminin (P1, E3, E8) and evidence that α6β1 but not α6β4 functions as a major receptor for fragment E8. J. Cell Biol. 110, 2145–2155 (1990).

Article
CAS
PubMed

Google Scholar
Delwel, G. O. et al. Expression and function of the cytoplasmic variants of the integrin alpha 6 subunit in transfected K562 cells. Activation-dependent adhesion and interaction with isoforms of laminin. J. Biol. Chem. 268, 25865–25875 (1993).

Article
CAS
PubMed

Google Scholar
Sonnenberg, A., Janssen, H., Hogervorst, F., Calafat, J. & Hilgers, J. A complex of platelet glycoproteins Ic and IIa identified by a rat monoclonal antibody. J. Biol. Chem. 262, 10376–10383 (1987).

Article
CAS
PubMed

Google Scholar
Anderson, J. M., Li, J. & Springer, T. A. Regulation of integrin α5β1 conformational states and intrinsic affinities by metal ions and the ADMIDAS. Mol. Biol. Cell 33, ar56 (2022).

Article
CAS
PubMed
PubMed Central

Google Scholar
Sato, T. et al. Paneth cells constitute the niche for Lgr5 stem cells in intestinal crypts. Nature 469, 415–418 (2011).

Article
CAS
PubMed

Google Scholar
Bazzoni, G., Shih, D. T., Buck, C. A. & Hemler, M. E. Monoclonal antibody 9EG7 defines a novel beta 1 integrin epitope induced by soluble ligand and manganese, but inhibited by calcium. J. Biol. Chem. 270, 25570–25577 (1995).

Article
CAS
PubMed

Google Scholar
Mould, A. P., Garratt, A. N., Askari, J. A., Akiyama, S. K. & Humphries, M. J. Identification of a novel anti-integrin monoclonal antibody that recognises a ligand-induced binding site epitope on the beta 1 subunit. FEBS Lett. 363, 118–122 (1995).

Article
CAS
PubMed

Google Scholar
Maqueda, A. et al. Activation pathways of α4β1 integrin leading to distinct T-cell cytoskeleton reorganization, Rac1 regulation and Pyk2 phosphorylation. J. Cell. Physiol. 207, 746–756 (2006).

Article
CAS
PubMed

Google Scholar
Arroyo, A. G. et al. Regulation of the VLA integrin–ligand interactions through the β1 subunit. J. Cell Biol. 117, 659–670 (1992).

Article
CAS
PubMed

Google Scholar
Bartfeld, S. et al. In vitro expansion of human gastric epithelial stem cells and their responses to bacterial infection. Gastroenterology 148, 126–136.e6 (2015).

Article
PubMed

Google Scholar
Boj, S. F. et al. Organoid models of human and mouse ductal pancreatic cancer. Cell 160, 324–338 (2015).

Article
CAS
PubMed

Google Scholar
Hu, H. et al. Long-term expansion of functional mouse and human hepatocytes as 3D organoids. Cell 175, 1591–1606.e19 (2018).

Article
CAS
PubMed

Google Scholar
Weitzman, J. B., Chen, A. & Hemler, M. E. Investigation of the role of β1 integrins in cell–cell adhesion. J. Cell Sci. 108, 3635–3644 (1995).

Article
CAS
PubMed

Google Scholar
Gelse, K., Poschl, E. & Aigner, T. Collagens—structure, function, and biosynthesis. Adv. Drug Deliv. Rev. 55, 1531–1546 (2003).

Article
CAS
PubMed

Google Scholar
Sorushanova, A. et al. The collagen suprafamily: from biosynthesis to advanced biomaterial development. Adv. Mater. 31, e1801651 (2019).

Article
PubMed

Google Scholar
Wang, Y., Wang, Z. & Dong, Y. Collagen-based biomaterials for tissue engineering. ACS Biomater. Sci. Eng. 9, 1132–1150 (2023).

Article
CAS
PubMed

Google Scholar
Fujii, M. et al. Human intestinal organoids maintain self-renewal capacity and cellular diversity in niche-inspired culture condition. Cell Stem Cell 23, 787–793 e786 (2018).

Article
CAS
PubMed

Google Scholar
Jee, J. H. et al. Development of collagen-based 3D matrix for gastrointestinal tract-derived organoid culture. Stem Cells Int. 2019, 8472712 (2019).

Article
PubMed
PubMed Central

Google Scholar
Wang, Y. et al. Self-renewing monolayer of primary colonic or rectal epithelial cells. Cell Mol. Gastroenterol. Hepatol. 4, 165–182.e7 (2017).

Article
PubMed
PubMed Central

Google Scholar
Tong, Z. et al. Towards a defined ECM and small molecule based monolayer culture system for the expansion of mouse and human intestinal stem cells. Biomaterials 154, 60–73 (2018).

Article
CAS
PubMed

Google Scholar
He, G. W. et al. Optimized human intestinal organoid model reveals interleukin-22-dependency of Paneth cell formation. Cell Stem Cell 29, 1718–1720 (2022).

Article
CAS
PubMed
PubMed Central

Google Scholar
Lee, R. T., Berditchevski, F., Cheng, G. C. & Hemler, M. E. Integrin-mediated collagen matrix reorganization by cultured human vascular smooth muscle cells. Circ. Res. 76, 209–214 (1995).

Article
CAS
PubMed

Google Scholar
Ibrahim, D. G., Ko, J. A., Iwata, W., Okumichi, H. & Kiuchi, Y. An in vitro study of scarring formation mediated by human Tenon fibroblasts: effect of Y-27632, a rho kinase inhibitor. Cell Biochem. Funct. 37, 113–124 (2019).

Article
CAS
PubMed
PubMed Central

Google Scholar
Chapman, S., McDermott, D. H., Shen, K., Jang, M. K. & McBride, A. A. The effect of rho kinase inhibition on long-term keratinocyte proliferation is rapid and conditional. Stem Cell Res. Ther. 5, 60 (2014).

Article
PubMed
PubMed Central

Google Scholar
Mole, M. A. et al. Integrin β1 coordinates survival and morphogenesis of the embryonic lineage upon implantation and pluripotency transition. Cell Rep. 34, 108834 (2021).

Article
CAS
PubMed
PubMed Central

Google Scholar
Broguiere, N. et al. Growth of epithelial organoids in a defined hydrogel. Adv. Mater. 30, e1801621 (2018).

Article
PubMed

Google Scholar
Wijnakker, J. et al. Integrin-activating Yersinia protein invasin sustains long-term expansion of primary epithelial cells as 2D organoid sheets. Proc. Natl Acad. Sci. USA 122, e2420595121 (2025).

Article
CAS
PubMed

Google Scholar
Yui, S. et al. Functional engraftment of colon epithelium expanded in vitro from a single adult Lgr5⁺ stem cell. Nat. Med. 18, 618–623 (2012).

Article
CAS
PubMed

Google Scholar
Watanabe, S. et al. Transplantation of intestinal organoids into a mouse model of colitis. Nat. Protoc. 17, 649–671 (2022).

Article
CAS
PubMed

Google Scholar
Sugimoto, S., Kobayashi, E., Kanai, T. & Sato, T. In vivo intestinal research using organoid transplantation. Keio J. Med. 71, 73–81 (2022).

Article
CAS
PubMed

Google Scholar
Dekkers, J. F. et al. High-resolution 3D imaging of fixed and cleared organoids. Nat. Protoc. 14, 1756–1771 (2019).

Article
CAS
PubMed

Google Scholar
Meyer, L. et al. A simplified workflow for monoclonal antibody sequencing. PLoS ONE 14, e0218717 (2019).

Article
CAS
PubMed
PubMed Central

Google Scholar
Fujii, M., Matano, M., Nanki, K. & Sato, T. Efficient genetic engineering of human intestinal organoids using electroporation. Nat. Protoc. 10, 1474–1485 (2015).

Article
CAS
PubMed

Google Scholar
Andersson-Rolf, A. et al. One-step generation of conditional and reversible gene knockouts. Nat. Methods 14, 287–289 (2017).

Article
CAS
PubMed
PubMed Central

Google Scholar
Boonekamp, K. E. et al. Identification of novel human Wnt target genes using adult endodermal tissue-derived organoids. Dev. Biol. 474, 37–47 (2021).

Article
CAS
PubMed

Google Scholar
Saxena, K. et al. Human intestinal enteroids: a new model to study human rotavirus infection, host restriction, and pathophysiology. J. Virol. 90, 43–56 (2016).

Article
CAS
PubMed

Google Scholar
Pleguezuelos-Manzano, C. et al. Establishment and culture of human intestinal organoids derived from adult stem cells. Curr. Protoc. Immunol. 130, e106 (2020).

Article
CAS
PubMed
PubMed Central

Google Scholar
Broutier, L. et al. Culture and establishment of self-renewing human and mouse adult liver and pancreas 3D organoids and their genetic manipulation. Nat. Protoc. 11, 1724–1743 (2016).

Article
CAS
PubMed

Google Scholar

Download references

Acknowledgements

We thank R. van Ineveld for help with immunofluorescence tissue staining, T. Grenier for processing primary colon tissue and assistance with qPCR, and L. Huang and S. Lim for flow cytometry. We thank A. Brousali, J. Salij and O. Kranenburg of the Utrecht Platform for Organoid Technology (U-PORT; UMC Utrecht) for patient inclusion and tissue acquisition. This study was supported by the NWO Gravitation Project Material Driven Regeneration (024.003.013 ZWK MDR) funded by the Ministry of Education, Culture and Science of the government of the Netherlands (W.d.L., J.W., D.K., H.C.) and by the European Union’s Horizon 2020 research and innovation program INTENS (668294)(H.C.)

Author information

Author notes

Fenna L. H. van Rijt

Present address: Synaffix/Lonza Company, Oss, The Netherlands
Hans Clevers

Present address: Roche Pharmaceutical Research and Early Development, Basel, Switzerland
These authors contributed equally: Wim B. M. de Lau, Joost J. A. P. M. Wijnakker.

Authors and Affiliations

Oncode Institute, Hubrecht Institute, Royal Netherlands Academy of Arts and Science (KNAW) and UMC, Utrecht, The Netherlands

Wim B. M. de Lau, Joost J. A. P. M. Wijnakker, Daniel Krueger, Daisong Wang, Matthijs S. Abendroth, Robin Schreurs, Johan H. van Es & Hans Clevers
Oncode Institute, Princess Maxima Center of Pediatric Oncology, Utrecht, The Netherlands

Gijs J. F. van Son, Claudia Y. Janda, Fenna L. H. van Rijt, Benjamin J. P. Chalopin & Emma Bokobza
Harvard Medical School, Boston Children’s Hospital, Boston, MA, USA

Timothy A. Springer
Division of Cell Biology, Netherlands Cancer Institute, Amsterdam, The Netherlands

Arnoud Sonnenberg

Authors

Wim B. M. de Lau
Joost J. A. P. M. Wijnakker
Gijs J. F. van Son
Daniel Krueger
Daisong Wang
Matthijs S. Abendroth
Robin Schreurs
Claudia Y. Janda
Fenna L. H. van Rijt
Benjamin J. P. Chalopin
Emma Bokobza
Johan H. van Es
Timothy A. Springer
Arnoud Sonnenberg
Hans Clevers

Contributions

W.d.L. and H.C. conceived the initial approach. W.d.L. and J.W. performed most of the culture experiments. W.d.L. designed the single-chain antibody version. J.W. did most of the protein purifications with help from F.R. and supervision by C.J. J.W. performed part of the confocal staining. D.K. did many confocal staining experiments. E.B. performed a single confocal experiment. B.C. performed production and purification of the single-chain antibodies in mammalian cells. J.W. performed gene deletion. G.S. did bulk RNA sequencing, single-cell RNA sequencing analysis and statistics. M.A. performed part of the cell culture during the first revision, qPCR analysis and media preparation. R.S. designed the PureCol EZ Gel media and did functional testing during the second revision. D.W. gave advice on the differentiation protocols. J.E. provided scientific feedback throughout the project. A.S. and T.A.S. provided reagents and advice and reviewed the paper. J.W., W.d.L. and H.C. wrote the paper. All authors approved the paper.

Corresponding author

Correspondence to
Hans Clevers.

Ethics declarations

Competing interests

W.d.L. and H.C. are inventors on a patent application (P095058WO) directly related to this study. H.C. is inventor of several patents related to organoid technology; his full disclosure is given at https://www.uu.nl/staff/JCClevers. H.C. is the Head of Pharma Research and Early Development of Roche, Basel, Switzerland. The other authors declare no competing interests.

Peer review

Peer review information

Nature Biotechnology thanks Gil Garnier, Victor Hernandez and Kim Jensen for their contribution to the peer review of this work.

Additional information

Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Extended data

Extended Data Fig. 1 Summary of integrin component expression in single cells throughout the gastrointestinal tract.

We consulted a series of tissue- and organoid-derived single cell RNA-seq databases to obtain a concise overview, specifically, of integrin β1 and selected integrin α-component expressions. Strong co-expression is inferred for collagen binder ITGA2, laminin binder ITGA3 and ITGA6, and RGD binder ITGAV. Script that we used to compose this table is to be found at; https://github.com/Hubrecht-Clevers/Integrin_B1_TS216.

Extended Data Fig. 2 Antagonizing integrin β1 Ab AIIB2 induces immediate cell cycle arrest as visualized in FUCCI-reporter organoid line.

A, N39 Ileum organoids, stably expressing the live cell cycle progression-reporter FUCCI, were cultured on BME-coated transwell inserts. Red signal identifies G1 cells, the green signal the S-G2- and M phases. Yellow signal arises as a mixed red/green signal, occurring at the transition from G1 to S-phase. B, The AIIB2 ab (30 nM) was added to medium in the bottom compartment, binding β1 integrins at the basal membrane of epithelium. Incubation times are indicated. Cells were then enzymatically processed to single cells and expression of the cell phase-specific fluorescent signals – in DAPI-neg cells – analyzed by flow-cytometry. The green signal, representing cycling cells, starts declining from 6 hrs on.

Extended Data Fig. 3 Integrin β1- antagonizing ab AIIB2 induces differentiation of duodenum organoids.

Duo-72 Duodenal organoids were cultured in 3D hydrogel in regular expansion medium or supplemented with 10 nM AIIB2 ab for the indicated times. RNA, isolated for each condition, was converted to cDNA and subjected to qPCR using primer sets for enterocyte markers ANPEP, CYP2C9, and CYP2C19, or goblet cell marker TFF3. ΔCt (delta Ct) values were calculated for indicated target genes and reference gene GAPDH and beta-actin within the same sample. The fold change values quantify the increase in expression of individual markers in AIIB2-treated cells versus cells in its absence. Their average and STDEV are indicated and based on 4 technical replicates (PCR reactions performed on mRNA of the same organoid sample). P values for one-tailed T.test no Ab vs 96 hrs AIIB2 are 2.7 ×10^-4 for ANPEP, 5.39 ×10^-7 for CYP2c9, 2.8 ×10^-5 for CYP2C19, and 1.92 ×10^-4 for TFF3.

Source data

Extended Data Fig. 4 DNA – and protein sequences of mTS2/16 antibody-derived hu IgG1 version.

DNA sequences were cloned into the vector pFUSEss-CHIg-hG1 and pFUSE2ss-CLIg-hK. Yellow and green marked amino acids sequences represent the V_Hregion and the V_L region of TS2/16, respectively.

Extended Data Fig. 5 Amino acid compositions of H-L and L-H orientations of scTS2/16.

V_H domain is marked yellow, the V_L domain is marked in green. The V-domains are interconnected by (G4S)3 linker marked in grey. Note C-terminal 6xHIS tag (underlined) is included in the pET21 bacterial expression vector.

Extended Data Fig. 6 DNA coding -, and protein sequences of Fc-tagged scTS2/16 (V_H-V_L orientation) and its purification from Expi293 cells.

A; The coding sequence contains the artificial human kappa light chain signal sequence and a C-terminal 8xHis tag, followed by the HRV 3 C protease cleavage site (LEVLFQ/GP), a Fc-tag (constant region of human IgG1) and a stop codon. B, the scTS2/16-Fc was captured from conditioned medium after which the Fc was enzymatically cleaved. Upon capturing this Fc-moiety, the remaining scTS2/16 component underwent final SEC-purification. C, the end result (N = 3) was evaluated on SDS-PAGE under reducing and non-reducing conditions. MW of relevant marker proteins indicated in kDa.

Extended Data Fig. 7 β1 ectodomain protein sequence and recognition by scTS2/16.

A; Protein sequence integrin β1 ectodomain with designation numbering: Yellow characters for Plexin-Semaphorin-Integrin (PSI) domain. Grey characters in regions 60-118 and 331-413 indicate hybrid domain1 and hybrid domain 2, respectively. Green characters designate the A domain. The four colors, seen from amino acid 414-590, represent the different EGFR domains. The FLAG-tag is underlined, and the stop codon is indicated with an asterisk. B; Recognition of native integrin β1-ectodomain subunit by huIgG1-, and scFv versions of mAb TS2/16 (N = 3). Upper panel; protein-G agarose beads were incubated with huIgG1 version of antibody TS2/16 and several control integrin β1-specific antibodies, as indicated. An integrin β4 subunit -specific antibody was included as a negative control. After washing, beads were incubated with FLAG-tagged recombinant integrin β1 ectodomain (110 kDa). Beads were again washed and finally incubated with reducing SDS sample buffer. SDS-PAGE gel was blotted and capture of the β1-ectodomain tested using HRP-linked anti-Flag M2 mAb (Sigma). Results were visualized using Enhanced chemiluminescence (ECL). Lower panels; Beads loaded with M2 anti-Flag mAb were incubated with Flag-tagged β1 ectodomain (in PBS + 2% BSA) (lane 2 + 3) or just PBS-2% BSA (lane1). After washing, beads were incubated with HIS-tagged scTS2/16 antibody (lane 1 + 3) or PBS–2% BSA (lane2). Beads were washed and incubated with reducing SDS sample buffer. SDS_PAGE gel was blotted and stained directly with an HRP-linked anti HIS-ab (lower right panel), or indirectly using mouse M2 anti-Flag antibody followed by anti-mouse HRP-conjugate (loer right panel). Results were in each case visualized using Enhanced chemiluminescence (ECL). The left margins indicate protein sizes in kDa.

Extended Data Fig. 8 P26N Colon organoid adhesion towards different isoforms of laminin (N = 3).

Wells of 96-well plate were coated overnight by incubation of indicated recombinant laminins at a final concentration of 20 ng/ml in PBS. Plates were then blocked by incubating for 30 min. with PBS/2%BSA. Colon organoids were enzymatically processed to single cells and resuspended in DMEM-F12 medium. Cells were plated (n = 3) to a final concentration of 50.000 cells/200 μl/well in the presence of 5 nM of scTS2/16, as indicated. For each coating condition the inhibitory AIIB2 ab (30 nM) was also included, to test the relative involvement of integrin ß1. Upon incubation for 1,5 hrs (37⁰C), wells were extensively washed with DMEMF12 medium and remaining cells quantified by CellTiter-GLO, an ATP-driven luminescence assay (Relative Light Units – RLU.10⁶). Data presented as mean + /- STDEV (n = 3).

Extended Data Fig. 9 3D BME-grown intestinal organoids maintain their transcriptional program with or without TS2/16.

Organoids derived from the duodenum, jejunum, ileum, and colon were cultured in expansion medium with or without scTS2/16. PolyA RNA was converted to cDNA and analysed using bulk sequencing. A; volcano plot showing log2fold changes and p-value (using a Wald test for negative binomial distribution) in expression between scFv and control samples. Using Bonferroni correction to adjust p-values for multiple testing showed no significant difference (padj < 0.01) between scFv and control samples. B; a heatmap showing the 30 most up and down regulated genes based on log2fold change expression of the same samples. Rows were clustered using hierarchical clustering based on Euclidian distance. C; A sample distance matrix showing Euclidian distance, comparing tissues under control- and scTS2/16 conditions, visualizes their close genetic identity D; Principal Component Analysis (PCA) plot indicating the virtually identical transcriptomic make-up of control and scTS2/16 pairs within the four tissues relative to inter-tissue variance. A sample distance matrix showing Euclidian distance, comparing tissues under control- and scTS2/16 conditions, visualizes their close genetic identity. Columns and rows are clustered using hierarchical clustering. E; heat map showing DESEQ2 normalised counts of cell type-specific markers for colon grown in control- and scFv-supplemented media. Rows were clustered using hierarchical clustering based on Euclidian distance. F; similar analysis for small-intestine as shown in E.

Extended Data Fig. 10 Duodenal or colonic organoids maintained in 2D BME or PureCol EZgel, along with pancreatic ductal cells cultured in BME, display no changes in their transcriptional profiles regardless of scTS2/16 treatment.

Organoids from duodenum and colon were cultured on BME or PureCol EZgel, in expansion medium, +/- scTS2/16. In addition, we grew pancreas ductal organoids in BME in the presence – or absence of scTS2/16. PolyA RNA, purified from these lines, was converted to cDNA and subjected to bulk-sequencing. A; volcano plot showing log2fold changes in expression and pvalue differences (using a Wald test for negative binomial distribution) between the averages of scFv and control samples. Using Bonferroni correction to adjust p-values for multiple testing showed no significant difference (padj < 0.01) between scFv and control samples. B; a heatmap showing the 30 most up- and down regulated genes based on log2fold change expression of the same samples. Rows were clustered using hierarchical clustering based on Euclidian distance. C; A sample distance matrix showing Euclidian distance, comparing tissues under control- and scTS2/16 conditions, visualizes their close genetic identity. D; Principal Component Analysis (PCA) plot indicating the virtually identical transcriptomes of control and scTS2/16 pairs within the three tissues relative to inter-tissue variance. Columns and rows are clustered using hierarchical clustering. E; heat maps showing DESEQ2 normalised counts of cell type-specific markers for colon grown in control-, and scTS2/16 supplemented media. Rows were clustered using hierarchical clustering based on Euclidian distance. F; similar analysis for pancreas as shown in E. G; similar analysis for small intestine as shown in E.

Supplementary information

Source data

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/.

Reprints and permissions

About this article

Cite this article

de Lau, W.B.M., Wijnakker, J.J.A.P.M., van Son, G.J.F. et al. A single-chain derivative of an integrin-activating antibody potentiates organoid growth in Matrigel and collagen hydrogels.
Nat Biotechnol (2025). https://doi.org/10.1038/s41587-025-02874-8

Download citation

Received: 24 May 2023
Accepted: 22 September 2025
Published: 28 October 2025
DOI: https://doi.org/10.1038/s41587-025-02874-8

A single-chain derivative of an integrin-activating antibody potentiates organoid growth in Matrigel and collagen hydrogels

Solution biophysics reveals the polydisperse structure of RNA lipid nanoparticles

Selective RNA sequestration in biomolecular condensates directs cell fate transitions

Selective RNA sequestration in biomolecular condensates directs cell fate transitions

A single-chain derivative of an integrin-activating antibody potentiates organoid growth in Matrigel and collagen hydrogels

Main

Results

Integrin β1 is indispensable for organoid growth

hIgG and scFv versions of mTS2/16 activate cell adhesiveness through integrin β1

scFv-TS2/16 enhances growth of intestinal organoids in BME

BME-established organoids from different regions of the gastrointestinal tract also rely on integrin β1

scTS2/16 antibody potentiates growth of established and primary isolated colon organoids in collagen I hydrogel

Discussion

Methods

Immunofluorescence analysis of apical–basal polarity and proliferation marker Ki67

TS2/16 hIgG1 production

TS2/16 scFv production in E. coli

scFv TS2/16 production in Expi293 cells

Integrin β1 ectodomain production and immunoprecipitation

Adhesion assays of K562 cells

FACS staining integrins

Mn2+ growth assay

N39 ileum FUCCI reporter organoid line

qPCR for intestinal differentiation markers

ITGA2 CRISPR–Cas9 base editing

Organoid electroporation

Western blot

Bulk RNA sequencing

BME and PureCol EZ Gel-based 3D organoid culturing

BME- and PureCol EZ Gel-based 2D organoid culturing

Organoid differentiation media

Organoid differentiation in 2D

PureCol EZ Gel polymerization conditions

Long-term 3D PureCol EZ Gel (Fujii medium)

3D collagen differentiation protocol

Organoid lines

Statistics

Organoid expansion medium

Reporting summary

Data availability

References

Acknowledgements

Author information

Authors and Affiliations

Contributions

Corresponding author

Ethics declarations

Competing interests

Peer review

Peer review information

Additional information

Extended data

Supplementary information

Source data

Rights and permissions

About this article

Cite this article

Solution biophysics reveals the polydisperse structure of RNA lipid nanoparticles

Selective RNA sequestration in biomolecular condensates directs cell fate transitions

Mn²⁺ growth assay