Female C57BL/6 mice and ERdj5 deficient mice were purchased as previously described [17]. All the mice were bred at the Kangwon National University Animal Laboratory Center.

Murine colonic organoids were cultured as previously described [26]. Briefly, mouse colonic tissues were washed to eliminate the gut microbiota, and dissociated colonic crypts were used as a recovery solution. Finally, intestinal crypts were plated in 50% Matrigel and filled with colon organoid growth medium (Dulbecco's Modified Eagle Medium F/12; Gibco, Carlsbad, CA, USA), 50% L-WRN growth medium, B27 (Gibco), N2 (Gibco), 1.25 mM N-acetylcysteine (Sigma Aldrich, St. Louis, MO, USA), 10 mM nicotinamide (Sigma Aldrich), 10 μM P38i (Sigma Aldrich), 16.7 pM mEGF (Invitrogen, Vienna, Austria), 10 μM Y-27632 (Abmole, Houston, TX, USA), and 0.5 μM A83-01 (Tocris, Bristol, UK). The colon organoid growth medium was replaced every 2 days, and the colonoids were passaged every 7 days. Colon organoids were observed using Nikon Eclipse Ts2 (Nikon, Minato, Tokyo, Japan).

Colonoids were incubated for 48 hours at 37°C on day 7 with 1 µg/mL of Pam₃CSK₄ (InvivoGen, San Diego, CA, USA), an agonist of Toll-like receptor (TLR)2/TLR1, to mimic inflammatory conditions.

Radiation exposure experiments inducing cell stress in colonoids were performed as previously described. In brief, we treated the colonoids after 3 days of culture with a 6 Gy dose of γ-radiation using Gammacell 40 Exactor (Best Theratronics, Ottawa, ON, Canada). After radiation treatment, the colonoids were incubated at 37℃ for 4 days.

Total RNA was extracted from the colon organoids using the TRIzol reagent (Ambion, Austin, TX, USA). RNA quality was confirmed using an ultraviolet-visible spectrophotometer (MicroDigital, Seongnam, Korea). The RNA was reverse-transcribed using the Goscript Reverse Transcription System (Promega, Madison, WI, USA). Additionally, quantitative real-time polymerase chain reaction (qRT-PCR) was performed using GoTaq^® qPCR and RT-qPCR Systems (Promega) by CFX96 Touch Real-Time PCR Detection System (Bio-Rad, Hercules, CA, USA). The primers used for qPCR were in Table 1.

Colon organoids were fixed in 4% paraformaldehyde and permeabilized with 2% Triton-X100 (Sigma Aldrich). Nonspecific antibody binding to the organoid surface was blocked with 3% bovine serum albumin (MP Biomedicals, Solon, OH, USA) in phosphate-buffered saline. To assess colon cell death, the ApopTag Fluorescein In Situ Apoptosis Detection Kit (Merck, Kenilworth, NJ, USA) was used for the terminal deoxynucleotidyl transferase dUTP nick end labeling (TUNEL) assay. The colonoids were then mounted with a Fluoroshield with 4′,6-diamidino-2-phenylindole (Sigma) to visualize the nuclei. Organoid images were visualized using a confocal laser scanning microscope (LSM880; Carl Zeiss, Göttingen, Germany) and analyzed using the ZEN program (Carl Zeiss).

GraphPad Prism 10 (San Diego, CA, USA) was used for data analysis. Differences between the 2 groups were analyzed using the t-test. One-way analysis of variance and post-hoc tests were also performed. The Tukey multiple-comparison test was utilized for comparisons of more than 2 groups. A threshold of p<0.05 was deemed statistically significant.

We cultured an equal number of crypts (5×10²/well) from the ERdj5 wild-type (WT)/ knockout (KO) mice colon and changed the medium every 2 days while observing the growth of the organoids for 15 days (Fig. 1A). We observed the morphology of the organoids at the indicated time points and obtained RNA to observe changes in gene expression (Fig. 1B). As observed in the colonoid images, no apparent differences in the early stages of growth were noted; however, from day 7, the ERdj5 KO organoids were smaller than the WT organoids. Moreover, Lgr5 levels on the indicated dates rapidly increased until day 7 in both the ERdj5 WT and KO groups, followed by a significant decrease thereafter (Fig. 1C). However, no significant difference in Lgr5 and Ki67 expression was observed between WT and ERdj5 KO organoids (Fig. 1C). We investigated differentiation because we did not observe significant differences in growth factors between ERdj5 WT and KO colon organoids over 15 days.

Previous studies have confirmed the appearance of differentiation-related characteristics that confer colonic identity on days 7 and 10 [27]. To evaluate the organoids, we analyzed budding on days 7 and 10. We observed that ERdj5 KO organoids formed fewer complex crypt-like structures than WT organoids (Fig. 2A). Furthermore, when evaluating the degree of differentiation on day 7 using microscopy, we observed that the level of differentiation per well was less pronounced in ERdj5 KO mice than in WT mice (Fig. 2B). Next, we confirmed the presence of RNA in functional cells that constitute the colon. The mucin 2 (Muc2) marker in goblet cells and lysozyme 1 (Lyz1) marker in Paneth cells demonstrated no significant differences between the groups. However, the enterocyte marker Villin 1 (Vil1) and the TA cell marker Prominin 1 (Prom1) exhibited lower levels in ERDj5 KO colonoids on day 7 than in WT colonoids (Fig. 2C). Collectively, the absence of ERdj5 resulted in reduced organoid differentiation.

ER stress modulates the levels of Lgr5, a prominent stem cell marker and a key factor in organoid growth [28]. As reported in our previous study, we induced inflammation using Pam₃CSK₄ (a TLR2 ligand) and observed external structural collapse, increased C-X-C motif chemokine ligand 1 secretion, and goblet cell depletion [17]. We investigated the impact of ERdj5 on functional cells such as Paneth cells and goblet cells in IECs; however, the influence of ERdj5 on the differentiation of stem cells into functional cell types remains unknown. Therefore, we aimed to investigate the effect of ERdj5 deficiency on growth factors in differentiated organoids by inducing TLR2-mediated inflammation. After 7 days of growth, colonoids were incubated for 48 hours at 37°C with 1 µg/mL of Pam₃CSK₄, which led to structural rupture in ERdj5-deficient organoids, unlike WT organoids (Fig. 3A). Similarly, as displayed in Fig. 3A, ERdj5 deficiency increased the number of TUNEL-positive cells (Fig. 3B). We examined various inflammation-related factors at the RNA level; however, no significant differences were identified except in interleukin-1 beta (Fig. S1). However, an increase in the PERK-related protein C/EBP homologous protein (CHOP) provided insight into the occurrence of cell apoptosis (Fig. 3C). Additionally, induction of ER stress was associated with a significant decrease in Lgr5 expression (Fig. 3D). Pam₃CSK₄ treatment induced ER stress, resulting in the exacerbation of apoptosis and stemness in organoids with ERdj5 deficiency.

Irradiation induces ER stress by directly or indirectly generating ROS [25]. We induced ER stress through another pathway to investigate how ERdj5 affects organoid growth. A previous study found that radiation with an intensity greater than 6 Gy resulted in the loss of budding and morphology maintenance in organoids [29]. Following this reference, we cultured organoids for 3 days and then irradiated them with 6 Gy to investigate the changes occurring. In ERdj5-deficient organoids, no expansion was observed, resulting in smaller organoids than the WT organoids (Fig. 4A). In WT organoids, Lgr5 exhibited a significant increase 4 days after irradiation, whereas, in ERdj5 KO organoids, no significant increase was observed (Fig. 4B). After irradiation, the markers Prom1, Vil1, and Ki67 decreased in ERdj5 KO organoids compared to WT organoids, indicating a decline before their differentiation into functional cells from stem cells. However, no significant differences were observed in the ER stress initiator binding immunoglobulin protein (BiP) after irradiation. Collectively, inflammation through irradiation induced mild ER stress in both groups, whereas ERdj5 deficiency eventually led to the downregulation of Lgr5. The colonoids were unable to defend against various inflammatory stimuli owing to a deficiency of the ERdj5-related ER stress resolution pathway.