MIF/CD74 axis is a target for novel therapies in colon carcinomatosis

Abstract

Background: Strategies aimed at obtaining a complete cytoreduction are needed to improve long-term survival for patients with colorectal cancer peritoneal carcinomatosis (CRC-pc).

Methods: We established organoid models from peritoneal metastases of two naïve CRC patients. A standard paraffin inclusion was conducted to compare their 3D structure and immunohistochemical profile with that of the corresponding surgical samples. RNA expression levels of the CRC stem cell marker LGR5 was measured by in situ hybridization. The secretome of organoids was profiled by mass spectrometry. Energy homeostasis of organoids was interfered with 4-IPP and metformin. Biochemical and metabolic changes after drug treatments were investigated by western blot and mass spectrometry. Mitochondria impairment was evaluated by electron microscopy and mitotraker staining.

Results: The two organoids recapitulated their corresponding clinical samples in terms of 3D structure and immmunoistochemical profile and were positive for the cancer stem cells marker LGR5. Proteomic analyses of organoids highlighted their strong dependence on energy producing pathways, which suggest that their targeting could be an effective therapeutic approach. To test this hypothesis, we treated organoids with two drugs that target metabolism acting on AMP-activated protein kinase (AMPK), the main regulator of cellular energy homeostasis, which may act as metabolic tumour suppressor in CRC. Organoids were treated with 4-IPP, an inhibitor of MIF/CD74 signalling axis which activates AMPK function, or metformin that inhibits mitochondrial respiratory chain complex I. As a new finding we observed that treatment with 4-IPP downregulated AMPK signalling activity, reduced AKT phosphorylation and activated a JNK-mediated stress-signalling response, thus generating mitochondrial impairment and cell death. Metformin treatment enhanced AMPK activation, decreasing the activity of the anabolic factors ribosomal protein S6 and p4EBP-1 and inducing mitochondrial depolarization.

Conclusion: We provide evidence that the modulation of AMPK activity may be a strategy for targeting metabolism of CRC-pc organoids.

Keywords: 4-IPP; AMPK; Macrophage migration inhibitory factor; Metabolism; Metformin; Organoids.

Figures

Fig. 1

Immunohistochemical profile of FFPE sections from C1 and C2 organoids and corresponding surgical samples. Haematoxylin and eosin (H&E) staining and CK AE1/AE3, CK19, CK20, CDX2 and Ki67 immunostaining of C1 a and C2 b. Arrows indicate the weak CK20 staining of C2 organoids and of the corresponding surgical sample. Magnifications at 20X

Fig. 2

In situ hybridization. in situ hybridization of LGR5 on FFPE sections of C1 and C2 organoids, and of the surgical specimens from which they originated

Fig. 3

Over-represented Gene Ontologies and pathways in the secretome of C2 organoids. a Gene Ontology analysis of secreted proteins. The Search Tool for the Retrieval of INteracting Genes/proteins (STRING) database was used for prediction of protein subcellular localization. STRING confidence score represented the probability of finding functional modules that were selected if the significance (p-value) of GO enrichment was less than 0.01. Statistical significance was ascertained with a false discovery rate (FDR) threshold of <0.05. b PANTHER overrepresentation test. Proteins involved in metabolic process and regulation of extracellular activities were also major components of the C2 secretome. The table shows the most representative identified overrepresented categories of proteins involved in Parkinson disease, glycolysis and pentose phosphate pathways. Experiments were performed in duplicate. Data were considerate significant when p ≤ 0.05. c Network analysis was performed using Ingenuity Pathway Analysis software. The analysis shows the connection between the MIF, DDT (MIF-2) with AMPK embedded within the network. The network predicts the potential value and validity of the MIF-AMPK connection. d Summary of the modulation of AMPK activity on organoids through MIF

Fig. 4

Effects of 4-IPP and metformin on C2 organoids. a MIF/CD74 signaling in C2 organoids. Lysates from C2 organoids expressed MIF, CD74 and presented activated AKT and AMPK. b Hematoxylin eosin and immunohistochemical staining of C2 tumour sample with anti-MIF and anti-CD74 antibodies. Magnifications at 10X. c Micrographs of C2 organoids before and after drug treatments. 4-IPP and metformin treatments induced a marked morphological change in C2 organoids, resulting in a loss of their original spheroid organization. Magnifications: 10X. Scale bars = 50 μm. d C2 organoids were treated with 5 mM metformin for 120 h or 50 and 100 μM 4-IPP for 24 and 48 h. The percentage of cell death was determined by trypan blue exclusion assay. Data are expressed as the mean ± SD. *P < 0.01 compared with untreated control and 50 μM 4-IPP. **P < 0.01 compared with untreated control and 100 μM 4-IPP. We have not observed significant increase in the percentage of cell death after metformin treatments compared with untreated controls. All the experiments were replicated at least three times. e Immunoblots of the principal proteins involved in CD74/MIF pathway. C2 organoids were untreated or treated with 50 and 100 μM 4-IPP for 24 h; lysates were resolved by 4–12% SDS–PAGE and immunoblotted. f Immunoblots of the principal proteins involved in metformin pathway. C2 organoids were untreated or treated with 5 mM metformin for 120 h; lysates were resolved by 4–12% SDS–PAGE and immunoblotted. g Global tyrosine and threonine phosphorylation changes in C2 organoids after 4-IPP and metformin treatments. C2 organoids were untreated or treated with 50 and 100 μM 4-IPP for 24 h and with 5 mM metformin for 120 h; lysates were resolved by 4–12% SDS–PAGE and immunoblotted. h Immunoblots showing the expression and activity of the PP2A and PP1 proteins before and after 4-IPP or metformin treatments. Blots are representative of the results from multiple (at least two) experiments. i Observed consequences of the 4-IPP or metformin treatments on C2 organoids

Fig. 5

Metabolic changes after 4-IPP or metformin treatments. The figure shows the bargraphs relative of the metabolites analysed by mass spectrometry after C2 treatments with 4-IPP at 100 μM for 24 h and metformin at 5 mM for 120 h. Values are expressed as ng/mg proteins and are normalized to the value of total proteins present in each sample. Each value represents the mean of five independent replicates. Metabolites values are also plotted; blue bars mark the amount with respect to the highest values. Metabolic organic acids intermediates of the TCA cycle (citrate, succinate and fumarate) were reduced both in 4-IPP and metformin treated cells. In the case of metformin treated cells, there was a significant increase in lactate production, suggestive of an inefficient pyruvate-dehidrogenase activity due to impaired TCA and OXPHOS. Citrate released by mitochondria is converted into Acetyl-CoA, NADPH and ATP for lipid (cholesterol and fatty acids) synthesis. Cholesterol and precursor sterols (markers of cholesterol synthesis) were reduced in both 4-IPP and metformin treated cells. Very long chain fatty acids (>C16) were reduced by both treatments

Fig. 6

Mitochondria impairment after 4-IPP and metformin treatments in C2 organoids. a Reactive oxygen species detection with ROS-H2O2-Glo assay kit. C2 cells were treated with metformin 5 mM, 4-IPP 100 μM or left untreated for 120 and 24 h respectively. Extracellular H2O2 formation was detected and quantified using the ROS-H2O2-Glo assay. Luminescence intensity was quantified using a microplate reader with a 500 ms integration time, reported as relative light units and normalised to 0 cells/well treatment conditions. Standard deviation was calculated for a set of triplicate values. b Confocal microscopy images of mitochondria using Mitotracker® Deep Red FM in C2 cells untreated (A2; D1; E1), treated with 50 μM 4-IPP for 24 h (B2), 100 μM 4-IPP for 24 h (C2), or 5 mM metformin for 120 h (D2; E2). Arrows indicate aggregated mitochondria. 4-IPP treatment leads to mitochondria impairment in C2 cells. Metformin treatment reduces signal intensity of Mitotracker (asterisk) suggesting a depolarization of mitochondrial membrane potential. The figure shows data from a representative experiment. DIC: differential interference contrast. Magnifications: 10X. Scale bars = 50 μm. All the experiments were replicated at least three times. C Flow cytometry analysis of the mitochondrial membrane potential using JC-1 assay on C2 organoids treated with metformin for 120 h or left untreated. Staining JC-1 revealed that metformin treatment caused a significant increase in the amount of FITC aggregates, a pattern that is characteristic for disruption of mitochondrial membrane potential. d The ultrastructure of the C2 cells after 4-IPP (D1-3) or metformin treatment (D4-6) and left untreated (D7). The white arrows show autophagic structures and red stars indicate apoptotic bodies present in 4-IPP treated cells. White triangles show irregular mitochondria, blue stars lipid droplets and the yellow star a multilamellar structure present in metformin treated cells. Additionally, no signs of autophagy, apoptosis or irregular mitochondria were observed in untreated cells (D7). e Functional consequences of 4-IPP or metformin treatments on C2 organoids

Fig. 7

4-IPP and metformin effects are recapitulated using C1 organoids. a Micrographs of C1 organoids before and after 4-IPP treatments. 4-IPP treatment induced loss of the original spheroid organization. Magnifications: 10X. Scale bars = 50 μm. b Immunoblots of the principal proteins involved in CD74/MIF signaling pathway. C1 organoids were untreated or treated with 50 and 100 μM 4-IPP for 24 h; lysates were resolved by 4–12% SDS–PAGE and immunoblotted. c Immunoblots showing the expression and activity of the α-pTyr and α-pThr proteins before and after 4-IPP treatment. C1 organoids were untreated or treated with 50 and 100 μM 4-IPP for 24 h; lysates were resolved by 4–12% SDS–PAGE and immunoblotted. d C1 organoids were treated with 50 and 100 μM 4-IPP for 24 and 48 h. The percentage of cell death was determined by trypan blue exclusion assay. Data are expressed as the mean ± SD. *P < 0.01 compared with untreated control and 50 μM 4-IPP. **P < 0.01 compared with untreated control and 100 μM 4-IPP. All the experiments were replicated at least three times. e Micrographs of C1 organoids before and after metformin treatments. Metformin treatment induced a moderate loss of the original spheroids organization. Magnifications: 10X. Scale bars = 50 μm. f Immunoblots of the principal proteins involved in the metformin pathway. C1 organoids were treated with 5 mM metformin for 120 h; lysates were resolved by 4–12% SDS–PAGE and immunoblotted. Images are representative of the results from at least two experiments. g C1 organoids were treated with 5 mM metformin for 120 h. The percentage of cell death was determined by trypan blue exclusion assay. Data are expressed as the mean ± SD. We have not observed significant increase in the percentage of cell death after metformin treatments compared with untreated controls. All the experiments were replicated at least three times

Fig. 8

CD74 and MIF IHC staining in three additional CRC-pc samples. In the upper part of the Figure, the normal colon mucosa from the three patients is shown. In all the cases, CD74 is highly expressed in epithelial cells and in the surrounding normal cellular microenvironment. Conversely, MIF is poorly expressed. MIF staining is mainly restricted to epithelial cells. In the lower part of the figure, the corresponding peritoneal carcinomatoses (C3, C4 and C5) are shown. Immunohistochemical staining for CD74 is heterogeneous with a granular, cytoplasmic and membranous pattern in C3, while focal, granular and restricted to the inner luminal aspect in C4 (inset) and C5. MIF immunostaining is homogeneous and very low expressed in C3 compared to C4 and C5. MIF staining is mainly cytoplasmic. Images were taken using a Nikon Eclipse 80i microscope (Nikon, Tokyo, Japan) and Nikon digital sight DS-Fi1 camera equipped with control unit-DS-L2 (Nikon). Images were assembled using Adobe Photoshop 6 (Adobe Systems, San Jose, CA, USA). Original magnification at 20X