Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2020 Nov 19;18(1):349.
doi: 10.1186/s12916-020-01806-4.

ADAM22/LGI1 complex as a new actionable _target for breast cancer brain metastasis

Sara Charmsaz  1 , Ben Doherty  1 , Sinéad Cocchiglia  1 , Damir Varešlija  1 , Attilio Marino  2 , Nicola Cosgrove  1 , Ricardo Marques  1 , Nolan Priedigkeit  3 , Siobhan Purcell  1 , Fiona Bane  1 , Jarlath Bolger  1 , Christopher Byrne  1 , Philip J O'Halloran  4 , Francesca Brett  5 , Katherine Sheehan  6 , Kieran Brennan  7 , Ann M Hopkins  7 , Stephen Keelan  1 , Petra Jagust  1 , Stephen Madden  8 , Chiara Martinelli  2 , Matteo Battaglini  2   9 , Steffi Oesterreich  3 , Adrian V Lee  3 , Gianni Ciofani  2 , Arnold D K Hill  7   10 , Leonie S Young  11
Affiliations

ADAM22/LGI1 complex as a new actionable _target for breast cancer brain metastasis

Sara Charmsaz et al. BMC Med. .

Abstract

Background: Metastatic breast cancer is a major cause of cancer-related deaths in woman. Brain metastasis is a common and devastating site of relapse for several breast cancer molecular subtypes, including oestrogen receptor-positive disease, with life expectancy of less than a year. While efforts have been devoted to developing therapeutics for extra-cranial metastasis, drug penetration of blood-brain barrier (BBB) remains a major clinical challenge. Defining molecular alterations in breast cancer brain metastasis enables the identification of novel actionable _targets.

Methods: Global transcriptomic analysis of matched primary and metastatic patient tumours (n = 35 patients, 70 tumour samples) identified a putative new actionable _target for advanced breast cancer which was further validated in vivo and in breast cancer patient tumour tissue (n = 843 patients). A peptide mimetic of the _target's natural ligand was designed in silico and its efficacy assessed in in vitro, ex vivo and in vivo models of breast cancer metastasis.

Results: Bioinformatic analysis of over-represented pathways in metastatic breast cancer identified ADAM22 as a top ranked member of the ECM-related druggable genome specific to brain metastases. ADAM22 was validated as an actionable _target in in vitro, ex vivo and in patient tumour tissue (n = 843 patients). A peptide mimetic of the ADAM22 ligand LGI1, LGI1MIM, was designed in silico. The efficacy of LGI1MIM and its ability to penetrate the BBB were assessed in vitro, ex vivo and in brain metastasis BBB 3D biometric biohybrid models, respectively. Treatment with LGI1MIM in vivo inhibited disease progression, in particular the development of brain metastasis.

Conclusion: ADAM22 expression in advanced breast cancer supports development of breast cancer brain metastasis. _targeting ADAM22 with a peptide mimetic LGI1MIM represents a new therapeutic option to treat metastatic brain disease.

Trial registration: ClinicalTrials.gov NCT01840293.

Keywords: ADAM22; Blood–brain barrier; Brain metastases; Breast cancer metastases; ECM signalling; LGI1; _targeted therapy.

PubMed Disclaimer

Conflict of interest statement

The authors declare that they have no competing interests. LY has previously had collaborative funding with Pfizer from Science Foundation Ireland.

Figures

Fig. 1
Fig. 1
ECM is a key pathway in breast cancer metastasis and the ECM signalling protein ADAM22 promotes distant metastatic disease burden in vivo. a Sequencing reads from exome capture RNAseq of patient-matched primary breast tumour with bone metastases (n = 11 patients) and brain metastases (n = 21 patients) and from RNA seq of matched primary and metastatic liver (n = 3 patients) were mapped against human reference transcriptome GRCh38.p10. DESeq2 (exome capture RNAseq) and edgeR (RNAseq) were used to identify differential gene expression separately in primary breast tumours compared to matched brain (266), bone (51) and liver (695) metastases (log2 fold change ± 1.5 FDR < 0.05). Dotplot (right) of functional annotation of differentially expressed genes using gene ontology biological process (hypergeometric test q-value < 0.05). b Differentially expressed genes (log2 FC > 1.5, adjusted p value < 0.01) are displayed in the volcano plot. Genes were cross referenced against the Matrisome database to identify extracellular matrix (ECM) related genes (labelled in the volcano plot). c Exome-capture RNAseq of ER-positive primary breast and matched brain metastatic tissues (n = 9 patients, 18 samples). Differentially expressed genes (log2 FC > 1.5, adjusted p value < 0.01) are displayed in the volcano plot. Genes were cross referenced against the Matrisome database to identify extracellular matrix (ECM)-related genes (labelled in the volcano plot). d A significant increase in ADAM22 expression was found in brain metastases in patients in comparison to matched primary tissue, n = 9, Wilcoxon matched-pairs signed-rank test *p = 0.0117. e Oncoprint of recurrent (> 3 patients) Matrisome and Druggable Genome (Drug-Gene Interaction Database)-related gene expression gains in ER-positive brain metastases patients (n = 9) showed ADAM22 (78%) as the second-ranked druggable genome in the list
Fig. 2
Fig. 2
ADAM22 expression promotes metastatic potential of endocrine resistant cells via upregulation of pro-metastatic signalling pathways. a LY2 WT, LY2 ADAM22 KI and LY2 ADAM22 KO (1 × 106) cells were luciferase tagged and injected into the mammary fat pad of NOD/SCID mice (n = 8, n = 7 and n = 7, respectively) and allowed to grow to approximately 150 mm3. b Tumour growth was assessed by IVIS 15 min post luciferin injection (150 μg/g). Tumour volume was measured by weekly calliper measurements using the following formula: Volume mm3 = (length × (smallest width) × 2) × 0.5. Statistical significance was calculated using 2-way ANOVA multiple comparison test *p = 0.0395, ***p < 0.0005, ****p < 0.0001. c Heatmap from RPPA analysis showing proteins/phospho-proteins differentially expressed in LY2 and LY2 ADAM22-KO cells following 4-OHT (10− 7 M) treatment for 15 min (LIMMA adjusted p value < 0.05; red = upregulated; blue = downregulated). d Pathway analysis using ClueGo represent the common pathways from KEGG pathway analysis using the common _targets from 4 or more differentially expressed proteins by ADAM22. e Top ranking KEGG pathways associated with ADAM22 dependent 4-OHT response. f ADAM22 promotes mammosphere formation. LY2, MCF7, LY2 ADAM22 KO and LY2 ADAM22 KI cells were plated in mammosphere-forming medium supplemented with 4-OHT (10− 8 M) for 5 days. Mammospheres (> 50 μm) were counted to determine the mammosphere forming efficiency (MFE). Bar graphs show relative (to LY2) MFE ± SEM from three independent experiments. Statistical significance was calculated using one-way ANOVA, ***p = 0.0001 ****p < 0.0001. g ADAM22 expression promotes anchorage independent colony formation. LY2, LY2 ADAM22 KO, MCF7 and LY2 ADAM22 KI cells cultured in an anchorage independent state for 14 days. Colonies were stained with p-iodonitrotetrazolium and counted. Bar graphs show relative (to LY2) colony formation ± SEM from three independent experiments. Statistical significance was calculated using one-way ANOVA, ****p < 0.0001
Fig. 3
Fig. 3
Clinically, ADAM22 expression is associated with relapse and brain metastasis and may be a suitable candidate for _targeted therapy. a Kaplan–Meier estimates of disease-free survival according to ADAM22 in all (n = 843 patients) (log rank, Pr > χ2 = 0.0208). b Kaplan–Meier estimates of disease-free survival according to ADAM22 in ER-positive breast cancer patients (n = 670 patients) (log rank, Pr > χ2 = 0.0432). c Immunohistochemical ADAM22 staining in 6 ER-positive endocrine-treated primary breast tumours and matched metastatic tissue. ADAM22 expression was maintained or elevated on metastasis. Scale bar, 50 μm. d Immunohistochemical ADAM22 staining in normal healthy tissues. Scale bar, 20 μm for brain and 50 μm for all other organs
Fig. 4
Fig. 4
The LGI1 peptide mimetic interacts with ADAM22 and inhibits pro-metastatic potential in vitro. a A 22 amino acid peptide mimetic (LGI1MIM) was designed based on the predicted ADAM22 binding domain of LGI1 (amino acids 441–462). A single cysteine to serine substitution (bold and underlined) was introduced in LGI1MIM to improve solubility. b Predicted interaction of LGI1MIM (pink) and the disintegrin domain (yellow) of ADAM22 (grey) using CABS-dock at a threshold of < 3A. c Contact map showing distribution of LGI1MIM contact points within the ADAM22 protein. Contact points within the disintegrin domain are highlighted (red box). d Western blot confirmation of the LGI1MIM/ADAM22 interaction. A no bait control (NBC), biotinylated scrambled peptide (SCRM) and biotinylated LGI1MIM were pre-incubated with LY2 lysate before pulling associated proteins with streptavidin Dynabeads. Interacting proteins were immunoblotted with ADAM22 antibody. e LGI1MIM (10 nM) significantly inhibits migration of endocrine-resistant LY2 and LetR cells, similar to full length recombinant LGI1 (5 nM) compared to scrambled peptide (SCRM) or vehicle. Two-way ANOVA, *p < 0.05 ***p = 0.0002 ****p < 0.0001. f LGI1MIM significantly inhibits mammosphere formation. LY2 cells were plated in mammosphere-forming medium supplemented with 4-OHT (10− 8 M) for 5 days in the presence or absence of LGI1MIM (25 nM). Mammospheres (> 50 μm) were counted to determine the MFE. Bar graphs show relative (to untreated) MFE ± SEM from three independent experiments. Unpaired two-tailed t-test ***p = 0.0004. g LY2 cells were cultured in an anchorage independent state for 14 days in the presence or absence of LGI1MIM (25 nM). Colonies were stained with p-iodonitrotetrazolium and counted. Bar graphs show relative (to LY2) colony formation ± SEM from three independent experiments. Bar graphs show relative (to vehicle) colony formation ± SEM from three independent experiments. Unpaired two-tailed t test ***p = 0.0002. h Schematic representation of an ex vivo explant experiment testing the effect of LGI1MIM treatment on patient brain metastatic tumour. i Proliferation rate of the tumour cells evaluated by Ki67 immunohistochemical staining (scale bar 100 μM) and represented as relative viable proliferating cells (T347, T2447 and T328). Bar graphs show relative (to vehicle) viable cell proliferation ± SEM, N = 3. j Brain metastatic cells (T347) grown as organoids in the presence of LGI1MIM (25 nM) or vehicle for 72 h, scale bar 50 μM. LGI1MIM significantly reduced cell proliferation as measured at 7 days using a 3D cell viability assay (p < 0.001, n = 8)
Fig. 5
Fig. 5
The anti-proliferative effect of LGI1MIM on metastatic tumour and the anti-cancer effect and potential of LGI1MIM-loaded liposomes (LGI1MIM/LSs) for crossing the blood–brain barrier (BBB). a Schematic of LGI1MIM/LSs functionalised with the antibodies against transferrin receptor. b Z-potential and hydrodynamic size (diameter) of the LGI1MIM/LSs. c WST-1 cell viability assay reveals the anti-proliferative effects of both LGI1MIM and LGI1MIM/LSs (single administrations; 72 h of treatment) on primary cells derived from brain metastatic tumour (T347). Mean ± SD *p < 0.05; **p < 0.01; ***p < 0.001, ANOVA HSD post hoc test is used. d Schematic of the multicellular 2D model of the BBB (left); 3D confocal imaging of the endothelial layer (top right) and of the astrocytes (bottom right); nuclei in blue, zonula occludens-1 (ZO-1) in green and f-actin in red. e BBB crossing of DiO-stained LGI1MIM/LSs incubated in the luminal compartment at increasing concentrations (15, 50, 150 and 500 μg/ml). LGI1MIM/LSs were detected in the abluminal compartment. f 3D confocal laser scanning microscopy imaging of T347 cells after treatment with 500 μg/ml LGI1MIM-LSs (single administration in the luminal compartment; 72 h of incubation). g WST-1 cell viability assay on T347 cells in response to a single administration in the luminal compartment of vehicle, 5 μM LSs, 500 μg/ml LGI1MIM, or 500 μg/ml LGI1MIM-LSs (72 h of incubation) *p < 0.05; **p < 0.01; ***p < 0.001, ANOVA HSD post hoc test is used. h Ki-67 expression on T347 cells in response to the treatments reported in (k). *p < 0.05; **p < 0.01; ***p < 0.001, ANOVA HSD post hoc test is used
Fig. 6
Fig. 6
The LGI1 mimetic inhibits metastatic burden in vivo. a LGI1MIM displays no toxicity in mice at low-, mid- and high-dose concentrations. NOD/SCID mice were administered a daily regimen of LGI1MIM by IP injection at three doses: 1 μg/mouse/day (n = 2), 10 μg/mouse/day (n = 2) and 100 μg/mouse/day (n = 2). Mouse weights were noted daily as a measure of LGI1MIM tolerability. b Schematic of LGI1MIM in vivo study to examine early metastatic seeding events. c Mean tumour volume in baseline study (top) determines baseline LGI1MIM tumour inhibition and the early seeding study (bottom) determines LGI1MIM effect on initial metastatic events (LGI1MIM = red; vehicle = blue). Either 1 × 106 (baseline study) or 8 × 105 (early seeding study) luciferase tagged LY2 cells were implanted into the mammary fat pad of NOD/SCID mice. Mice were treated with either vehicle (n = 2 mice baseline study; n = 7 mice early seeding study) or LGI1MIM at (100 μg/mouse/per day) (n = 2 control study; n = 7 metastatic study) for 6 weeks. LGI1MIM induced a substantial decrease in local tumour volume in the baseline study (mean 450 mm3 versus 150 mm3, calliper measurement), whereas no alteration was detected in the early metastatic seeding model. d Cumulative metastatic burden in vehicle- and LGI1MIM-treated mice. LGI1MIM significantly reduces metastatic burden (as measured by BLU log2 p/s). Two tailed Mann–Whitney test* p = 0.0111. e The ex vivo luciferase activity from brain metastases in vehicle-treated (top) and LGI1MIM-treated (bottom) mice. f Brain metastatic burden, BLU log2 p/s, was significantly reduced in LGI1MIM treated mice (red; n = 7 mice) versus vehicle control (blue; n = 7 mice). Two-tailed Mann–Whitney test* p = 0.0373
See this image and copyright information in PMC

Similar articles

See all similar articles

Cited by

References

    1. Mehlen P, Puisieux A. Metastasis: a question of life or death. Nat Rev Cancer. 2006;6(6):449–458. doi: 10.1038/nrc1886. - DOI - PubMed
    1. McBryan J, Fagan A, McCartan D, Bane FT, Vareslija D, Cocchiglia S, et al. Transcriptomic profiling of sequential tumors from breast cancer patients provides a global view of metastatic expression changes following endocrine therapy. Clin Cancer Res. 2015;21(23):5371–5379. doi: 10.1158/1078-0432.CCR-14-2155. - DOI - PubMed
    1. Vareslija D, Priedigkeit N, Fagan A, Purcell S, Cosgrove N, O'Halloran PJ, et al. Transcriptome characterization of matched primary breast and brain metastatic tumors to detect novel actionable _targets. J Natl Cancer Inst. 2019;111(4):388–398. doi: 10.1093/jnci/djy110. - DOI - PMC - PubMed
    1. Mead TJ, Apte SS. ADAMTS proteins in human disorders. Matrix Biol. 2018;71–72:225–239. doi: 10.1016/j.matbio.2018.06.002. - DOI - PMC - PubMed
    1. Turner SL, Blair-Zajdel ME, Bunning RA. ADAMs and ADAMTSs in cancer. Br J Biomed Sci. 2009;66(2):117–128. doi: 10.1080/09674845.2009.11730257. - DOI - PubMed

Publication types

MeSH terms

Associated data

  NODES
admin 4
INTERN 2
Note 1
twitter 2