Matrisome proteins at the heart of precision medicine and fertility restoration

Consider This
Matrisome proteins at the heart of precision medicine and fertility restoration
Like

Authors:

Emna Ouni, PhD1, Didier Vertommen, PhD2, Christiani A. Amorim, PhD3

1Pôle de Recherche en Gynécologie, Institut de Recherche Expérimentale et Clinique, Université Catholique de Louvain, Brussels, Belgium
2PHOS Unit & MASSPROT platform de Duve Institute, Université Catholique de Louvain, Brussels, Belgium
3Pôle de Recherche en Physiopathologie de la Reproduction, Institut de Recherche Expérimentale et Clinique, Université Catholique de Louvain, Brussels, Belgium

Consider This:

Chemotherapy has been known for its gonadotoxicity through several mechanisms by (i) having a direct detrimental effect on the DNA of primordial follicles constituting the ovarian reserve leading to apoptosis; (ii) induction of massive growth of dormant follicles, which are then destroyed; or (iii) vascular ovarian damage. While a recent study of our group highlights the involvement of ovarian extracellular-matrix (ECM) proteins in the described cellular mechanisms (Fig. 2) (3), chemotherapy gonadotoxicity has never been related to the follicle microenvironment. Therefore, as a concrete demonstration of the ECM-chemotherapy relationship, we show the example of possible undiscovered interactions between chemotherapy and ovarian matrisome proteins using advanced text-mining technology, shedding light on novel facets of the gonadotoxicity of chemotherapy. Our goal is to help understand how chemotherapy can affect differently ovarian cell microenvironment in an age-dependent manner, leading to different alterations depending on the treatment type and the patient age. We hope our analytical workflow can inspire other researchers and oncologists to conduct similar analyses to better consider ECM implications in cancer therapy.

Reverting back to a native non-permissive tumor environment using reverse bioengineering

Tumors need constant support from previously ‘unsupportive’ microenvironments. Novel therapeutic strategies that inhibit such microenvironmental support to tumor cells would reduce chemoresistance and relapse. Thus, an important area of future cancer research will be determining whether abnormal ECM could be an effective cancer therapeutic target. A daunting task in this regard will be determining the kind of ECM changes that have causative effects on disease progression and how these changes, alone or in combination, may affect cancer cells and their surrounding stromal cells.

Reverse engineering, which means applying the engineering concept of taking apart a process or mechanism to understand and re-engineer it, offers an excellent opportunity to understand the native non-permissive cancer environment and guide future targeted therapies and more biomimetic models for cancer research and regenerative medicine. The best example of such a strategy is available in the context of human ovarian tissue, where ECM was deconstructed to better understand its proteomic, architectural and mechanical cues and their role in the ovarian activity. Our team has generated the first proteomic atlas of the human ovarian tissue to fingerprint 98 proteins characterizing healthy ovarian matrisome (an extended definition of the ECM) (1-3). Among these proteins, 26 potential fertility biomarkers were differently expressed between prepubertal, reproductive age, and menopausal tissue and have been used to build a mathematical model. The model has also been designed based on machine learning to predict the probability of a specific ovarian tissue matrisome to resemble prepubertal, reproductive-age, or menopausal tissue (3). The complete list of ovarian matrisome proteins can be downloaded from (4): https://data.mendeley.com/datasets/22t56fbd6t.

Beyond the classical ECM definition: the human matrisome

ECM proteins represent one of the earliest recognized and most elaborate examples of exon (domain) shuffling during evolution (5). This distinctive feature of ECM proteins allows bioinformatic identification of the proteome encoded by any given genome, using around 50 domains to establish a list of ECM protein candidates (5). Using such methods combined with manual annotation, Hynes and Naba groups have built an in-silico matrisome database and generated a robust list of the proteins defining the mammalian matrisome by analysis of the human, mouse, and zebrafish genomes (6, 7). Thus, by definition, the matrisome is a subset of the entire proteome dedicated to ECM. It is divided into two main categories: core- and associated-matrisome. The core-matrisome is the set of high molecular weight and structural proteins synthesized by cells to form ECMs. The human genome encodes 44 collagen genes, 195 ECM glycoproteins and 35 proteoglycans, forming the core-matrisome (6). The associated-matrisome encompasses proteins secreted and either localized to the ECM or remodeling the ECM. Thus, the associated-matrisome is further divided into secreted factors localizing to ECMs (344 proteins), ECM-regulators (238 proteins), and ECM-affiliated (171 proteins) (6). Overall, in humans, the matrisome includes 1027 genes, representing about 4% of the human protein-encoding genome (8). However, each organ at each age represents a unique matrisome composition tailored to its function.

Advancing oncofertility through text-mining

We studied all possible effects of chemodrugs on the 98 detected ovarian matrisome proteins (1) that have been previously discovered by our group (Supp Table.1). Thus, by improving our understanding of the mechanisms involved, we hope to mediate the development of solutions limiting the negative impact of chemotherapy on ovarian reserve. To this end, we relied on Pathway Studio software (mammalcedfx.pathwaystudio.com) (Fig. 1). This Elsevier-associated text-mining software extracts relevant information from scientific publications on a large scale. It uses its proprietary Nature Language Processing (NLP) MedScan technology (9). Collected data are derived from more than 36 million abstracts from MEDLINE and more than four million full-text journal articles from Elsevier and other major scientific literature publishers. NLP is a collection of methods for semantic analysis of an unstructured text that allows for extracting specific items or facts of scientific interest. Basic NLP works by recognizing and capturing information triplets in the form of subject-verb-object statements. These triplets are the basic units of NLP technology. The recognition of these triplets is referred to as syntactic analysis (namely, recognition of these relationships in a sentence). For example, all the different ways in which a single gene or protein can be named is referred to as the dictionary. This is one of the very powerful assets in MedScan technology, as it identifies many synonyms that refer to the same object or entity. These information triplets, once recognized, must be understood in the context of particular scientific domains and definitions in a process known as semantic analysis. Defining these domain ontologies is part of the NLP approach.

Click to view Supplementary Table 1. NLP technology for deciphering possible ECM-chemotherapy interactions. Report of the results and references used to describe chemodrugs impact on ovarian matrisome proteins.

Here, Pathway Studio was used to elucidate the impact (e.g., positive/negative regulation; binding; molecular transport) of FDA-approved anticancer drugs on ovarian matrisome proteins (4, 10). The results obtained after entering the FDA-approved drug list described by Sun et al. (10)  interrogate on Pathway tudio the possible direct targets of each drug among the matrisome list (4) and describe the nature of their interactions.

Ovarian matrisome and chemotherapy: a potentially toxic relation

Angiotensinogen (AGT) is among the affected matrisome proteins with a characteristic expression level during reproductive age. AGT expression could be upregulated by Decitabine or Doxorubicin; both described as DNA-targeting drugs respectively prescribed for head and neck carcinoma and breast cancer or neuroblastoma (Suppl. Table 1). Conversely, AGT expression is inhibited by another DNA-targeting chemodrug, Actinomycin D, which is used for treating Ewing’s sarcoma and uterine carcinoma. Clearly, chemodrugs have collateral effects beyond their primary targets, reaching matrisome proteins and leading to their abnormal expression, which can affect ovarian homeostasis and negatively impact follicle reserve and fertility. We detected ovarian ECM proteins that have been tightly related to chemodrug resistance (TGFBI, COL6A3, ANXA1/5/7, ITIH4, POSTN, F2). These proteins have been differently expressed in prepubertal, reproductive-age, and menopausal ovarian tissues, which reveals different responses of ovarian tissue to chemodrugs according to these protein expression levels and prompts careful selection of cancer therapy according to patient age (1).  For instance, some studies have suggested that TGFBI, one of the matrisome proteins notably upregulated during prepuberty, sensitizes ovarian cancer cells to Paclitaxel by inducing microtubule stabilization and the loss of TGFBI induces drug resistance and mitotic spindle abnormalities in ovarian cancer cells (Suppl. Table 1). Another protein exclusively upregulated during prepuberty is POSTN, which has been associated with the induction of chemoresistance of colorectal cancer cells by activating Pi3k/Akt/survivin signaling pathway. At the same time, POSTN silencing could reverse chemoresistance in pancreatic cancer (Suppl. Table 1; Fig. 2). During menopause, we have detected a uniquely high level of ANXA5 expression (3). This protein was also associated with chemoresistance to Telozolomide in glioblastoma through a PI3K-dependent mechanism (Suppl. Table 1). In light of current results, it is tempting to think of therapeutic solutions targeting the expression of matrisome proteins to counter chemodrug toxicity at the follicle level or chemoresistance. However, the overexpression of some ECM proteins beyond their homeostatic level during reproductive age may lead to ovarian pathologies such as polycystic ovarian syndrome (PCOS) (Fig. 2), despite their upregulation during prepuberty or menopause (ANXA5, POSTN, SERPINA1) (3). This points to potential risks related to ECM-targeted therapies and prompts in-depth validations and safety monitoring in vitro and in vivo. Thus, overlooking the properties of the human ovarian ECM might jeopardize the efficacy of prescribed cancer treatment depending on the patient age and its corresponding matrisome.

Perspectives and reasons for caution

To overcome chemoresistance, it is imperative to study the ECM contribution and identify its alteration upon tumorigenesis by comparison to a reliable, detailed reference of the affected organ or tissue. Only lately have we started to tap into ovarian ECM implication in fertility, which prompts the development of new drugs targeting ECM deregulations. Caution must be exercised, as ECM modification can promote cancer metastasis (11). Modifying or degrading even part of the ECM might create a highway through which cancer cells can migrate to other tissues or organs.

Just like the survival of cancer patients matters, their well-being after cancer is also of major importance. Fertility restoration is one of the increasingly debated concerns, and ovarian matrisome reverse engineering will be the first sparkle shedding light on gonadotoxicity at the level of ECM and leading the way toward developing a transplantable engineered ovary as a fertility restoration alternative (12).

Acknowledgments

We thank Marta Da Pian and Hangbao Cao from Elsevier for their guidance on using the Pathway Studio program.

References

  1. Ouni E, Valerie N, Da Pian M, Cao H, T. Haas K, Peaucelle A et al. Proteome-wide and matrisome-specific atlas of the human ovary computes fertility biomarker candidates and open the way for precision oncofertility. Matrix Biology Journal (under review) 2022.
  2. Senthebane DA, Rowe A, Thomford NE, Shipanga H, Munro D, Mazeedi M et al. The Role of Tumor Microenvironment in Chemoresistance: To Survive, Keep Your Enemies Closer. Int J Mol Sci 2017;18.
  3. Ouni E, Nedbal V, Da Pian M, Cao H, Haas KT, Peaucelle A et al. Proteome-wide and matrisome-specific atlas of the human ovary computes fertility biomarker candidates and open the way for precision oncofertility. Matrix Biol 2022;109:91-120.
  4. Ouni E, Nedbal V, Da Pian M, Cao H, Haas KT, Peaucelle A et al. Proteome-wide and matrisome-specific atlas of the human ovary from prepuberty to menopause. , Mendeley Data, V1, doi: 10.17632/22t56fbd6t.1
  5. Naba A, Clauser KR, Hoersch S, Liu H, Carr SA, Hynes RO. The matrisome: in silico definition and in vivo characterization by proteomics of normal and tumor extracellular matrices. Mol Cell Proteomics 2012;11:M111 014647.
  6. Nauroy P, Hughes S, Naba A, Ruggiero F. The in-silico zebrafish matrisome: A new tool to study extracellular matrix gene and protein functions. Matrix Biol 2018;65:5-13.
  7. Naba A, Pearce OMT, Del Rosario A, Ma D, Ding H, Rajeeve V et al. Characterization of the Extracellular Matrix of Normal and Diseased Tissues Using Proteomics. J Proteome Res 2017;16:3083-91.
  8. Hynes RO, Naba A. Overview of the matrisome--an inventory of extracellular matrix constituents and functions. Cold Spring Harb Perspect Biol 2012;4:a004903.
  9. Novichkova S, Egorov S, Daraselia N. MedScan, a natural language processing engine for MEDLINE abstracts. Bioinformatics 2003;19:1699-706.
  10. Sun J, Wei Q, Zhou Y, Wang J, Liu Q, Xu H. A systematic analysis of FDA-approved anticancer drugs. BMC Syst Biol 2017;11:87.
  11. Winkler J, Abisoye-Ogunniyan A, Metcalf KJ, Werb Z. Concepts of extracellular matrix remodelling in tumour progression and metastasis. Nat Commun 2020;11:5120.
  12. Amorim CA, Asiabi P, Ouni E, Vilela J, Camboni A, Chiti MC. Transplantation of Isolated Follicles and the Engineered Ovary. In: Grynberg M, Patrizio P, eds. Female and Male Fertility Preservation. Cham: Springer International Publishing, 2022:353-67.

 

Please sign in or register for FREE

Your Fertility and Sterility Dialog login information is not the same as your ASRM or EES credentials. Users must create a separate account to comment or interact on the Dialog.