Hyaluronan metabolism in naked mole-rat

PhD thesis defended by Delphine DEL MARMOL (Prof. Bruno FLAMION) - 05/11/2021



Prof. Bruno FLAMION, UNamur, Molecular Physiology Research Unit (URPhyM) - promoter

Prof. Yves POUMAY, UNamur, Molecular Physiology Research Unit (URPhyM), Cell and Tissue Laboratory (LabCeTi) - co-promoter

  • Prof. Aaron PETREY, University of Utah, Salt Lake city, USA
  • Prof. Marielle BOONEN, Université de Namur
  • Prof. Bruno FLAMION, Promoteur, Université de Namur
  • Prof. Romain FONTAINE, Institut Cochin, Paris, France
  • Prof. Pierre GARIN, Président, Université de Namur
  • Prof. Suzanne HOLZE, Leibniz-Institut für Zoo – und Wildtierforschung (IZW), Berlin, Allemagne
  • Prof. Michel JADOT, Université de Namur
  • Prof. Yves POUMAY, Co-promoteur, Université de Namur
  • Prof. Jean-Michel VANDEWEERD, Université de Namur

Hyaluronan (HA) is a glycosaminoglycan polymer found ubiquitously in the extracellular matrix of all vertebrate tissues:  the vitreous humor of the eye, umbilical cords, synovial fluid, heart, and skin are especially rich in HA. Together with other connective tissue components, the HA molecules create a network in which they have structural and hydration roles. HA itself interacts with diverse cellular receptors and binding proteins to generate different physiological and pathological responses depending on its molecular weight (MW). High MW (HMW) HA (> 1000 kDa) shows immunosuppressive, anti-angiogenic, and anti-inflammatory features whereas low MW HA has opposite effects. HA undergoes fast turnover with an in situ degradation of ca 30% while the other 70% are drained through the lymphatic system where most of them are degraded in the lymph nodes. HA that is discharged into the blood is taken by the liver and degraded. HA is produced, at the plasma membrane of many cells, with slightly different average sizes and at different rates by HA synthases (HAS1, HAS2, HAS3). HA is degraded by hyaluronidases at the cell surface (HYAL2) and in lysosomes (HYAL1) after endocytosis. The role of HYAL3 remains unclear even though it could increase HYAL1 activity.

The naked mole-rat (NMR, Heterocephalus glaber) is a eusocial, long-lived, tumor-resistant, subterranean rodent found in East Africa. Beyond its eusociality (well-defined hierarchy similar to bees and termites) and the protective underground habitat which contribute to its extended lifespan (>30 years), NMR harbours some remarkable features characterizing its healthy ageing. Indeed, only minor changes in basal metabolic rates, skeletal muscles fibers, bone mineral density, cardiovascular, neurovascular and gastrointestinal functions, or fertility are observed in ageing NMR. The NMR is seen as a healthy ageing animal that does not develop age-associated diseases like cancer, neurodegenerative disease, or cardiovascular diseases. Several hypotheses have been proposed to explain these features. Reactive oxygen species tolerance, low metabolic rate, low mutation rate, better repair and cytoprotective pathways, improved proteostasis, special mitochondria, stronger telomeres, and enhanced proteasome activity are among the diverse NMR adaptations but their protective mechanisms are not fully understood so far.

In 2013, the NMR was described by Tian et al. in a prominent Nature article as harbouring very high levels of very-HMW HA (vHWM) (6-12 000 kDa) in comparison to laboratory mouse (Mus musculus) and guinea pig (GP, Cavia porcellus) tissues. Tian et al. claimed that the NMR vHMW was the key factor endowing this rodent with its astonishing tumor resistance. This claim was based on the observation that the larger MW HA produced by NMR fibroblasts was required to produce enhanced early contact inhibition (ECI). Contact inhibition, i.e., the stop of cell growth when they touch each other or the ECM, is a fundamental mechanism which is lost in cancer cells. It appears from Tian et al.'s experiments that NMRs have developed a hypersensitive ECI triggered by its vHWM HA binding to CD44 and subsequently activating the tumor suppressor Nf2/Merlin pathway. This ECI is inhibited by adding hyaluronidase to the culture medium of NMR fibroblasts. Lower hyaluronidase activity (in fibroblasts and tissues) combined with a unique sequence of HAS2 were proposed as additional factors responsible for vHMW HA production. Moreover, the Has2 gene was shown to be overexpressed in NMR skin fibroblasts whereas Has1 and Has3 did not show any peculiarity. Later on, it was shown that the vHMW HA produced by NMR fibroblasts, in contrast to mouse fibroblast HA, exhibited special cytoprotective properties against various stresses. NMR HA has also been identified as exhibiting specific folded structures and assemblies in NMR vs mouse tissues.

However, the initial results of Tian et al. were obtained using a non-specific Alcian blue staining to assess global HA amount and pulsed-field electrophoresis to assess HA MW. Surprisingly, the observations on NMR tissue and serum HA, or on ECI, have not been reproduced so far. In contrast, I decided to use specific and state-of-the-art methods such as histochemistry with a HA binding protein, quantification using an enzyme-linked immunosorbent-like assay, and determination of HA molecular size using size exclusion chromatography, to better characterize the features of HA in the NMR. GP, a close relative of NMR that does not exhibit cancer resistance and healthy ageing, was used as a negative control. Some experiments were also performed in another long-lived and tumor resistant mole-rat, the Ansell’s mole-rat (AMR, Fukomys anselli). Using RNA sequencing, we also measured the expression of several “HA family” genes such as HASs, HYALs, HABPs, and HA receptors, in various NMR tissues.

Our results confirm a higher average HA size in NMR than in GP and mouse peripheral tissues and serum, and constant exposure of NMR tissues and blood to a high proportion of HMW HA compared with LMW HA. Nevertheless, we were unable to detect any relevant amount of HA molecules larger than 2500 kDa, even in the supernatant of fibroblast cultures. We did not observe any ECI in NMR skin fibroblast cultures. We did not find direct explanation for the discrepancy between our results and those of Tian et al. apart from the difference in methods. We cannot exclude the possibility that NMR tissues contain particular HA assemblies with or without specific proteins that resolve themselves in size-exclusion chromatography and agarose gel electrophoresis but not in pulse-field electrophoresis.

In addition, RNAseq data in NMR tissues revealed extremely large up-regulation of two HA-related genes: Hyal3 and Tnfaip6. Regarding Hyal3, in human and mouse, there is a clear link between HYAL3/HYAL3 and HYAL1/HYAL1; therefore, we decided to explore HYAL1 activity in blood and lymph nodes of NMR, GP, and mouse. HYAL1 activity was lower in NMR lymph node and higher in NMR blood vs GP and mouse. NMR HYAL1 may show a greater extracellular release than in other species, as reflected by circulating blood levels. This increased extracellular release may in turn be due to high levels of HYAL3.  Regarding Tnfaip6, its protein product, TSG6, is a HA-binding protein with a key role in inflammation as well as in ovulation. TSG6 is known to be induced by progressive inflammation and, conversely, to reduce inflammatory cues. TSG6 is able to interact with HA and induce deep structural changes to the HA network. TSG6 also shows interaction with CD44; a feedback between HA, CD44, and TSG6 has been described but is still poorly understood. The link between HA, CD44, and TSG6 could be the basis of a dynamic and unique mechanism in NMR cells and tissues, which may be involved in the generation of large HA structures. Preliminary results on AMR show similar HA profiles compared to NMR; due to similar cancer resistance in NMR and AMR, HA turnover and structure should be explored further in AMR.

In parallel with the study of HA in NMR, I also performed experiments in Hyal2 KO mice (the Hyal2-/- mouse also harbors HMW HA). However, unlike the NMR, the Hyal2-/- mouse has a decreased lifespan and shows blood abnormalities. The Hyal2-/- mouse is affected with a mild and chronic thrombotic microangiopathic anemia with the presence of schizocytes (fragmented red blood cells) and peripheral thrombocytopenia. HYAL2 was found to play a main role in megakaryocyte maturation and platelet activation in inflammatory diseases like colitis. So, we decided to build a new mouse model (C57Bl/6J-Hyal2fl/flPf4-creERT2) with HYAL2 extinction specifically targeted to megakaryocytes and platelets, using the CreERT2 specific and inducible system. Preliminary results show the efficiency of the system with an extinction of HYAL2 in platelets. This new mouse model opens new horizons for the understanding of HA and HYAL2 implications in inflammatory diseases.