Another look


Hyaluronic acid and tissue regeneration

Denis Couchourel PhD, MBA

Denis Couchourel, PhD

Scientific & Medical Director Vivacy Laboratories


The first publications describing the clinical results of hyaluronic acid (HA) used as dermal filling material for aesthetic purposes date from the very early 2000s (Duranti F et al, 1998; Pollack S, 1999). (2,20)
Since then, the number of products registered for this type of indication has exploded, along with the number of articles published.
However, when the bibliographic search was expanded to include keywords such as "hyaluronic acid AND skin", the first data appeared in 1941 and the number of articles multiplied by 3. (Figure 1)
This simple observation illustrates the fact that endogenous HA has been studied for decades in basic biology, particularly in the context of the skin.
The number of these studies has also increased exponentially over the last 20 years, as analytical techniques were perfected and as research teams became aware of the importance of this molecule in the structure and especially in the physiology of the skin.
Nevertheless, it is clear that there are still some difficulties in bringing together the results generated by the two types of approaches.
For clinicians, the means of measuring the effectiveness of products after injection are rather well developed and reliable, especially since the advent of digital quantification techniques for volumes created on the face.
Numerous subjective, but validated, scales are also used to systematize published clinical evaluations. On the other hand, the fine interactions between the injected gel and its target tissue are still difficult to evaluate in humans, in particular because of the technical difficulties encountered and the impossibility of easily performing biopsies during the clinical efficacy of the product.
However, the data generated in basic sciences mentioned above could be of great help to us in trying to deepen the clinical results improving the patient's condition but also to work on the hypotheses allowing to explain some side effects, in particular delayed ones.

In a previous article in "Look & Medicine", we mentioned that a large number of cellular membrane and transmembrane receptors specifically dedicated to the biological action of HA were now well characterized.

The very existence of these structures within tissues such as the dermis or hypodermis is indisputable proof that HA can induce massive, complex, and sometimes essential cellular responses to major functions such as :
  •  angiogenesis,
  • inflammation,
  • the hydration of the tissues,
  • the innate immune response
  • or the healing process (Fallacara et al, 2018).
All these processes are essential for tissue regeneration and the role of HA can be highlighted by focusing on the consequences of the activation of its most well-known receptors. We will limit ourselves here to the 3 main ones: CD44, RHAMM and TLR.

1 The main HA-specific receptors

1-1 CD44

CD44 is a transmembrane receptor, of which 11 isoforms have been identified to date.

Functionally, CD44 is present in most tissues, but with a strong specificity for pluristratified epithelia, keratinized or not.

Its main characteristic is that it can bind with 2 types of HA:

  • High molecular weight HA (HMW-AH)

  • and low molecular weight HA (LMW-AH).

Activated CD44 then induces drastically different intracellular signals in either situation.

When HMW-HA interacts with CD44, the responses :

  • anti-inflammatory,

  • anti-angiogenic,

  • Inhibition of migration

  • and stimulation of cell proliferation are favored (Noble, 2002)(17).

Conversely, if we consider the binding of LMW-HA oligomers to CD44, activities :

  • pro-inflammatory,

  • cell mobility

  • and migration are clearly favored (Gao et al, 2008; Wang et al, 2011). (8,31)


RHAMM (Receptor for Hylauronan-Mediated motility) was the first HA receptor to be cloned by Dr. Turley's team in Canada.

It is present in many cell types, as is CD44.
Its cellular distribution is somewhat broader than that of CD44 because in addition to its membrane version, it has also been detected in the cytosol and nucleus of cells (Turley EA et al, 2002).(30)
The system of second messengers that it is able to trigger in the cell is complex and involves numerous kinases.
Like CD44, the interaction of RHAMM with HA also depends on the size of the fragment considered. For example, it was shown that if HA fragments binding to RHAMM were 8 to 2600 sugar units long, then the mitotic response was increased, as well as cell motility (Toole BP, 1997). (28)
In particular, this propensity of cells to be able to move is explained by the mobilization of cilia involving cytoskeleton proteins such as actin microfilaments and molecules specialized in adhesion function, such as connexins, (Sugahara KN et al, 2009; Nagi JI, 1996),(16,25)) or integrins.
Needless to say, this action also occurs in collaboration with the CD44-dependent migration functions discussed above.
Another example of the importance of these movements is of course embryogenesis. During these processes, intense phases of mitosis are systematically preceded by a phase of endogenous HA synthesis, first to allow the still totipotent cells to isolate and 'detach' themselves from the surrounding extracellular matrix, but also, once divided, to be able to guide their migrations along the different sheets of the growing embryo.
Then, there is a decrease in the amount of surrounding HA as the cell begins to differentiate. (Toole BP, 1997; Brecht M et al, 1986; Knudson CB et al, 1993).(13,28)

1-3 LRT

The large family of 'Toll-like' receptors (TLR) was initially discovered in Drosophila but the teams quickly realized that there were indeed homologous genes in humans.

The Toll-like receptor (TLR) family was initially discovered in Drosophila but the teams quickly realized that there were indeed homologous genes in humans.
These are transmembrane receptors that are present on the vast majority of cell types.
They are also likely to bind to a large number of ligands, among which HA figures prominently, especially for TLR2 and TLR4 receptors (Jiang D et al, 2005; Termeer C et al, 2002). (12,27)
For example, the interaction of HMW-HA with these TLRs protects the pulmonary epithelium from external aggression and decreases apoptosis (Jiang D et al, 2005). (12)
Conversely, if the ligands consist of LMW-HA, the expression of potent pro-inflammatory cytokines causes, among other things, the recruitment of immune cells.
The influx of macrophages and NK (natural killer) cells, the maturation of dendritic cells and Langherans cells, the migration of CD4+ T cells and to some extent the attraction of B cells cause the creation of a highly inflammatory and immunologically reactive environment, but which is a physiological step in the wound healing process (Singampalli KL et al, 2020; Fukui M et al, 2000; Jiang D et al, 2005; Hauck S et al, 2021; Termeer C et al, 2002; Larkin J et al, 2006; Iwata Y et al, 2009). (7,10,11,12,15,22,27)
There are of course many other receptors that have been identified as being capable of inducing a signal, a response to a stimulus caused by their binding to some form of endogenous HA.
However, in the 3 examples described above, it is interesting to note that the cellular responses can be drastically different depending on whether the stimulus is with high molecular weight HA or short chain HA.
Several theories have been put forward to explain this difference, but the one known as 'rafting' is the most commonly accepted.
It is illustrated by the figure 3:

As an illustration of the importance of HA in the vascularization process, we can cite Olivares et al (18) who demonstrated in 2016 the critical role that HA could have for angiogenesis.

Angiogenesis is of course a fundamental step in tissue repair and wound healing.

From this point of view, D'Agostino et al (3) showed very well in 2015 that HMW-HA, LMW-HA or even the combination between the 2 sizes of molecules did not have the same potentiality (Figure 5).

LMW-HA are much more active in the repair process, but it is also the mixture of the 2 species that gives the best results in the considered study model (D'Agostino A et al, 2015).(3)

Basic sciences show us that HA is not only an essential GAG for the mechanical strength of extracellular matrices, but also a biologically active molecule, capable of delivering a very broad spectrum of instructions to the surrounding cells through an extremely complex and refined system based essentially on the sizes of the HA considered.

Knowing that the human HA molecules produced are initially characterized by a high molecular weight (between 6000 and 7000 kDa for example in the synovial fluid of a healthy joint (Fraser JRE et al, 1997) (6), we must now look at the degradation mechanisms available to the organism to vary this size.

2 The degradation of HA by hyaluronidases

In mammals, hyaluronidases are endo-b-acetyl-hexosaminidases which are also able to degrade (but more slowly) other GAGs such as chondroitin sulfate.
In humans, 3 hyaluronidases have been identified (HYAL1, HYAL2 and HYAL3) with specific respective roles.
There is a 4th one (PH-20) but it can only be detected in the acrosome of the spermatozoon.
Furthermore, it is interesting to note that there is a syndrome in humans characterized by the absence of biological activity of HYAL-1.
Affected patients are generally of small stature and, above all, systematically present with generalized skin edema.
They also suffer from painful tissue masses around the joints and bilateral joint effusions. (Natowicz MR et al, 1996; Triggs-Raine B et al, 1999). (29)
Their macrophages and fibroblasts contain large vacuoles with a flocculent appearance and their blood HA concentration is about 40 times higher than normal.
HYAL1 and HYAL2 do not have the same activity. These enzymes are not interchangeable and intervene at different stages of the degradation of the HA molecule.
Figure 6 describes this sequence from the outside of the cell to the inside:
In this process, HMWs are therefore internalized into the cell by binding to CD44-HYAL2 complexes.
In the endosome thus formed, HMWs will be degraded into fragments of about 20 kDa.
The lysis process then continues with the intervention of HYAL1 which will result in the formation of very small tetramers, of about 800 Da.
All these steps are obviously very finely controlled by external actors (inhibitors of HYAL1 and HYAL2, ion exchangers since these enzymes are mostly active at acidic pH, etc).
This control is all the more important as we now know the potential potency and conflicting roles that HA fragments can hold depending on their size.
Feedback loops play a central role in this regulation and explain, for example, why the addition of exogenous hyaluronidase causes an increase in HA synthesis (Prehm P, 1984). (21)
All of these complex mechanisms make clinical sense when placed within physiological processes such as wound healing.
This is perhaps the best possible example to demonstrate the central role of HA in tissue regeneration. 


3. HA and wound healing (Hemostasis, inflammation, proliferation and remodeling)

Immediately after the skin injury, the healing process begins to restore the skin's architecture as soon as possible and stop the bleeding (Tavianatou AG et al, 2019). (26)
To do this, the platelets release a large amount of high molecular weight HA (HMW-HA) which facilitates the deposition of fibrinogen that will form the initial clot.
At this stage, HA is also extremely present in the edematous fluid (Figure 6).
Indeed, HA is also a molecule that can play a chemotactic role by attracting neutrophils involved in the phagocytosis of tissue debris and the subsequent release of 3 cytokines that are essential for the continuation of tissue regeneration (TNF-a, IL-1b, Il-8) Tavianatou AG et al, 2019). (26)
As the inflammatory phase progresses, the accumulated HA begins to depolymerize under the action of hyaluronidases. The resulting LMW-HA, associated with proteins such as fibronectin, guide the fibroblasts to the area to be recolonized.
Immediately, they begin to synthesize massive amounts of procollagen type I, de type III and for some, they are differentiated in myofibroblasts to be able to induce a "contraction" of the wound (Webber et al, 2009; Stern R et al, 2006) (24,32)

4 Injectable HA and tissue interactions

Given the above, it is obviously legitimate to wonder what the tissue consequences of injecting a cross-linked gel made from hyaluronic acid of bacterial origin might be.
We will briefly recall here that the process of cross-linking consists of: reacting HA with a molecule endowed with groups capable of covalently bonding with the available HA molecules.
In doing so, the HA chains physically move closer to each other and the degradation kinetics of the product formed slows down (Figure 7).
Thus, the clinical life of the product becomes compatible with the expectations of patients and practitioners. Gel manufacturers have all developed specific cross-linking techniques and also use different molecular weights in their initial formulation.
On the other hand, such a modification can have a significant impact on the propensity of HA molecules composing the cross-linked gel to bind to cellular receptors. In particular for CD44, a decrease in affinity of up to 40% has been demonstrated depending on the chemical modification considered. (Kwon MY et al, 2019). (14)
Thus, just because a cross-linked gel is made from small molecules of HA, it would not necessarily trigger more inflammatory reactions if we compared it to another product made from HMW-HA.
On the other hand, for products containing a portion of non-cross-linked HA, care must undoubtedly be taken not to use molecules that are too small to avoid activating a signaling pathway leading, for example, to significant recruitment of immune cells. 
However, the fact remains that even if it is slower, the enzymatic degradation of a cross-linked gel takes place normally in the body.
  • Indeed, the active sites of HYAL 1 and 2 are located at the glycosidic bonds between glucuronic acid and acetylglucosamine (DeBoulle et al, 2013) (5). Therefore, they are not modified by the BDDE cross-linking reaction.
  • This degradation will therefore necessarily lead to the generation of HA fragments of varying lengths in the immediate vicinity of the injected bolus.
  • In theory, these fragments are likely to play a major role in the evolution of tissues located immediately in contact with the gel. Thus, the synthesis of collagen or elastin via ligand-dependent activation of fibroblasts can be perfectly envisaged, whatever the size of HA initially used in the manufacture of the gel.
  • Conversely, the triggering of a delayed inflammatory reaction cannot be excluded after a certain level of product degradation.
  • Everything will depend on the size of the initial molecules, the intensity of the enzymatic activity developed in a particular patient, and of course the state of activation of his immune system.

5 Conclusion

HA is a major component present in all extracellular matrices. Because of its hydrophilic properties, it has long been restricted to a role in the management of interstitial fluids.

However, thanks to the progressive discovery of numerous receptors capable of interacting with it, a large number of studies have gradually brought to light its central role in processes as vital as angiogenesis or tissue regeneration in the context of wound healing.

These functions are tightly controlled via very complex mechanisms of action, but mostly based on size-dependent differential effects of HA.

These findings also help us to understand why the turnover of this endogenous molecule is so rapid, since the sequential degradation of HA allows the organism to finely modulate tissue responses in response to stress.

These results also open many perspectives.

To what extent will we be able to use these new molecular targets to add new functions to injectable HA gels?  

Or more precisely, would it be possible to control the length (and thus the biological effect) of HA fragments generated during the slow degradation of a cross-linked gel?

These questions are already topical because solving them would undoubtedly reduce even more the rate of delayed side effects still observed in aesthetic medicine.     


  1. Brecht M, Mayer U, Schlosser E, Prehm P. Increased hyaluronate synthesis is required for fibroblast detachment and mitosis. Biochem J. 1986 Oct 15;239(2):445-50.
  2. Duranti F, Salti G, Bovani B, Calandra M, Rosati Injectable hyaluronic acid gel for soft tissue augmentation. A clinical and histological study Dermatol Surg. 1998 Dec;24(12):1317-25.
  3. D'Agostino A, Stellavato A, Busico T, Daddy A, Tirino V, Papaccio G, La Gatta A, De Rosa M , Schiraldi In vitro analysis of the effects on wound healing of high- and low-molecular weight chains of hyaluronan and their hybrid H-HA/L-HA complexes. BMC Cell Biol. 2015 Jul 11;16:19.
  4. D'AgostinoAMaritato R,  Gatta A, Fusco AReale SStellavato SAdriana Pirozzi AVDe Rosa MDonnarumma GSchiraldi In Vitro Evaluation of Novel Hybrid Cooperative Complexes in a Wound Healing Model: A Step Toward Improved Biorepair. Int J Mol Sci. 2019 Sep 24;20(19):4727.
  5. From Boulle KGlogau R, Kono T, Nathan M, Tezel A, Roca-Martinez JX, Paliwal S, Stroumpoulis  A review of the metabolism of 1,4-butanediol diglycidyl ether-crosslinked hyaluronic acid dermal fillers. Dermatol Surg. 2013 Dec;39(12):1758-66.
  6. Fraser J.R.E., Laurent T.C., Laurent U.B.G. Hyaluronan: Its nature, distribution, functions and turnover. J. Intern. Med. 1997;242:27-33.
  7. Fukui, M.; Whittlesey, K.; Metcalfe, D.D.; Dastych, J. Human Mast Cells Express the Hyaluronic-Acid-Binding Isoform of CD44 and Adhere to Hyaluronic Acid. Clin. Immunol. 2000, 94, 173-178
  8. Gao F, Yang CX, Mo W, Liu YW, He YQ. Hyaluronan oligosaccharides are potential stimulators to angiogenesis via RHAMM mediated signal pathway in wound healing. Clin Invest Med 2008;31:E106-16.
  9. Graça MFPMiguel SP Cabral CSDCorreia Hyaluronic acid-Based wound dressings: A review. Carbohydr Polym 2020 Aug 1;241:116364.
  10. Hauck, S.; Zager, P.; Halfter, N.; Wandel, E.; Torregrossa, M.; Kakpenova, A.; Rother, S.; Ordieres, M.; Räthel, S.; Berg, A.; et al. Collagen/hyaluronan based hydrogels releasing sulfated hyaluronan improve dermal wound healing in diabetic mice via reducing inflammatory macrophage activity. Bioact. Mater. 2021, 6, 4342-4359.
  11. Iwata, Y.; Yoshizaki, A.; Komura, K.; Shimizu, K.; Ogawa, F.; Hara, T.; Muroi, E.; Bae, S.; Takenaka, M.; Yukami, T.; et al. CD19, a Response Regulator of B Lymphocytes, Regulates Wound Healing through Hyaluronan-Induced TLR4 Signaling. Am. J. Pathol. 2009, 175, 649-660.
  12. Jiang D, Liang J, Fan J, Yu S, Chen S, Luo Y, et al. Regulation of lung injury and repair by Toll-like receptors and hyaluronan. Nature Medicine. 2005;11(11):1173-9.
  13. Knudson CB, Knudson W. Hyaluronan-binding proteins in development, tissue homeostasis, and disease. FASEB J 1993;7:1233-41.
  14. Kwon MYWang CGalarraga JHPuré EHan LBurdick JA. Influence of hyaluronic acid modification on CD44 binding towards the design of hydrogel biomaterials. Biomaterials. 2019 Nov;222:119451.
  15. Larkin, J.; Renukaradhya, G.J.; Sriram, V.; Du, W.; Gervay-Hague, J.; Brutkiewicz, R. CD44 Differentially Activates Mouse NK T Cells and Conventional T Cells. J. Immunol. 2006, 177, 268-279.
  16. Nagy JI, Hossain MZ, Lynn BD, Curpen GE, Yang S, Turley EA. Increased connexin-43 and gap junctional communication correlate with altered phenotypic characteristics of cells overexpressing the receptor for hyaluronic acid-mediated motility. Cell Growth Differ 1996;7:745-51.
  17. Noble PW. Hyaluronan and its catabolic products in tissue injury and repair. Matrix Biol 2002;21:25-9
  18. Olivares CN,  Alaniz LD,  Menger MD, Barañao RI, Laschke MW,  Meresman Inhibition of Hyaluronic Acid Synthesis Suppresses Angiogenesis in Developing Endometriotic Lesions. PLoS One. 2016 Mar 28;11(3):e0152302.
  19. Pardue EL, Ibrahim S, Ramamurthi Role of hyaluronan in angiogenesis and its utility to angiogenic tissue engineering Organogenesis 2008 Oct;4(4):203-14.
  20. Pollack S Some new injectable dermal filler materials: Hylaform, Restylane, and Artecoll.
  21. J Cutan Med Surg. 1999 Dec;3 Suppl 4:S27-35.
  22. Prehm, P., 1984. Hyaluronate is synthesized at plasma membranes. Biochem. J. 220, 597 } 600
  23. Singampalli, K.L.; Balaji, S.; Wang, X.; Parikh, U.M.; Kaul, A.; Gilley, J.; Birla, R.K.; Bollyky, P.L.; Keswani, S.G. The Role of an IL-10/Hyaluronan Axis in Dermal Wound Healing. Front. Cell Dev. Biol. 2020, 8, 636.
  24. Stern R. Hyaluronan catabolism: a new metabolic pathway. Eur J Cell Biol 2004 Aug;83(7):317-25
  25. Stern, R., Asari, A. A., & Sugahara, K. N. (2006). Hyaluronan fragments: An information rich system. European Journal of Cell Biology, 85(8), 699-715.
  26. Sugahara KN. Hyaluronan fragments: informational polymers commanded by cancers. In: Stern R, editor. Hyaluronan in cancer biology. San Diego, CA, USA: Academic Press, 2009:221-54
  27. Tavianatou, A. G., Caon, I., Franchi, M., Piperigkou, Z., Galesso, D., & Karamanos, N. K. (2019). Hyaluronan: Molecular size-dependent signaling and biological functions in inflammation and cancer. The FEBS journal, 286(15), 2883-2908.
  28. Termeer C, Benedix F, Sleeman J, Fieber C, Voith U, Ahrens T, et al. Oligosaccharides of Hyaluronan activate dendritic cells via toll-like receptor 4. J Exp Med. 2002;195(1):99-111.
  29. Toole BP. Hyaluronan in morphogenesis. J Intern Med 1997;242:35-40.
  30. Triggs-Raine,B., Salo, T. J., Zhang, H., Wicklow,B. A., Natowicz, M. R., 1999. Mutations in HYAL1, a member of a tandemly distributed multigene family encoding disparate hyaluronidase activities, cause a newly described lysosomal disorder, mucopolysaccharidosis IX. Natl. Acad. Sci. USA 96, 6296 - 6300.
  31. Wang YZ, Cao ML, Liu YW, He YQ, Yang CX, Gao F. CD44 mediates oligosaccharides of hyaluronan-induced proliferation, tube formation and signal transduction in endothelial cells. Exp Biol Med (Maywood) 2011;236:84-90.
  32. Webber, J, Jenkins, R. H., Meran, S., Phillips, A., & Steadman, R. (2009). Modulation of TGFβ1-dependent myofibroblast differentiation by hyaluronan. The American Journal of Pathology, 175(1), 148-160.