Therapeutic applications of hyaluronan

Abstract

Hyaluronan (HA), a multifunctional, high molecular weight glycosaminoglycan, is a component of the majority of extracellular matrices. HA is synthesised in a unique manner by a family of hyaluronan synthases, degraded by hyaluronidases and exerts a biological effect by binding to families of cellular receptors, the hyaladhedrins. Receptor binding activates signal pathways in endothelial cells leading to proliferation, migration and differentiation collectively termed angiogenesis. HA and associated enzymes are implicated in the aetiology of cardiovascular disease and cancer and manipulation of HA expression offers a therapeutic target. HA microspheres have been developed as drug delivery agents to deliver HA to sites of disease and also in diagnosis. In this review we discuss some of the recent therapeutic applications of hyaluronan in tissue repair, as a drug delivery system and the synthesis, application and delivery of hyaluronannanoparticles to target drugs to sites of disease.

---

Introduction

Hyaluronan (HA), a ubiquitous, non-sulfated glycosaminoglycan (GAG), is a linear polysaccharide of 2000–25 000 repeating units of the dissacharide [–β(1,4)-D-glucuronic acid-β1,3-N-acetyl-D-glucosamine]_n. The molecular mass of HA is in the range 10²–10⁴ kDa (n = 25–25 000) and it has an extended length of 2–25 μm. The biological roles of hyaluronan are in part dependent on its hydrophilic and hydrodynamic properties which allow it to retain water and play a structural role, for example in the adult vitreous, synovial fluid and dermis.¹HA forms loose, hydrated matrices with a pseudo-random coil configuration creating space to facilitate cell division and migration in the embryo² and increase cell motility and decreasing cell–cell contact in post-embryonic development.³ At a cellular level HA interacts in an autocrine manner with HA receptors and in a paracrine manner with receptors on neighbouring cells. Because of its size HA may interact with more than one cell and these interactions may play a role in tissue assembly. Finally, HA may bind to diverse extracellular matrix (ECM) proteins which in turn link to the plasma membrane through HA receptors.⁴ Reed et al.⁵ showed that in the rat about 50% of the total HA was located in the skin, 25% in the skeleton and the remainder divided equally between muscles and viscera. In man⁶ the concentration of HA in the skin is 200 μg g⁻¹ and in the brain 35–115 μg g⁻¹.

Hyaluronan synthesis

Unlike other GAGs which are synthesised in the Golgi, HA is synthesised on the cytoplasmic surface of the plasma membrane and translocated to the pericellular space. HA is synthesised as a free glycan, not attached to a protein, a primer is not required but Mg²⁺ or Mn²⁺ is essential for synthesis. HA is produced by hyaluronan synthases (HAS; EC 2.4.1.212) located on the plasma membrane and unusually a single enzyme attaches both sugar residues to the nascent HA.⁷ Structurally all HASs are integral membrane proteins with several transmembrane regions and large cytoplasmic loops. Prehm proposed a mechanism for HA synthesis in which the enzyme contains two separate recognition sites for N-acetyl-D-glucosamine (GlcNAc) and glucuronic acid (GlcUA),⁸ which are attached to the growing HA as their uridine diphosphate (UDP) conjugates.⁹ There are two methods of elongating the HA molecule: by addition of the new sugar to the reducing end (eqn (A)) or to the non-reducing end (eqn (B)) of the HA molecule.⁹ In both cases the formation of the new glycosidic bond involves displacement of UDP. Human HASs use mechanism B.

HA–GlcNAc + GlcUA–UDP → UDP + HA–GlcNAc(β1,4)GlcUA (A)

GlcUA–UDP + GlcNAc–HA → UDP + GlcUA(β1,3)GlcNAc(β1,4)–HA (B)

In chordates HASisozymes are encoded by three to four genes and the mammalian HAS-1, -2 and -3 enzymes exhibit 55–70% structural homology.¹⁰ Both HAS-1 and -2 generate high molecular weight HA (RMM > 2–4 × 10⁶ Da) although HAS-2 is more catalytically active. HAS-2 synthesises HA in response to shock, inflammation and in tissue repair while HAS-3 has the highest catalytic activity and produces low molecular weight HA (RMM < 2 × 10⁵ Da). The precise molecular weights of the HA produced by the HASs depend on a number of undetermined factors and in particular the cell type. HAS-3 may be responsible for synthesis of the HA constituting the pericellular coat of the cell.¹¹ However, other studies have indicated that this coat may be as thick as 20 μm and corresponds to an extended high molecular weight HA molecule.¹² The pericellular matrix is assembled through HA crosslinking with the proteins versican and tenascin-C in a calcium-dependent manner.¹² HASs may not only synthesise and secrete HA but bind and retain it at the cell surface.

Turnover of hyaluronan

HA degradation in peripheral tissue takes place both in the tissue and on release into the lymph and vascular systems. The HA entering the lymph system is removed in the lymph nodes and that entering the circulation is removed by the liver and kidney.⁹ The uptake of HA from the circulation is mediated by two receptors: the HA receptor for endocytosis (HARE) and lymphatic vessel endothelial HA receptor (LYVE). It has been estimated that the half-life of HA in the skin is about 1 day, in the eye 1–1.5 h, in cartilage 1–3 weeks and in the vitreous humour 70 days.⁹HA degradation occurs in lysosomes which contain hyaluronidases (HYALs) and other degradative enzymes.

Mammalian hyaluronidases (for example testis hyaluronidase, EC 3.2.1.35) are endo-β-N-acetylhexaminadases and cleave β-1-4 glycosidic bonds not only in HA but also in chondroitin and chondroitin sulfates. The major products are tetra- and hexaoligosaccharides which are further degraded by two exoglycosidases (β-glucuronidase and β-N-acetyl hexosaminadase) which remove the terminal sugar residues.¹³HYALs also possess transglycosidase activity. In humans six hyaluronidasegenes have been identified in somatic tissue. Sequences for HYAL-1, -2 and -3 are located on chromosome 3p21.3 and HYAL-4, a pseudogene (HYALP1) and sperm adhesion molecule 1 (SPAM1) are located on 7p31.3. However, only HYAL-1 and -2 are expressed to any extent in somatic tissue. HYAL-2 is linked through a glycophosphoinositol (GPI) anchor to the cell membrane¹⁴ and cleaves high molecular weight HA to fragments of approximately 20 kDa (50 dissaccharide units). HYAL-1, a lysosomal enzyme, acts in concert with HYAL-2 in somatic tissue to produce tetrasaccharides of hyaluronan;¹⁵ little is known of the specificity of HYAL-3. In all cases HA is taken up by the hyaluronan receptor CD44 receptor for degradation.

Hyaluronan receptors

The HA receptors (the hyaladherins) can be divided into those containing a common domain of about 100 amino acid residues, the link module or proteoglycantandem repeats (PTRs) and those unrelated to each other. The PTR module has two α-helices and two triple stranded anti-parallel β-sheets arranged around a hydrophobic core.¹⁶ The most studied of this group are CD44, RHAMM (the receptor for hyaluronan-mediated motility) and TSG-6 (tumour necrosis factor-α-stimulated gene-6). A cluster of four basic amino acids in the link domain (Arg-41, Tyr-42, Arg-78, Tyr-79) are essential for HA binding.¹⁷

CD44, a cell surface receptor for HA, is expressed on the majority of cells of neuroectodermal origin. It is a type 1 transmembrane glycoprotein coded by a single gene and produced in multiple isoforms by rearrangement of splice variants. The CD44 gene contains 21 exons of which 7 code for the extracellular domain in the most widely expressed variant CD44v6.¹⁸ CD44 may also be modified by post-translational modifications such as glycosylation and attachment to other GAGs. Unlike other hyaladhedrins, CD44 requires additional amino acid sequences distant from the link module, other cell-specific factors and only becomes functional on binding to multiple HA molecules.¹⁹HA binding to CD44 occurs at the cell surface to multiple CD44 receptors each with a low affinity. It is thought that CD44 may exist in a variety of states: inactive and unable to bind HA, constitutively active or inducible (can bind HA on contact with inducers). The cellular response following CD44 binding to HA is determined by the size of the HA fragment. Low molecular weight HA (3–10 disaccharide units; o-HA) is pro-angiogenic, induces the formation of new blood vessels, and activates a signal transduction pathway leading to endothelial cell proliferation and migration. In contrast, native high molecular weight HA (n-HA) is anti-angiogenic and will inhibit blood vessel formation.

We have recently shown an involvement of CD44 in phosphorylation of γ-adducin leading to a change of endothelial cell (EC) shape. Using confocal microscopy we showed a diffuse perinuclear distribution of γ-adducin in ECs and the organisation of F-actin stress fibres gave the cell a distinct triangular shape.²⁰ In CD44 knockout cells there was a rearrangement of F-actin fibres with a change of morphology. On addition of o-HA increased staining for γ-adducin was observed with more pronounced actin fibres and on knockout of CD44 the effect was substantially reduced (Fig. 1). These data indicate a role for CD44 in cytoskeletal rearrangement determining cell shape in the presence of o-HA.

Fig. 1 Photomicrographs showing the involvement of CD44–phospho-γ-adducin in o-HA-induced F-actin cytoskeleton rearrangement. Confocal photomicrograph (original magnification ×400) and insert (×630) showed a diffuse and perinuclear (blue) distribution of phospho-γ-adducin in endothelial cells in the absence of o-HA. The organisation of F-actin stress fibres (red) gave cells a triangular shape. CD44 knockout cells treated with anti-CD44 siRNA, showed an elongated shape with a rearrangement of F-actin stress fibres organised in parallel. After addition of o-HA increased staining for phospho-γ-adducin was observed with a perinuclear distribution and F-actin stress fibres were more pronounced. CD44 knockout cells showed a general reduction in fluorescent staining of phospho-γ-adducin and a disappearance of F-actin stress fibres.²⁰

TSG-6 is a 35 kDa protein not constitutively expressed but up-regulated in tissue in response to inflammatory mediators, for example epidermal growth factor (EGF), fibroblast growth factor-2 (FGF-2) and transforming growth factor-β (TGF-β)²¹ and which also interacts with HA through a link module. RHAMM is a 55 kDa protein distributed both on the cell surface and intracellularly (in the cytoskeleton, mitochondria and nucleus). These receptors will not be discussed any further.

HA fragment size affects the angiogenic response

Both native, high molecular weight (n-HA) and o-HA bind to receptors and initiate signal transduction pathways leading to the different stages of angiogenesis.

n-HA is anti-angiogenic inhibiting EC proliferation, migration and capillary tube formation.²² The mechanism of inhibition is not clear but in a variety of cell lines , n-HA binds to a 34 kDa member of the hyaladhedrins, increasing phosphorylation of a range of proteins including phospholipase Cγ.²³ The pathways through which o-HA modulates EC migration and proliferation are better understood. Activation of Raf-1, extracellularly regulated kinases 1/2 (ERK1/2) and of early response genes including c-fos and c-jun²⁴ occurs 2–7 minutes after stimulation in bovine aortic endothelial cells (BAEC). There is evidence that both n-HA and o-HA compete for the same receptor since addition of n-HA to BAEC blocked activation of c-jun and c-fos.

The role of hyaluronan in disease progression

Cerebrovascular disease

Cerebrovascular disease is one of the major causes of death and disability in developed countries and HA, HA metabolising enzymes and receptors have been implicated in the progression and recovery from disease. The ECM of the central nervous system is poorly characterised but is unusually rich in proteoglycans including hyaluronan and tenascins which have a role, in conjunction with other proteins, in guiding and limiting neurite outgrowth.^25,26 Several studies have shown HA is a major component of the central nervous system and it is expressed by astrocytes and to a lesser extent other glial cells (reviewed in ref. 27).The expression of HA-binding molecules and their ability to compete with proteoglycans for HA binding determines the properties of the brain ECM and enhances the migration of cells in the brain. In white matter HA is associated with myelinated fibres and in grey matter HA is located around neuroncell bodies. Galtrey et al.²⁸ found HA associated with perinuclear networks in grey and white matter and identified the presence of HASs but did not specify which ones were expressed.

Cerebral ischaemia is a pathophysiological condition produced by blockage of cerebral vessels resulting in occlusion of blood flow and deprivation of glucose and oxygen leading to cell death.²⁹ The extent of recovery from stroke is related to the survival of neurons, especially in the peri-infarcted regions (the area of potentially viable tissue surrounding the infarcted core). There is also a relationship between morbidity, survival time and the density of blood vessels in the area of infarction which in turn is affected by production of angiogenic forms of HA.

Data on the cellular expression of specific HYALs, HASs and HA receptors after stroke in human subjects are limited (reviewed in ref. 24). However, Wang et al.³⁰ used suppression subtractive hybridisation and Western and northern blotting to study gene up-regulation following induced focal stroke in a rat model. They found concomitant CD44 and HAS-2mRNA up-regulation suggesting increased biosynthesis of high molecular weight HA. The two molecules co-localised to inflammatory areas in the brain and may be involved in tissue remodelling which occurs after stroke. Our more detailed study using a set of human subjects who survived for different times after stroke showed a time-dependent distribution of HA-associated proteins.³¹ Using a HA-specific biotinylated probe HA accumulation in both infarcted tissue and serum was observed up to 37 days after stroke. The newly synthesised HA was localised in blood vessels and the nuclei of neurons in peri-infarcted regions (Fig. 2). Using immunohistochemistry (IHC) we found expression of HAS-1 (Fig. 2) and HAS-2 in inflammatory cells and of HAS-2 in neurons in peri-infarcted regions. This would suggest the synthesis of high molecular weight HA following stroke. HAS-3 was not expressed. HYAL-1 (Fig. 2) and -2 were also up-regulated in stroke tissue suggesting degradation of high molecular weight HA to pro-angiogenic forms is a feature of stroke. This is an interesting finding since HYAL-2 is not normally expressed in brain tissue since the gene is inactivated by hypermethylation³² but it has been shown to be expressed in intracerebral tumours. Interestingly, HYAL-1 has also been seen in the spinal cords of rats after injury, together with HA degradation.³³HYAL-1 was also detected in ECs in microvessels suggesting an association with angiogenesis and CD44 was expressed in both large neurons and ECs. CD44 expression is elevated in astrocytes and some populations of microglia in a variety of animals after injury^34,35 and in neurones after transection,³⁶ suggesting that the cells have an increased HA affinity. One technical problem with these types of studies is that an antibody capable of discriminating different sizes of HA is not available and HA size can only be determined by tissue extraction and separation. Other studies³⁷ inducing stroke using middle cerebral artery occlusion (MCAO) in the rat have shown increased staining for HA throughout the stroke area indicating increased synthesis possibly accompanying tissue remodelling. Using the reverse transcriptasepolymerase chain reaction (RT-PCR) increased mRNA for HYAL-1 and -2 was seen in stroke-affected tissue within 1 h of stroke induction, reaching a maximum after 3 days and persisting for at least 21 days. IHC showed expression of the enzymes in neurones suggesting a rapid degradation of hyaluronan at the site of injury which could enhance neuronal plasticity or stimulate angiogenesis.³⁷mRNA for CD44 was also up-regulated for the same time periods as the HYALs and the expression was greater in peri-infarcted tissue. CD44 was localised to blood vessels and microglia by IHC. CD44 plays a role in inflammation following stroke including leukocyte adhesion, cell–matrix interactions, metabolism of HA and matrix remodelling. CD44 gene expression was found to be regulated by HA fragments and interleukin-1β³⁸ both of which are up-regulated in stroke tissue. Wang et al.³⁰ employed a dominant negative strain of CD44 −/− mice subjected to middle cerebral artery occlusion and found infarct size and neurological deficit were reduced compared to the wild type. The difference was due to a reduced inflammatory response mediated by CD44. There is therefore clear evidence of the involvement of n- and o-HA and HA receptors in tissue remodelling and angiogenesis following stroke. Since early angiogenesis after stroke is related to survival, manipulation of this process would have therapeutic benefits.

Fig. 2 A. (i) Normal human grey matter showing weak staining for HA around blood vessels but (ii) in infarcted grey matter staining was more intense, demonstrating increased synthesis of HA. B. (i) Staining for HAS-1 in neurons in infarcted grey matter and (ii) HYAL-1 localisation in neurons (arrows) in peri-infarcted areas of the human brain. C. (i) In normal grey matter there was little staining for CD44 but (ii) neurons in peri-infarcted regions showed intense staining. All photomicrographs are from ref. 31.

Tumourigenesis

Tumourigenesis is associated with increased HA synthesis, for example in gastric carcinoma³⁹ and stromal levels of HA correlated with poor prognosis in ovarian carcinoma.⁴⁰ Toole and Hascall⁴¹ found a correlation between increased expression of HA and tumour growth and metastasis in animal models and that HA promoted tumour cell invasion and the effects were mediated by signal transduction pathways initiated through CD44. Malignant transformation is often involved by overproduction of HA⁴² and is accompanied by transcriptional regulation of HASgenes (reviewed in ref. 43). Experimental manipulation of HAS-2 and HAS-3genes in melanoma and mesothelioma⁴⁴cell lines led to increased tumourogenicity and conversely suppression of these genes reduced HA synthesis with concomitant reduction in tumourogenicity. However, the levels of HA in the tumour microenvironment may be more important since in glioma and prostate overexpression of HAS-2 suppresses growth which is restored by induction of HYAL-1.⁴⁵ Other studies have demonstrated tumour vascularisation adjacent to HA-rich areas associated with the generation of low molecular weight o-HA. In contrast recent reports have suggested that o-HA may counteract tumour development. Ghatak et al.⁴⁶ reported inhibition of in vivo growth of LX-1 human lung carcinoma by o-HA of 6–20 disaccharides and it has also been noted that HA fragments exert a pro-apoptotic effect in tumours.⁴⁷ These effects may be explained by o-HA perturbing the interaction between CD44 and n-HA.⁴⁸

Increased expression of CD44 variants has been noted on gastric tumours,⁴⁷ in invasive breast cancer and on microvessels from human melanoma and cancers of the glottis.⁴⁹

Since HA and its metabolism is perturbed in major groups of disease it has been proposed as a therapeutic target.

Therapeutic applications of hyaluronan

Hyaluronan has been utilised as a drug delivery system, providing a support for tissue repair and, in conjunction with nanoparticles, in targeting drugs to tumour cells.

HA as a drug delivery system

HA particles have advantages over other drug delivery systems of biocompatibility, non-immunogenicity and biodegradability. The use of HA as a delivery system has been developed in animal models and limited clinical trials in cancer patients are now in progress. A recent example of the use of HAnanoparticles involved delivery of the tumour necrosis factor-related apoptosis inducing ligand (TRAIL)⁵⁰ to areas of rheumatoid arthritis in a rat model. TRAIL has shown promise as an anti-tumour agent and as a therapeutic agent for rheumatoid arthritis and is being evaluated in Phase I and II trials. TRAIL activates overexpressed death receptor pathways on cancer cells inducing apoptosis without affecting healthy cells. If delivered intravenously TRAIL has a very short half-life due to proteolytic degradation but nano-sized complexes of HA and TRAIL of diameters of 200–1200 nm were stable for 5 days in the rat compared with a half-life of less than 30 min for TRAIL alone. HA–TRAIL nanocomplexes can be delivered by subcutaneous injection without side effects and initial results suggest therapeutic potential.

An alternative method of protecting proteindrugs against degradation is by the chemical attachment of polyethylene glycol which shields the surface of globular proteins from proteolytic enzymes. However, this reduces the biological activity of proteins which is avoided by the use of HA making it a more promising alternative.

HA synthesis and uptake in breast tissue increases accompanying breast cancer progression and metastasis⁵¹ and breast cancer cells overexpress CD44 and proliferation is dependent on CD44-mediated uptake of HA. These characteristics make breast cancer a suitable target for HA directed drug delivery. Cai et al.⁵² developed a cisplatin–HA conjugate using AgNO₃ as an activating agent. The drug was chemically linked to the carboxyl moiety of D-glucuronic acid with a drug : HA ratio of 1 : 4 and in an animal model of breast cancer had significant anti-tumour activity. A further advantage of this mode of delivery is that it allows the highly toxic cis-platinum to be delivered to a specific site, reducing the side effects associated with other forms of delivery.

A growing application of HA is as a macromolecular carrier of cytotoxic drugs to cancer cells. Most anti-cancer drugs distribute throughout the body and are toxic to healthy dividing cells, as well as neoplastic cells. Patients with colorectal cancer who had stopped responding to 5-fluorouracil were treated with a complex of HA and the anti-cancer drug irinotecan. A minority of patients (17%) had a partial response and in 50% of patients the disease stabilised indicating combination with HA had not affected efficacy.⁵³HA contains several functional groups including the glucuronic acid carboxylate group and the hydroxyl moiety on N-acetylglucosamine to which drugs can be chemically linked. On linking the drug to one of these potential binding sites a pro-drug is created which becomes active on release at the site of delivery. Platt and Szoka⁵⁴ give several examples of this procedure including using hydrazide to link the glucuronate COO⁻ group on low molecular weight HA to the drug paclitaxel and linking the drugcarboranevia an ester link to the same site on high molecular weight HA. In both cases the pro-drug was taken up through CD44 expressed on the tumour and exhibited cytotoxicity.

Because HApolymers appear to be an effective drug delivery system Hoare et al.⁵⁵ have undertaken a study of their rheological properties. They used different mixtures of HA and hydroxypropylmethyl cellulose (HPMC) to optimise drug delivery and drug release kinetics using bupivacaine as a model. They found the carrier had excellent biocompatibility and water binding properties and the kinetics of drug release could be adjusted according to the ratio of the two polymers. This suggests a potential for delivery of a range of drugs.

An example of increasing the effectiveness of drug delivery comes from the application of an anti-tubercular drug. Tuberculosis (TB) affects 1.8 billion people worldwide and multi-drug resistant strains of the causative bacillus are contributing to increased incidence of disease. Pulmonary TB is characterised by the presence of bacilli in pulmonary macrophages where they are inaccessible to anti-tubercular therapy. Spheres with a diameter of 2–5 μm were prepared by co-spraying oflloxacin with Na⁺hyaluronate in a spray drier and contained 50.0 ± 2.5% by weight of the drug. The microspheres were delivered more favourably to the lung via an aerator and had greater efficacy than other routes of administration.⁵⁶

Since HA synthesis and metabolism is a feature of ischaemic stroke and angiogenesis is an indicator of survival, manipulation of HAmetabolism in stroke patients may be beneficial. One possible approach is to use o-HA linked gold nanoparticles to deliver angiogenic o-HA to stroke areas. Preliminary studies have shown an ability of gold nanoparticles of 15–50 nm diameter coated with albumin to cross the blood–brain barrier.^57,58 To enhance the targeting of the nanoparticles to ECs, an antibody to a protein found only on EC can be incorporated into the particle. In particular the cell determinant CD31 found on ECs is of interest. We are actively pursuing some of these ideas with industrial collaborators.

HA in tissue engineering and repair

Hydrogels, crosslinked hydrophilic polymers, have been used as scaffolds to allow tissue repair or regeneration at sites of injury. They may absorb up to 1000× their dry weight in water giving them characteristics resembling tissue. Hydrogel introduction into the body requires surgical implant, producing irritation and tissue damage, prompting a search for an injectable hydrogel. Ideally this would be non-toxic, and should be degraded by tissue enzymes after repair is completed. Hydrogels have the capacity to form into the desired shape at the injury site and to adhere to tissue during gel formation, strengthening the tissue–hydrogel interface. HA is easy to produce and modify, hydrophilic and naturally biodegradable and so a suitable material. Kurisawa et al.⁵⁹ synthesised a hydrogel of a HA–tyramine complex using horseradish peroxidase in the presence of H₂O₂. The reaction, an oxidative coupling which linked nine tyramine residues per 100 disaccharide units of HA, can be carried out in situ using two syringes: one containing the enzyme and the other the gel -forming components. Gels have also been used to deliver cells to sites of injury. Tan et al.⁶⁰ created a complex of HA and water soluble N-succinyl-chitosan which linked through a Schiff base reaction. The gel could encapsulate viable chondrocytes with potential for delivery to sites of injury. An injectable cell-containing hydrogel that supports cell proliferation⁶¹ has been derived from thioylated HA coupled to polyethylene glycol (PEG3400). The materials crosslinked forming a network in which fibroblasts could survive and proliferate. The gels were implanted into the flanks of nude mice and secreted extracellular matrix, confirming their potential for tissue engineering. A photopolymerisable scaffold has been developed incorporating glycidyl methacrylate–HA (GMHA) conjugates which were polymerised on exposure to ultraviolet radiation to form crosslinked GMHA hydrogels. The hydrogels supported the growth of aortic endothelial cells in a subcutaneous rat model. A recent development has seen the incorporation of single walled carbon nanotubules in HA solutions and subsequent induction of reinforced crosslinking with divinyl sulfone. The resulting structure had no loss in water retention compared to HAgels but had increased mechanical strength.⁶²

Use of anti-CD44 antibodies

To target the delivery of anti-cancer drugs to tumour cells, antibody–drug conjugates have been developed. In particular anti-CD44 antibodies have shown promise in clinical trials and their use is predicated on overexpression of CD44 in the target tumours. A radiolabelled anti-CD44 antibody has been used in clinical trials to target head and neck carcinoma and early stage breast cancer.^63,64 The antibody was conjugated to Tc-99m or Re-186 using a linking agent 5-benzoyl-mercaptoacetylglycine. Mertansine linked to the same antibody has been used against head and neck and metastatic breast cancer (reviewed in ref. 54). Solid tumours have a less developed blood supply than healthy tissue with an increased number of leaky capillaries. This means that particles of <500 nm diameter can leak into tumour cells from the circulation. Tumours also have an underdeveloped lymph system which reduces clearance of the particles from the tumour.

It is interesting that the anti-CD44 antibody, as well as targeting drugs to the cancer cells also disrupts CD44 matrix interactions and by binding to CD44 activates signalling pathways inducing apoptosis.

Hyaluronan–metal nanoparticles

The development of nanoparticles with high specificity is of interest in medical diagnosis and treatment⁵⁴ (Fig. 3). Inorganic nanoparticles provide a large surface area for attachment and in the case of Au and Ag have photothermal properties which can be exploited for drug release. Gold and silver nanoparticles stabilised with hyaluronan have been synthesised by a number of groups. Kemp et al.⁶⁵ added HAuCl₄ to a boiling solution of HA which acted as a reducing agent to link to and stabilise the metal. The solution was further heated to generate the gold nanoparticles. Ag nanoparticles were prepared in a similar way but required extended heating and both particles had a size distribution between 5 and 30 nm and inhibited oedema in a rat injury model. A gold nanoprobe has been synthesised for the detection of hyaluronidase in tissue.⁶⁶ The probe was produced by incorporating near-infrared fluorescence dye labelled HA onto the surface of gold nanoparticles. When the immobilised HA was cleaved by hyaluronidase strong fluorescence was detected. In animal models of metastatic tumour expressing hyaluronidase tumour sites were clearly identified. An alternative delivery system has been produced using Fe₂O₃ particles incorporated into HA layers with a diameter of less than 160 nm.⁶⁷ The particles were able to deliver drugs effectively to epithelial cell lines.

Fig. 3 HA is synthesised on the cell surface and extruded to the ECM and can interact with cell surface receptors. HA microparticles can attach to HA receptors and be internalised with any attached bioactive material. Adapted from a diagram in ref. 54.

Conclusion

HA is a widely spread, biologically active glycosaminoglycan which regulates cellular function through interaction with receptors. The perturbation of HAmetabolism following injury and in neoplastic disease can be used in drug targeting and delivery. HA is an emerging vehicle for drug delivery because of the ability to target it to sites of overexpression of CD44 and it will not induce an immune response. Future work will allow more efficient incorporation of drugs into HAnanoparticles and the production of hyaluronannanoparticles as biosensors .

References

1. B. P. Toole and M. G. Slomiany, Semin. Cancer Biol., 2008, 18, 244–250.
2. A. J. Day and G. D. Prestwich, J. Biol. Chem., 2002, 277, 4585–4588.
3. R. Stern, Glycobiology, 2003, 13, 105R–115R.
4. B. P. Toole, Nat. Rev. Cancer, 2004, 4, 528–539.
5. R. K. Reed, K. Lilja and T. C. Laurent, Acta Physiol. Scand., 1988, 134, 405–411.
6. J. R. Fraser, T. C. Laurent and U. B. Laurent, J. Intern. Med., 1997, 242, 27–33.
7. P. H. Weigel, V. C. Hascall and M. Tammi, J. Biol. Chem., 1997, 22, 13997–14000.
8. P. Prehm, Biochem. J., 1983, 211, 191–198.
9. T. C. Laurent and J. R. Fraser, FASEB J., 1992, 6, 2397–2404.
10. P. H. Weigel and P. L. DeAngelis, J. Biol. Chem., 2007, 282, 36777–36781.
11. R. Stern, Pathol. Biol. (Paris), 2005, 53, 372–382.
12. S. P. Evanko, M. I. Tammi, R. H. Tammi and T. N. Wight, Adv. Drug Delivery Rev., 2007, 59, 1351–1365.
13. L. Roden, P. Campbell, J. R. Fraser, T. C. Laurent, H. Pertoft and J. N. Thompson, CIBA Foundation Symp., 1989, 143, 60–86.
14. S. K. Rai, F. M. Duh, V. Vigdorovich, A. Danilkovitch-Miagkova, M. I. Lerman and A. D. Miller, Proc. Natl. Acad. Sci. U. S. A., 2001, 98, 4443–4448.
15. A. B. Csoka, G. I. Frost and R. Stern, Matrix Biol., 2001, 20, 499–508.
16. A. J. Day, Biochem. Soc. Trans., 1999, 27, 115–121.
17. J. Bajorath, B. Greenfield, S. B. Munro, A. J. Day and A. Aruffo, J. Biol. Chem., 1998, 273, 338–343.
18. E. Pure and R. K. Assoian, Cell Signal., 2009, 21, 651–655.
19. J. Lesley, V. C. Hascall, M. Tammi and R. Hyman, J. Cell Biol., 2000, 275, 26967–26975.
20. S. Matou, J. Gaffney, S. Kumar, J. Krupinski and M. Slevin, Int. J. Oncol., 2009, 35, 761–773.
21. C. M. Milner and A. J. Day, J. Cell Sci., 2003, 116, 1863–1873.
22. P. Rooney, S. Kumar, J. Ponting and M. Wang, Int. J. Cancer, 1995, 60, 632–666.
23. C. M. Rao and K. Datta, Biochem. Mol. Biol. Int., 1996, 40, 327–337.
24. M. Slevin, J. Krupinski, J. Gaffney, S. Matou, D. West, H. Delisser, R. C. Savini and S. Kumar, Matrix Biol., 2007, 26, 58–68.
25. R. J. Gilbert, R. J. McKeon, A. Darr, A. Calabro, V. C. Hascall and R. V. Bellamkonda, Mol. Cell Neurosci., 2005, 29, 545–558.
26. G. Haddock, A. K. Cross, S. Allan, B. Sharrack, J. Callaghan, R. A. Bunning, D. J. Buttle and M. N. Woodroofe, Biochem. Soc. Trans., 2007, 35, 692–694.
27. L. S. Sherman, J. N. Struve, R. Rangwala, N. M. Wallingford, T. M. Tuohy and C. Kuntz, Glia, 2002, 38, 93–102.
28. C. M. Galtrey, J. C. Kwok, D. Carulli, K. E. Rhodes and J. W. Fawcett, Eur. J. Neurosci., 2008, 27, 1373–1390.
29. X. Wang, L. Xu, H. Wang, Y. Zhan, E. Pure and G. Feurstein, J. Neurochem., 2002, 83, 1172–1179.
30. H. Wang, Y. Zhan, L. Xu, G. Feuerstein and X. Wang, Stroke, 2001, 32, 1020–1027.
31. A. Al’Qteishat, J. Gaffney, J. Krupinski, F. Rubio, D. West, S. Kumar, P. Kumar, N. Mitsios and M. Slevin, Brain, 2006, 129, 2158–2176.
32. G. Lepperdinger, J. Muellegger and G. Kreill, Matrix Biol., 2001, 20, 509–514.
33. J. Struve, P. C. Maher, S. Kinney, M. G. Fehlings, C. Kuntz and L. S. Sherman, Glia, 2005, 52, 16–24.
34. W. S. Kang, J. S. Choi, Y. J. Shin, H. Y. Kim, J. H. Cha, J. Y. Lee, M. H. Chun and M. Y. Lee, Brain Res., 2008, 1228, 208–216.
35. L. L. Jones, Z. Liu, J. Shen, A. Werner, G. W. Kreutzberger and G. Ravich, J. Comp. Neurol., 2000, 426, 468–492.
36. L. L. Jones, G. W. Kreutzberger and G. Raivich, Eur. J. Neurosci., 1997, 9, 1854–1863.
37. A. Al’Qteishat, J. Gaffney, J. Krupinski and M. Slevin, NeuroReport, 2006, 17, 1111–1114.
38. L. C. Foster, P. Wiesel, G. S. Huggins, R. Panares, M. T. Chin, A. Pellacani and M. A. Perrella, FASEB J., 2000, 14, 368–378.
39. A. D. Theocharis, D. H. Vynios, N. Papageorgakopoulou, S. S. Skandalis and D. A. Theocharis, Int. J. Biochem. Cell Biol., 2003, 35, 376–390.
40. M. A. Antilla, R. H. Tammi, M. I. Tammi, K. J. Syrjanen, S. V. Saarikoski and V. M. Kosma, Cancer Res., 2000, 60, 150–155.
41. B. P. Toole and V. C. Hascall, Am. J. Pathol., 2002, 161, 745–747.
42. D. Hamerman, G. J. Todaro and H. Green, Biochem. Biophys. Acta, 1965, 101, 343–351.
43. N. Itano and K. Kimata, Semin. Cancer Biol., 2008, 18, 268–274.
44. H. Li, L. Guo, J. W. Li, N. Liu, R. Qi and J. Liu, Int. J. Cancer, 2000, 17, 927–932.
45. C. Forster-Hovarth, L. Meszaros, E. Raso, B. Dome, A. Ladanyi, M. Morini, A. Albini and J. Timar, Microvasc. Res., 2004, 68, 110–118.
46. S. Ghatak, S. Misra and B. P. Toole, J. Biol. Chem., 2002, 277, 38013–38020.
47. R. I. Cordo Russo, M. G. Garcia, L. Alaniz, G. Blanco, E. Alvarez and S. E. Hajos, Int. J. Cancer, 2008, 122, 1012–1018.
48. Y. Li and P. Heldin, Br. J. Cancer, 2001, 85, 600–607.
49. B. Enegd, J. A. King, S. Stylii, I. Paradiso, A. H. Kaye and U. Novak, Neurosurgery, 2002, 50, 1311–1318.
50. S. J. Na, S. Y. Chae, S. Lee, K. Park, K. Kim, J. H. Park, I. c. Kwon, S. Y. Jeong and K. C. Lee, Int. J. Pharm., 2008, 363, 149–154.
51. M. Gotte and G. W. Yip, Cancer Res., 2006, 66, 10233–10239.
52. S. Cai, Y. Xie, T. R. Bagby, M. S. Cohen and M. L. Forest, J. Surg. Res., 2008, 147, 247–252.
53. P. Gibbs, T. J. Brown, R. Ng, R. Jennens, E. Cinc, M. Pho, M. Michael and R. M. Fox, Chemotherapy, 2009, 55, 49–59.
54. V. M. Platt and F. C. Szoka Jr., Mol. Pharm., 2008, 5, 474–486.
55. T. Hoare, D. Zurakowski, R. Langer and D. S. Kohane, J. Biomed. Mater. Res., 2009 , epub ahead of print.
56. S. M. Hwang, D. D. Kim, S. J. Chung and C. K. Shim, J. Controlled Release, 2008, 129, 100–106.
57. J. Lopez-Viota, S. Mandal, A. V. Delgado, J. L. Toca-Herrara, M. Moller, F. Zanuttin, M. Balestrano and S. Krol, J. Colloid Interface Sci., 2009, 332, 215–223.
58. G. Sonavane, K. Tomoda and K. Makino, Colloids Surf., B, 2008, 66, 274–280.
59. M. Kurisawa, J. E. Chung, Y. Y. Yang, S. J. Gao and H. Uyama, Chem. Commun., 2005, 4312–4314.
60. H. Tan, C. R. Chu, K. A. Payne and K. G. Marra, Biomaterials, 2009, 30, 2499–2506.
61. X. Z. Shu, Y. Liu, F. Palumbo, Y. Luo and G. D. Prestwich, Biomaterials, 2004, 25, 1339–1348.
62. J. B. Leach, K. A. Bivens, C. N. Collins and C. E. Schmidt, J. Biomed. Mater. Res., 2004, 70, 74–82.
63. P. K. Borjesson, E. J. Postema, J. C. Roos, D. R. Colnot, H. A. Marres, M. H. van Schie, G. Stehle, R. de Bree, G. B. Snow, W. J. Oyen and G. A. van Dongen, Clin. Cancer Res., 2003, 9, 3961S–3872S.
64. D. R. Colnot, J. C. Roos, R. de Bree, W. J. Wilhelm, J. A. Kummer, G. Hanft, K. H. Heider, G. Stehle, G. B. Snow and G. A. van Dongen, Cancer Immunol. Immunother., 2003, 52, 576–582.
65. M. M. Kemp, A. Kumar, S. Mousa, T.-J. Park, P. Ajayan, N. Kubotera, S. A. Mousa and R. J. Linhardt, Biomacromolecules, 2009, 10, 589–595.
66. H. Lee, K. Lee, I. K. Kim and T. G. Park, Biomaterials, 2008, 29, 4709–4718.
67. A. Kumar, B. Sahoo, A. Montpetit, S. Behera, R. F. Jockey and S. S. Mohapatra, Nanomed.: Nanotechnol., Biol. Med., 2007, 3, 132–137.

---