Research Article: Gene Expression Profiling of Peri-Implant Healing of PLGA-Li+ Implants Suggests an Activated Wnt Signaling Pathway In Vivo

Date Published: July 21, 2014

Publisher: Public Library of Science

Author(s): Anna Thorfve, Anna Bergstrand, Karin Ekström, Anders Lindahl, Peter Thomsen, Anette Larsson, Pentti Tengvall, Xing-Ming Shi.


Bone development and regeneration is associated with the Wnt signaling pathway that, according to literature, can be modulated by lithium ions (Li+). The aim of this study was to evaluate the gene expression profile during peri-implant healing of poly(lactic-co-glycolic acid) (PLGA) implants with incorporated Li+, while PLGA without Li+ was used as control, and a special attention was then paid to the Wnt signaling pathway. The implants were inserted in rat tibia for 7 or 28 days and the gene expression profile was investigated using a genome-wide microarray analysis. The results were verified by qPCR and immunohistochemistry. Histomorphometry was used to evaluate the possible effect of Li+ on bone regeneration. The microarray analysis revealed a large number of significantly differentially regulated genes over time within the two implant groups. The Wnt signaling pathway was significantly affected by Li+, with approximately 34% of all Wnt-related markers regulated over time, compared to 22% for non-Li+ containing (control; Ctrl) implants. Functional cluster analysis indicated skeletal system morphogenesis, cartilage development and condensation as related to Li+. The downstream Wnt target gene, FOSL1, and the extracellular protein-encoding gene, ASPN, were significantly upregulated by Li+ compared with Ctrl. The presence of β-catenin, FOSL1 and ASPN positive cells was confirmed around implants of both groups. Interestingly, a significantly reduced bone area was observed over time around both implant groups. The presence of periostin and calcitonin receptor-positive cells was observed at both time points. This study is to the best of the authors’ knowledge the first report evaluating the effect of a local release of Li+ from PLGA at the fracture site. The present study shows that during the current time frame and with the present dose of Li+ in PLGA implants, Li+ is not an enhancer of early bone growth, although it affects the Wnt signaling pathway.

Partial Text

Orthopedic and dental implant therapies have evolved into important treatments for deranged joints and lost teeth or to provide fixation of bone in the case of fractures. Osteogenesis, i.e. the differentiation of mesenchymal stem cells (MSCs) into mature osteoblasts is essential in bone growth, fracture healing and osseointegration. The hallmarks of osteogenesis around implants are increased alkaline phosphatase (ALP) activity and the formation of a calcium-rich mineralized extracellular matrix (ECM). This contains bone-related proteins, such as type I collagen (COL1), osteocalcin (OCN), bone sialoprotein (BSP) and osteopontin (OPN)[1]. The runt-related transcription factor 2 (RUNX2) is indicated as the master switch in osteogenesis, although other factors, such as the canonical Wnt signaling pathway are pivotal for the guidance of MSCs into the osteoblastic lineage and bone homeostasis[2], [3]. The canonical Wnt signaling pathway involves several Wnt proteins that, upon activation, bind to receptors frizzled (FZD) and co-receptor LDL-related proteins 5/6 (LRP5/6). This interaction initiates a signaling cascade that leads to the inactivation of the β-catenin (CTNNB1) degradation complex, consisting of AXIN, adenomatous polyposis coli (APC), casein kinase I (CK1) and glycogen synthase kinase 3b (GSK-3β). The intracellular β-catenin concentrations then become stabilized, translocated into the nucleus and activate the transcription of downstream canonical Wnt-related genes. The non-canonical pathways, which function independently of β-catenin, are less well studied but attract increasing interest. Recent studies suggests the FZD9 receptor as a positive regulator of both intramembranous and endochondral ossification during fracture healing via the non-canonical pathways[4], [5], although the canonical Wnt signaling pathway is implicated as the dominant mechanism in bone biology. The critical role of Wnt signaling is well recognized, not only during embryonic bone formation, post-natal bone homeostasis and regeneration but also for osseointegration of implants[6]–[8]. Of several ways to modulate the Wnt signaling pathway, lithium ions (Li+) are often regarded as the simplest activator[9], [10]. Oral Li+ treatment, widely used as stabilizer in bipolar and depressive disorders, is reported to activate the canonical Wnt signaling pathway via the inhibition of the β-catenin degradation enzyme, GSK-3β[11], [12]. Previous studies indicated that Li+ induces increased bone formation and bone mass in mice, as well as reduce the risk of fractures in patients on Li+ treatment[13], [14]. However, due to the fact that Li+ has a narrow therapeutic index and the therapy is associated with multiple side-effects[11], it would be beneficial in biomaterial applications to enhance bone regeneration via a local release of Li+ from bioactive degradable materials. The objective of the present study was to incorporate Li+ into poly(lactic-co-glycolic acid) (PLGA) implants, to monitor the local release of Li+ and to evaluate the local biological effects by gene expression, immunohistochemistry, histology and histomorphometry in a rat tibia model. PLGA is a biodegradable co-polymer that is widely used in pharmaceutics as a controlled drug delivery system[15] and it has evolved into a frequently used synthetic polymer within the field of bone regeneration[16]. One reason for its widespread use is its biodegradability and biocompatibility. In the presence of water, PLGA degrades via hydrolysis into lactic and glycolic acids, natural compounds that are metabolized and excreted as carbon dioxide and water[16], [17]. PLGA is in clinical use since decades and its biosafety is proven in many medical applications[16], [18]. It has been used as a carrier for the delivery of osteogenic factors for cell adhesion, differentiation and improved bone regeneration, as pure PLGA implants or in combination with hydroxyapatite (HA) to obtain an improved mechanical strength[19]–[23]. To study the effect of Li+, the ions have previously been incorporated into various materials or added directly to cell culture media[24]–[27]. However, there is no previous study using Li+-PLGA implants with the aim to modulate the Wnt signaling pathway in vivo. In order to fully investigate the underlying cellular and molecular mechanisms of peri-implant healing within this context, we performed a genome-wide microarray analysis, followed by validation of selected results by qPCR, in combination with a histomorphometric evaluation. Previous works used gene expression profiling during in vivo bone healing, with or without implants[28]–[32], but this is to the best of the authors’ knowledge, the first in vivo study to evaluate the bone healing aspects in the vicinity of Li+-containing PLGA implants. We were able to show that the present dose of Li+ activates the Wnt signaling pathway but is not an inducer of early bone growth. In addition to providing insights into Wnt signaling during peri-implant healing around bone-anchored implants, this study shows that a local release of Li+ at the fracture site can be used to modulate bone cell signaling but needs further optimization in order to induce early bone growth.

The inclusion of Li+ in PLGA was obtained by a hot-melt procedure, resulting in the incorporation of 10% (w/w) Li2CO3/implant. The SEM and TOF-SIMS imaging data demonstrated how the polymer implant transformed in appearance from a dense to a porous morphology with time. The increased porosity was initiated at the surface of the shaft and spread towards the center. The recognized degradation mechanisms of biodegradable polymers are surface or bulk erosions. Bulk erosion is the dominant mechanism, as water penetrates into the matrix and the degradation occurs in the inner parts of the matrix, leading to an equally distributed porosity. For matrices with dimensions below 7 cm, like our implants, the suggested degradation mechanism is bulk erosion[38]. However, our experimental findings, showing increased porosity close to the surface, indicate that surface erosion contributed largely to the erosion of the device. Furthermore, the in vitro Li+ release rate was almost constant over time, indicating that the release mechanism follows Case II, i.e. water penetration was the rate-limiting factor and not the drug diffusion out from the device. Inside the PLGA matrix, the highly water-soluble Li2CO3 was distributed in the shape of particles. When the water front reached the Li2CO3 particles, these dissolved rapidly and Li+ diffused out from the matrix, after which the location of the former Li2CO3 particles turned into water-filled pores. The rate-limiting step for Li+ release was therefore the water penetration rate. Simultaneously, the pore formation observed in SEM and TOF-SIMS was also controlled by water penetration. After 28 days of exposure to buffer, about 50% of the Li+-salt content was released, and even though highly perforated the implants still held together. In order to obtain a complete PLGA degradation or a larger amount of released salt, the formulation has to be further optimized. The mode of degradation in bone is also expected to be different from that of a fully submerged implant in PBS buffer, because of the uniqueness of the bone environment. However, assuming an equal in vitro and in vivo release rate, this means that during 28 days, about 2.3 10−5 mol Li+ was released to the bone environment around the implants.

In the present study, Li+ was successfully released from PLGA implants out to surrounding bone in a sustained release fashion and the gene expression profile during peri-implant healing in a rat tibia model was investigated. Microarray analysis revealed a large number of significantly differentially regulated genes within the 2 implant groups (PLGA with/without Li+) over time. The annotation cluster analysis demonstrated, for both implant groups, skeletal system morphogenesis/development and, surprisingly, the Li+-containing implants were also related to cartilage development and condensation as well as to the Wnt signaling pathway. Li+ appears not to be a strong inducer of early bone growth around PLGA implants although the qPCR analysis demonstrated that Li+ activated the Wnt signaling pathway at 7 days post-surgery, as demonstrated by a significant increase in FOSL1 expression. A possible impact on the Wnt signaling pathway via ASPN expression was observed at 28 days post-surgery. The decrease in bone area that was observed around both implant groups may to some extent be explained by the presence of multinucleated cells, possibly caused by the polymeric degradation products. The collected data indicate that Li+ is a mild bone growth modulator in the context of bone-anchored implants and the Wnt signaling pathway when administered locally in the present dose and during the time frame of the study, up to 28 days.