Abstract
The research on the development of new biomaterials that promote bone tissue regeneration is receiving great interest by the biomedical scientific community. Recent advances in nanotechnology have allowed the design of materials with nanostructure similar to that of natural bone. These materials can promote new bone formation by inducing the formation of nanocrystalline apatites analogous to the mineral phase of natural bone onto their surfaces, i.e. they are bioactive. They also stimulate osteoblast proliferation and differentiation and, therefore, accelerate the healing processes. Silica-based ordered mesoporous materials are excellent candidates to be used as third generation bioceramics that enable the adsorption and local control release of biological active agents that promote bone regeneration. This local delivery capability together with the bioactive behavior of mesoporous silicas opens up promising expectations in the bioclinical field. In this review, the last advances in nanochemistry aimed at designing and tailoring the chemical and textural properties of mesoporous silicas for biomedical applications are described. The recent developed strategies to synthesize bioactive glasses with ordered mesopore arrangements are also summarized. Finally, a deep discussion about the influence of the textural parameters and organic modification of mesoporous silicas on molecules adsorption and controlled release is performed.
1. Introduction
Currently, the biomedical scientific community is demanding new advances in the design of third generation bioceramics for bone tissue regeneration technologies [1]. The main goal consists in tailoring bone grafts that mimic as much as possible natural bone tissue biomineralization processes. This involves the growth of nanocrystalline carbonate hydroxyapatites similar to the biological ones onto the materials surfaces, that is, bioactive behavior. Moreover, these biomaterials should stimulate osteoblasts adhesion, proliferation, and differentiation to reduce the postoperative periods [1–6].
Natural hard tissues in vertebrates are natural composite materials, composed of an organic matrix and an array of inorganic nanoapatites. The inorganic phase consists of nanocrystalline apatites with average length of 50 nm, width of 25 nm, and thickness of 2–5 nm [7]. Such inorganic phase is located into an organic matrix mainly consisting of type I collagen [8, 9]. The type I collagen molecules are bonded forming linear chains that are in turn arranged in fibers, giving rise to various macroscopic structures and leaving small empty interstitial compartments where needle-like apatite nanocrystals are deposited during a controled biomineralization process. The integration of both organic and inorganic phases at the nanometric scale modulates the mechanical properties of each kind of bone. The physiological formation mechanism of biological nanocrystalline nanoapatites proposed by Brown [10, 11] involves the growth of biological apatites from two metastables intermediates. Thus, according to the model proposed by Brown for in vivo conditions, the first compound formed onto the bioactive material surface consists in amorphous calcium phosphate which is subsequently transformed into octacalcium phosphate and, finally, the hydrolysis step leads to the formation of a nanocrystalline calcium deficient carbonate hydroxyapatite [12, 13]. Bone is a dynamic tissue that is constantly being formed and resorbed and this remodeling process is mainly governed by osteoblasts and osteoclasts cells [14, 15]. Osteoblasts govern bone formation processes whereas osteoclasts modulate bone resorption phases and both processes are controled by mechanical and hormonal stimuli. Skeletal tissues display different arrangement and porosity to maintain bone performance, including mechanical, biological, and chemical functions, such as mechanical support, protection of essential body organs, and storage of mineral ions, mostly calcium and phosphate. Bone has a hierarchical organization at many levels of scale which is essential to allow an appropriate physiological performance [16, 17]. Therefore, pore sizes in cortical bone range from 1 m to 100 m, being 200 m to 400 m in trabecular bone. Moreover, trabecular bone shows 50–95% of porosity, where 55–70% of such pores are interconnected.
The efforts committed to develop bioceramics that could promote new bone formation began in 1971 when Hench reported on the first bioactive composition [18]. Bioactive materials exhibit highly reactive surfaces that are able to bond to living tissues through the formation of nanocrystalline calcium deficient carbonate hydroxyapatites similar to the biological ones. The newly formed apatite, together with the adsorption of biological moieties, constitutes an excellent layer for the action of osteoblasts and the subsequent bone-implant fixation. During the last decades, the development of bioactive materials has allowed important advances in the fields of bone and dental grafting for hard tissue replacement [19–21]. However, the new trends go further than just ensuring the implant-bone fixation. Therefore, bone tissue engineering is a new interdisciplinary research field that emerged because of the excellent potential of certain materials to repair diseased or damaged tissue via regeneration rather than replacement. Such interdisciplinary research area is mainly based on the design of 3D-scaffolds made of biocompatible and bioactive materials that incorporate biologically active agents (drugs, cell, gens, and proteins) to stimulate bone tissue regeneration [22, 23]. 3D-scaffolds will play a significant role in bone tissue regeneration by preserving tissue volume, providing temporary mechanical function, locally delivering biologically active agents, and allowing regenerated tissues to assume their function as scaffold degrades [24]. Thus, one of the main challenges for the scientific community is the development of scaffolds with 3D interconnected porosity and with pores similar in size and number to those existing in natural bone. The size range of these pores is quite wide (1–1000 m) and at the moment there is a lack of consensus regarding the ideal porosity necessary for bone in growth [25].
Up to date, several porous bioceramics, such as calcium phosphates, glasses, cements, or different biphasic combinations exhibiting open texture and high surface area and porosity, have been reported as excellent candidates to be used as local carriers for local controled delivery systems [1, 26, 27]. Recently, supramolecular chemistry has allowed the emergence of a new generation of advanced bioceramics which exhibit fascinating properties for regenerative purposes. Among such bioceramics, ordered mesoporous materials show potential features to be used as starting materials for the subsequent design and fabrication of scaffolds for bone tissue engineering purposes. These materials exhibit enhanced bioactive capability, and a very precise control over loading and controled release of biologically active agents can be achieved (see Figure 1). This review deals with the recent nanotechnological advances aimed at tailoring the structural, textural, and chemical properties of nanostructured ordered mesoporous silicas with biomedical purposes. To reach this goal, two main aspects concerning mesoporous silicas will be tackled: (i) bioactive behavior and (ii) capability of loading and releasing biologically active molecules in a sustained fashion. Such properties will be described and critically analyzed on the basis of their textural and structural properties and on their composition and the possibility of undergoing organic modification.
2. Silica-Based Ordered Mesoporous Materials
Since the discovery of silica-based ordered mesoporous materials in the early 1990s by scientists of the Mobil Corporation [28] and Waseda University [29], these materials have attracted enormous interest. Such nanostructured mesoporous materials have been widely employed in catalysis as host matrices of several guest molecules. Recently, the potential applications of mesoporous silicas as molecule carriers have been expanded into the biomedical field for bone tissue regeneration purposes [3].
Silica-based mesoporous materials constitute a new generation of materials that show ordered arrangements of channels and cavities of different geometry built up from units [30]. These materials exhibit variable pore sizes (2–50 nm), high surface areas (ca.), high pore volumes (ca.), and homogeneous nanostructures, which can be tailored by varying the synthesis procedure [31]. The synthesis of silica-based ordered mesoporous materials is based on the formation of liquid crystalline mesophases of amphiphilic molecules (surfactant) that serve as template for the in situ polymerization of orthosilicic acid [32]. Well-known examples are MCM-41 and MCM-48 materials, which are synthesized by using cationic surfactants under basic conditions. These mesoporous matrices exhibit pore sizes of ca. 2.5 nm and different pore arrangements: 2D-hexagonal (p6mm plain group) for MCM-41 and 3D-bicontinuos cubic (Ia–3d space group) for MCM-48 [32, 33]. On the contrary, the employment of nonionic surfactants and acid conditions lead to an increase in the pore wall thickness and to an enlargement of the pore size (ca. 10 nm) of the resulting mesoporous matrices [34]. SBA-15 [35] and FDU-5 [36] materials constitute two representative examples of such mesoporous materials. SBA-15 exhibits 2D-hexagonal structure with micropores interconnecting adjacent mesopores and FDU-5 presents a 3D-bicontinuos cubic structure similar to that of MCM-48.
It was in 2001, owing to their outstanding adsorptive properties, when mesoporous silicas were proposed for the first time as drug delivery system and ibuprofen, a common anti-inflammatory, was loaded into the pore channels of MCM-41 and subsequently released in a sustained manner [37]. Since then, this research field has experienced a significant growth and much effort has been devoted to tailor the nanostructure and textural properties (i.e., pore diameter, surface area, and pore volume) of mesoporous materials that allow to achieve a better control over molecule loading and controled release. All the parameters that govern the adsorption and release kinetics of biologically active molecules, with special emphasis on the organic functionalization of the silica walls, will be tackled in this review.
Moreover, the bioactive behavior of silica-based ordered mesoporous materials is undoubtedly an added value when considering these materials for bone tissue regeneration technologies. The combination of both properties, that is, bioactivity and controled delivery capability, is a remarkable synergy that makes mesoporous silicas excellent candidates to be used as starting materials for the manufacture of 3D-scaffolds for bone tissue engineering [38]. The different chemical strategies developed to accelerate the bioactive response of silica-based ordered mesoporous materials will be also discussed in this manuscript.
2.1. Bioactivity of Silica-Based Ordered Mesoporous Materials
Silica-based mesoporous materials are characterized by exhibiting high surface areas and high surface density of silanol groups (Si–OH). It is well known that silanol groups, which are present in conventional bioactive glasses, are able to react with physiological fluids to produce nanometer-sized carbonated apatites [21, 39]. Therefore, nanostructured silica-based ordered mesoporous materials, which present high amounts of silanol groups covering their surfaces, are expected to induce bioactive responses. It was in 2006 when several in vitro bioactivity assays were performed by soaking three mesoporous materials, SBA-15, MCM-48, and MCM-41, into simulated body fluid (SBF) [40]. The obtained results revealed that SBA-15 and MCM-48 developed a nanocrystalline carbonate hydroxyapatite layer onto their surfaces after 30 and 60 days, respectively. On the contrary, no evidence of formation of an apatite-like layer was observed onto the MCM-41 surface. The different bioactive behavior of SBA-15, MCM-48, and MCM-41 matrices is related to the concentration of silanol groups, and also to the textural and structural properties of the mesoporous matrices. In this case, MCM-41 showed lower concentration of silanol groups (c.a.) than SBA-15 and MCM-48 (c.a.), which could explain the absence of bioactivity in MCM-41 after 60 days. In addition, the textural and structural properties also contribute to the bioactive behavior, since large and accessible pores systems favor the ionic diffusion into the mesoporous framework and consequently the apatite formation is accelerated. Hence, SBA-15, which shows pore size around 9 nm and 3D-pores system with interconnected microporosity, and MCM-48, with 3D-cubic pores system exhibited bioactive response. On the contrary, MCM-41, which presents 2D-pores system with pore sizes close to 2.5 nm did not show bioactive behavior after 60 days of assay. In fact, traditional pure silica mesoporous materials such as SBA-15 and MCM-48 showed bioactive behavior, but the rate of formation of the hydroxyapatite layer is too slow (30–60 days) compared to that of conventional sol-gel bioactive glasses (3 days). Figure 2 displays the bioactivity studies using Fourier transform infrared spectroscopy (FTIR) and scanning electron microscopy-energy dispersive spectroscopy (SEM-EDS) techniques of a conventional sol-gel glass in the system versus SBA-15 mesoporous matrix. The results indicated that the band at , which corresponds to amorphous calcium phosphate, began to split into a doublet at after 3 days in the case of conventional sol-gel glass and after 30 days in the case of SBA-15. Moreover, SEM-EDS studies revealed the formation of spherical agglomerate of particles consisting of needle-like crystals with a Ca/P ratio of 1.5 onto the materials surfaces after different periods of time, confirming the formation of a calcium-deficient nanocrystalline apatite layer.
(a)
(b)
These findings suggest that high surface areas and porosities are not enough conditions to achieve satisfactory biomimetic behavior. Consequently, different chemical strategies have been achieved to promote the bioactive response of silica-based mesoporous materials and, therefore to accelerate their bioactive kinetics. Some approaches consisted in the incorporation of units into the silica mesoporous network [41] or the addition of small amount of bioactive sol-gel glasses [42]. However, the real advance has been to synthesize sol-gel glasses with the outstanding textural properties (surface area of 195–427 () and pore volume of ) and ordered porous arrangements of silica-based mesoporous materials [43, 44]. In this sense, the synthesis of ordered mesoporous glasses was carried out employing the evaporation-induced self-assembly (EISA) method [45]. The synthesis involved the use of a nonionic triblock copolymer surfactant () as structure-directing agent and the employment of tetraethyl orthosilicate (TEOS), triethyl phosphate (TEP), and calcium nitrate, as , , and CaO sources, respectively. The textural and structural properties of this new family of mesoporous glasses can be controled by changing the CaO content of mesoporous glasses. A progressive evolution from 2D-hexagonal to 3D-bicontinuous cubic structure (see Figure 3) with an increase in the textural properties was observed when decreasing the CaO content [46]. The possibility of tailoring both structural and textural features of mesoporous glasses is undoubtedly an attractive advance towards the development of biomaterials that is able to fulfill the essential requirements for specific biomedical applications.
(a)
(b)
(c)
(d)
It has been evidenced that this new family of mesoporous glasses exhibits enhanced bioactive behavior, with even faster apatite phase formation than conventional bioactive sol-gel glasses. Besides, it has been revealed that the kinetics of apatite formation in mesoporous glasses is governed by their textural and structural properties, differing from conventional bioactive glasses, where the formation kinetics depended on compositional and textural properties. In this sense, the bicontinuous network in 3D-cubic mesoporous glass with low CaO content (10% in mol), which exhibits high accessibility to the diffusion process in pore systems, has led to the most accelerated bioactive response reported up to date [46]. Transmission electron microscopy (TEM) studies revealed that the formation of the crystalline apatite formation took place after only 4 hours of assay (see Figure 4). Their bioactive response was faster than mesoporous glasses with higher CaO content and 2D-hexagonal structure. Moreover, Figure 4 also shows the concentration and pH evolutions in SBF as a function of soaking time. It should be highlighted that there was not a significant increase in calcium and pH, which remained almost constant exhibiting values of 7.45 and 100 ppm, respectively [46]. This is a very important factor to avoid the cytotoxic effect that high local pH values could produce in living tissues. On the other hand, mesoporous glasses with 2D-hexagonal pore arrangements and high CaO content (37% in mol) provided a biomimetic mechanism where a sequential transition from amorphous calcium phosphate (ACP) to octacalcium phosphate (OCP) and to calcium deficient carbonate hydroxyapatite (CDHA) maturation, similar to the in vivo bone biomineralization process, was observed by TEM (see Figure 5) [47]. Such ACP-OCP-CDHA maturation sequence had never been observed before in any other bioactive ceramic. In this case, the open channel array and the textural parameters of 2D-hexagonal structure together with high CaO content allow an intense exchange that results in local acid pH values that favor the OCP phase formation.
(a)
(b)
(a)
(b)
(c)
Therefore, the possibility of tailoring the reactive properties of mesoporous glasses as a function of their textural, structural, and compositional properties allows to obtain materials with high biomimetism-derived biocompatibility. In addition, the fast and intense apatite-phase formation makes mesoporous glasses groundbreaking materials that will probably point out the future guidelines for the synthesis of the next bioactive bioceramics.
2.2. Silica-Based Ordered Mesoporous Materials as Controled Delivery Systems
Since 2001, when MCM-41 was purposed for the first time as controled delivery system [37], much research effort has been devoted to tailor the chemical properties of mesoporous carriers at the nanometer scale to achieve a better control over loading and release of molecules. The mesoporous carrier is selected according to the features of the guest molecule and the targeted application. Molecule loading is usually performed by impregnation methods by soaking the mesoporous carrier into a concentrated molecule solution. Loading solvent is selected attending to the chemical nature of the guest molecule [48]. Then, the release process is performed by placing the molecule-loaded mesoporous material into SBF solution [49], which exhibits ionic concentrations similar to those in the human plasma, or in physiological serum to make the detection process easy. Therefore, different guest molecules have been successfully confined into mesoporous silicas. Some of these molecules are drugs, such as ibuprofen [37, 50–52], amoxicillin [53], gentamicin [54, 55], erythromycin [51, 56], vancomycin [57], naproxen [58], aspirin [59, 60], diflunisal [58], captopril [61], itraconazole [62], and alendronate [63, 64]. Other guest molecules consisted of biologically active species, such as proteins (bovine serum albumin (BSA) [65–67] and certain amino acids (-tryptophan (-Trp) [68]).
The textural properties (i.e., pore diameter, surface area, and pore volume) of mesoporous materials are key factors that govern molecules adsorption and release [69, 70]. Moreover, functionalization of silica walls using different organic groups has been revealed as the main strategy to modulate molecule loading and release.
2.2.1. Influence of Textural Properties
The textural properties of mesoporous materials (pore diameter, surface area, and pore volume) play an important role in molecule adsorption and release. The influence of all these parameters on performance of controled delivery systems will be discussed in the next sections.
(a) Pore Diameter
The adsorption of molecules into mesoporous carriers obeys size
selectivity criteria, that is,
pore diameter will determine the size of the guest molecule to be hosted. Most
of the drugs commonly used in biomedicine are in the range of nanometers.
Therefore, they could be easily confined into the pores of nanostructured mesoporous
silica. However, pore diameter is a limiting factor when the adsorption of
large molecules, such as proteins, is aimed. This is the case of serum
albumins, which are one of the major components in plasma proteins. They are
involved in several relevant physiological functions of humans and upper
mammals, such as binding, transport and delivery of biologically active
molecules, including drugs, to specific places in the body [71]. Serum albumins
are usually composed of single chains of 582 amino acids and exhibit average length of 10 nm and width
of 6 nm [72, 73]. The adsorption of globular proteins, such as bovine serum
albumin (BSA) onto MCM-41 matrices, which exhibits pore diameters in the 2–5 nm range, has
been reported [74, 75]. The small diameter of MCM-41 impeded the confinement of
the protein into the mesopores and consequently, the protein was adsorbed on
the outer surface of the mesoporous carrier.
These results evidenced that when the confinement of large size proteins
into mesoporous channels is targeted, large pore matrices are needed. Our
research group recently applied an improved method to increase the pore
diameter of SBA-15 and BSA adsorption and release tests were performed [67].
The increase in the mesopore diameter was achieved by increasing the time of
hydrothermal treatment during the mesoporous synthesis, as described elsewhere
[76, 77]. The pore diameter of the resulting SBA-15 materials ranged from 8.2
to 11.4 nm employing hydrothermal treatments ranging from 1 to 7 days,
respectively. BSA loading tests revealed that there was a direct dependence of
protein adsorption on pore size, that
is, the greater the pore diameter, the higher the amount of protein
loaded. Therefore, conventional SBA-15, which presented pore sizes of 8.2 nm,
loaded 151 mg/g of BSA. The amount of BSA loaded increased to 234 mg/g, 242 mg/g, and 270 mg/g for mesoporous SBA-15 materials with pore sizes of 9.5 nm,
10.5 nm, and 11.4 nm, respectively (see Figure 6.a.1). In vitro delivery tests were carried out by soaking disc-shaped
mesoporous matrices in saline solution (NaCl 0.9%, pH 7.4). Release patterns
exhibited and initial burst profile where more than 90% of the adsorbed protein
was quickly released during the first 24 hours, followed by a sustained release
and the complete delivery was achieved after 192 hours of assay in all tested
samples.
Pore diameter has been revealed as a critical factor that modulates
release rate of molecules to the delivery medium. This fact was firstly
evidenced when MCM-41 materials exhibiting different pore sizes ranging from
2.5 to 3.6 nm were synthesized using cationic surfactants with different length
alkyl chains [78]. Ibuprofen, a common anti-inflammatory,
was selected and release tests were carried out in a simulated body fluid (pH
7.4) [49]. The resulting release profiles, which are displayed in Figure 6.a.2,
revealed that release kinetics strongly depended on mesopore size. Thus, the
amount of ibuprofen released to the delivery medium diminished as the pore size
decreased from 3.6 to 2.5 nm, evidencing that drug dosage can be modulated
depending on the mesopore size.
(a) Influence of pore diameter
(b) Influence of surface area
(c) Influence of pore volume
(b) Surface Area
The adsorption of molecules into mesoporous matrices depends on the
adsorptive properties of the silica surface. Therefore, the chemical
interactions between the silanol groups covering the silica surface and the
functional groups of the guest molecule will determine the amount of molecule
loaded. Surface area of mesoporous matrix is the textural parameter that points to the silica
surface susceptible to interact with the guest molecule and, consequently, it
is expected that the higher the surface area, the higher the amount of molecule
loaded. This fact was confirmed when several MCM-41 matrices with different
surface areas were employed as ibuprofen delivery systems [78]. MCM-41
mesoporous materials exhibiting surface areas of 768, 936, 1087, and 1157 /g
loaded 110, 190, 230, and 340 mg/g of ibuprofen, respectively, evidencing the
increase in the drug loading with the surface area increase.
With the aim of further establishing the effect of surface area on molecule adsorption, two
mesoporous materials, MCM-41 and SBA-15, which exhibit the same structure (2D-hexagonal, p6mm symmetry) but different surface
areas (1157 /g for MCM-41 and 719 /g for SBA-15) were
tested as alendronate delivery systems [63]. Alendronate is a potent
bisphosphonate commonly employed for osteoporosis treatments that inhibits bone resorption
by osteoclasts and it is a promising drug to be locally applied in bone implant
technologies. As displayed in Figure 6(b), the amount of alendronate loaded into
MCM-41 (139 mg/g) was higher than that into SBA-15 (83 mg/g), confirming the dependence of drug
loading on surface area. Release profiles from both mesoporous carriers into
the delivery medium (NaCl 0.9%, pH 7.4) exhibited an initial burst effect when ca. 20% and ca. 50% of alendronate loaded were quickly released from MCM-41 and SBA-15,
respectively (see Figure 6(b)). Afterwards, alendronate was released to the
delivery medium in a sustained manner, following first-order kinetics for MCM-41
and zero-order or linear kinetics for SBA-15 materials.
(c) Pore Volume
As previously mentioned, the adsorption of molecules into mesoporous
matrices is a surface-molecule interaction [79]. For this reason, pore diameter,
as the size-selectivity limiting factor, and surface area are key parameters
that control the adsorption of molecules. However, when the confinement of
really large molecules is targeted, the pore volume available to host the guest
molecule plays also an important role in molecules adsorption.
Recently, mesostructured cellular foams (MCFs) have been tested as host
matrices for the adsorption of several enzymes and proteins [80, 81]. The
synthesis of MCF materials was carried out by using triblock copolymers as
surfactants and introducing a swelling agent, commonly trimethylbenzene, into
the structure directing template solution [82, 83]. TEM image of SBA-15 (see Figure 6(c)) revealed the presence of two-dimensional hexagonal arrays of pores,
typical of hexagonally ordered mesoporous materials with pore sizes of ca. 8 nm. In the case of MCF, spherical
cells of ca. 28 nm with windows of ca. 18 nm were observed (see arrows in Figure 6(c)). These matrices seem suitable to be used as delivery systems of
large-size molecules, such as proteins. The amount of BSA loaded was higher in
MCF (240 mg/g) than in SBA-15 (150 mg/g), following the same trend than pore
volume, being 1.9 /g and 1.1 /g for MCF and SBA-15,
respectively (see Table inset Figure 6(c)). This results evidence that pore
volume is an important parameter to consider when aiming at confining large-size
and large-volume molecules. However, the effect of other parameters, such as
pore diameter into protein loading cannot be overruled. The pore size of SBA-15
is just on the limit of BSA dimensions, which could complicate the protein
adsorption into the mesopores. However, such steric hindrance would not take
place when MCF, with a pore opening of 18 nm, is used as host matrix.
2.2.2. Influence of Organic Functionalization
A straightforward way to optimize the performance of mesoporous matrices as controled delivery systems consists in organically modifying the mesoporous silica walls with appropriate functional groups [69, 70]. On the surface of silica walls, there is a high surface density of silanol groups that can experience organic modification by covalently anchoring organic silanes, SiR’ [84]. The adsorption and release of molecules can be effectively controled by using the adequate organic functionalization. To reach this goal, two main strategies have been widely employed. The first one involves the use of organic groups that promote attracting host-guest interactions with the functional groups of the guest molecule. The choice of the functional group would depend on the targeted molecule. For instance, ibuprofen delivery tests from MCM-41 functionalized with several organic groups (chloropropyl, phenyl, benzyl, mercaptopropyl, cyanopropyl, and butyl) revealed that polar groups induced greater ibuprofen adsorption than nonpolar groups [85]. The second strategy concerns the functionalization of mesoporous silica walls using hydrophobic species. In the next sections, some relevant results regarding functionalization using amino groups and functionalization employing hydrophobic groups will be described.
(a) Functionalization Using Amino Groups
Functionalization
using amino groups has been a widely employed strategy to attain a better
control over molecule loading and release. Therefore, amino functionalized MCM-41
and SBA-15 mesoporous matrices have been tested as alendronate delivery systems
[63]. The obtained results indicated that the amount of drug loaded into
amino-modified materials was almost 3 fold that of unmodified matrices. This
fact could be explained by the stronger attracting interactions taking place
between phosphonate groups in alendronate and amino groups of functionalized
matrices compared to the weaker interactions through hydrogen bonds between
phosphonate groups and silanol groups of unmodified materials. Therefore,
amino-modified MCM-41- loaded more alendronate (366 mg/g) than
unmodified MCM-41 (139 mg/g). Also, amino-modified SBA-15- matrix
loaded more amount of drug (220 mg/g) than unmodified SBA-15 (83 mg/g). Moreover, drug release assays revealed
that amino functionalization of mesoporous silica allowed a better control on
the drug release. Thus, the amount of alendronate released after 24 hours was
28% and 58% of the total amount loaded for amino-modified and unmodified
MCM-41, respectively. In the case of SBA-15, the 11% and 56% of the total
alendronate loaded were
released from amino-modified and unmodified materials, respectively, after 24
hours of assay.
Recently,
the amino modification of SBA-15 mesoporous materials using different functionalization
degrees has been revealed as key factor to modulate alendronate dosage [64]. The
amino-modification method was improved by adding a catalyst during the
functionalization process. Alendronate loading as a function of experimental
functionalization degree is displayed in Figure 7. The amount of alendronate
loaded linearly increased as the functionalization degree increased, ranging
from 88 mg/g, for the lowest functionalization degree (0%), to 121 mg/g, for
the highest functionalization degree (47%). Moreover, functionalization degree
also allowed a better control over alendronate release (see Figure 7), that is, the higher the
functionalization degree, the smaller the percentage of alendronate released.
This work evidences the possibility of modulating drug dosage by varying the
functionalization degree of mesoporous matrix, which is particularly
interesting when using bisphosphonates since the high potency of these drugs
implies that only small local doses are needed.
The
organic modification of mesoporous matrices was observed to control release
kinetics of proteins, such as BSA [66, 67]. As previously described, SBA-15
materials exhibiting different pore diameters were obtained by varying the time
of hydrothermal treatment during their synthesis. These materials were then
modified using amino groups to favor attracting interactions with amide groups of protein. It should
be highlighted that organic functionalization always leads to a diminishing of pore
diameter and consequently, as BSA size is on the limit of SBA-15 mesopore
dimensions, after amino functionalization, the amount of BSA loaded decreased
compared to unmodified materials. However, organic functionalization of SBA-15
had a noticeable effect on BSA kinetic release rate. As a consequence of
organic functionalization, the initial burst effect observed in BSA release
from unmodified matrices (ca. 90%) was
drastically reduced after amino modification (ca. 30%). Moreover, the release of protein from the mesopores of
all amino-modified materials was incomplete, whereas the total amount of loaded
BSA was released after 192 hours from unmodified materials.
Moreover,
functionalization of large-pore MCF materials with amino groups has been
demonstrated to increase the affinity for BSA [81]. There was an increase in
BSA adsorption after amino-modification MCF- (270 mg/g) compared
to unmodified MCF (240 mg/g), despite of the decrease in pore diameter due to
the functionalization. However, amino functionalization has a strong influence
on BSA release rate. Unmodified MCF exhibited an initial burst effect when
almost 60% of the protein loaded was quickly released to the delivery medium. This burst effect
was drastically reduced to ca. 10%
after amino modification. After 24 hours of assay, 62% of the loaded BSA was
released from unmodified MCF material and this percentage decreased to 22%
after functionalization with amino groups.
These
studies evidence that organic functionalization of silica mesoporous matrices
using amino groups allows the control of release kinetics of several molecules
that exhibit functional groups able to undergo attracting interaction with
amino functions.
(b) Functionalization Using Hydrophobic Groups
The functionalization of mesoporous silica walls with hydrophobic groups
is an alternative to control the release of certain molecules. Therefore,
SBA-15 mesoporous matrix has been functionalized with different alkyl chains, octyl
() and octadecyl (), and erythromycin, a hydrophobic antibiotic was
selected to carry out adsorption and delivery assays [56]. As previously
commented, organic modification led to a decrease of the effective pore
diameter and surface area of SBA-15. The decrease in surface area resulted in diminishing the amount
of erythromycin loaded, 130 mg/g and 180 mg/g for SBA-15- and
SBA-15-, respectively, compared to unmodified SBA-15 (340 mg/g). Comparable
results were reported using mesoporous materials modified by silylation as
ibuprofen delivery systems, showing a lower drug loading after silylation [86].
In addition, as a consequence of functionalization with hydrophobic chains,
there was a decrease in the wettability of the surface, which made difficult
the penetration of the aqueous delivery solution inside the mesopores
originating a decrease in the delivery rate. In fact, the erythromycin release
rate from SBA-15-C18 material was one order of magnitude lower than that from
unmodified SBA-15.
Recently,
functionalized SBA-15 has been proposed as delivery system of -tryptophan (-Trp) [68], a hydrophobic amino acid present in the
three-dimensional structure of many peptides, proteins, and growth factors of
interest in bone tissue regeneration technologies [87–89]. -Trp presents an aromatic indol ring
that makes necessary
to modify the silanol-rich walls of SBA-15. In fact, unmodified SBA-15 loaded less than
5 mg/g of -Trp, probably due to the
extremely different chemical nature of hydrophobic amino acid and hydrophilic
SBA-15. The small amount of -Trp
adsorbed into SBA-15 could be interactions through hydrogen bonds between
deprotonated carboxylic group of amino acid and silanol groups covering the
silica walls (see Figure 8(a)). For this reason, SBA-15 matrix was organically
modified using quaternary amines with different alkyl lengths ( and ),
as illustrated in Figures 8(b) and 8(c) [68]. Functionalization with short
alkyl chains ()
allowed coulombic attracting interactions between deprotonated carboxylic
groups of amino acid () and protonated quaternary amines ()
covering the mesoporous surface. In this case, the amount of -Trp loaded into SBA-15- matrix was higher (43 mg/g) than in unmodified SBA-15 (<5 mg/g). On the
other hand, using long hydrocarbon chains (),
two-thirds of the silica surface was functionalized. This high degree of
functionalization with hydrophobic chains promoted interaction of mesoporous
surface with indol group of -Trp,
and consequently, the amount of amino acid loaded increased to 82 mg/g.
Moreover, release profiles of amino acid from functionalized SBA-15 materials
are displayed in Figure 8(d). In both cases, there is an initial burst effect,
where most of the -Trp loaded is
quickly released to the delivery medium. After such burst effect, the rest of
amino acid is released in a sustained manner, following first-order and
zero-order or linear kinetics from SBA-15- and SBA-15- matrices,
respectively.
(a)
(b)
(c)
(d)
3. Conclusions
Nanostructured mesoporous silicas are promising candidates to be used as controled delivery systems of a wide range of biologically active molecules of interest in bone tissue regeneration applications. Recently, much research effort is being dedicated to enhance the bioactive capability of ordered mesoporous silicas. Supramolecular chemistry has allowed the design and synthesis of ordered mesoporous glasses with fascinating textural and structural features that open many paths for the research on high-bioactive nanostructured materials for bone tissue regeneration purposes. These materials have shown high bioactive responses and biomimetic behavior not reported for any bioceramic up to date, which open promising expectations in the biomedical field. Furthermore, much attention has been devoted to tailor the textural properties of nanostructured mesoporous carriers, pore diameter, surface area, and pore volume, with the aim of optimizing their loading capacity depending on the guest molecule. Moreover, organic modification of mesoporous silica walls brings up many possibilities in molecule adsorption and release by promoting host-guest interactions. The outstanding properties of mesoporous silicas make them suitable to be used as starting materials for the design of scaffolds for bone tissue engineering technologies.