Using hydrogels for the tissue engineering of cartilage represents potentially massive societal and company benefits. Hydrogels form a 3 dimensional scaffold which can facilitate the growth of native-like articular cartilage in a bioreactor, with cells extracted from non-weight bearing areas of the patient with minimum cosmetic impact. This report highlights the process by which, within an acceptable timeframe, viable articular cartilage can be produced, and transplanted back into the patient, ending the potentially debilitating degradation of the native cartilage and limited use of the joint.
Executive Summary 1
Process overview 3
Process Conditions 4
Cell extraction and isolation 4
Cell proliferation 4
Incubation conditions 5
Hydrogel Mass Transport 6
Hydrogels Ltd is a biotechnology company that specialises in the production of hydrogels. The company sees its position occupying a large sector of the biotech market as a good one from which to move into tissue engineering, given our expertise in hydrogels.
The company has chosen to specialise in one human tissue in particular, which will be the most advantageous to the company in the long term. Out of bone, skin, cartilage, ligaments and the bladder we have chosen cartilage tissue to base our design process around. There are a number of reasons for choosing to engineer cartilage, the most significant of these reasons is the large market place where engineered cartilage would be enormously beneficial to the patient. Osteoarthritis is a severe form of arthritis and is caused largely by the degeneration of the articular cartilage surrounding the bone at the joint. In particular the metacarpal trapezium joint in the thumb and knee joints are of particular interest, so hydrogel structures will be used for these clinically relevant sites. (Pellegrini, V. D. 1991)
There is very little reparation that naturally takes place in the cartilage, (Buckwalter and Mankin 1998), so once the tissue becomes damaged it is highly likely that the joint will produce less than satisfactory performance for the rest of the patient life without medical intervention. (Ochi, M. et al 2002) Osteoarthritis affects 8 million people in the UK, many more in America, and these numbers are sure to grow considering the aging population of 1st world countries. The continuing obesity epidemic coupled with the typical western diet of processed, denatured and pasteurised foods will amplify the problem inflating the demand for tissue engineered cartilage worldwide.
Cartilage donor cells are harvested from the patient. The autographed cartilage can be removed from a number of sites including the joint where the engineered tissue is required. However removing a sample from the affected area is obviously intrusive, costly and there is potential for complications. Jakob et al, (2003) discuss the creation of additional injury as a result of cell harvesting from the affected area and the low cell yield. There is, thankfully a much simpler option, which is to remove cartilage from the patients ear or nose, also known as non-weight bearing cartilage. The size of the sample to be taken is considered to have minor cosmetic impact. Moreover studies have shown NC (nasal chondrocytes) to have higher proliferation rate and more reproducible chondrogenic capacity in vitro than articular cartilage. (Kafienah et al., 2002; Tay et al., 2004)Ear and nasal cartilage has the additional advantage over articular cartilage in that the quality of the regenerated tissue appears to be independent of the age of the donor. (Rotter et al., 2002)
The use of allogenic (human derived) and xenogenic (animal derived) cell sources has been considered, however currently the risk of adverse immunogenic responses are considered too high. (Platt, 1996)
Articular cartilage consists of four distinguishable layers which vary in collagen fibre alignment, thickness, ECM composition, cellular morphology and the composition of proteoglycan as shown in figure 1. (Kim, Mauck et al. 2011) The improvement of bioreactors aims to condition the cartilage into mimicking the native cartilage as closely as possible
Figure 1. Cellular alignment in articular cartilage showing four distinct layers. H&E: hematoxylin and eosin; GAGs; alcian blue stain for glycosaminoglycans. (Kim, Mauck et al. 2011)
The harvested cells are isolated, and allowed to proliferate in vitro on a grow-medium. Once the required numbers are produced the cells are seeded onto the hydrogel structure so the extracellular matrix can form. This process takes place in a bioreactor where the structure is subjected to forces designed to imitate the forces found in the particular joint. After this the cartilage is transplanted into the patient.
If bacteria or viruses come in contact at any time of the process the whole operation will be compromised. The best case scenario is the tissue is thrown away and started again which costs time and money, the worst case scenario is the contaminated tissue is replanted into the patient severely affecting the patient’s health. Therefore from the time of extraction, isolation and proliferation to when the cells are seeded onto the hydrogel and then placed in the bioreactor aseptic conditions are maintained. All operations take place in ISO 5, Class 100 sterile conditions, and are protected by HEPA filtered laminar air, which removes ; 99.7% of all particles ; 0.3 micron in size.
A pressure differential between the critical zone, where the process takes place, and the cleanroom and changing room is created so that air flows from the critical zone into the other rooms and not the other way around. This means only class 100 air is in the sterile area. The highest risk areas for contamination are often at the interface between the processes, for example seeding the cells into the hydrogel so these areas are validated for safety with safety procedures before-hand .Personnel enter and leave the cleanroom through airlocks and wear full biological cleanroom suit. The ISO/EN 10993 part 5 protocol is used to assess the cytotoxicity of the bioreactor components. Good Manufacturing Practice (GMP) and Good Laboratory Practice (GLP) ARE followed throughout the process.
Cell extraction and isolation
A combination of type II collagenase and pronase is used to remove the chondrocyte cells from the extracellular matrix before the cells can be cultured. The isolated chondrocytes are seeded onto the culture medium to begin stage 1 of cell proliferation
Cells grown in monolayer grow quicker, however growth is often limited by nutrient deficiency after a certain time period. In suspension cultures doubling times are generally 3 times as long however final cell densities when compared with monolayer growth are higher and final cells numbers are considerably higher. Suspension cultures are also significantly less labour intensive and hence less costly. With the aim to minimise growth time and labour costs the cell proliferation stage is divided into two parts. Cell de-differentiation during monolayer expansion as discussed by Benya and Shaffer (1982) is also largely overcome using the staged process. Generally chondrocytes express a greater differentiation capacity when cultivated in a 3 dimensional environment, (suspension culture). (Albrecht et al 2006)
Acharya et al, (2011) have suggested that chondrocyte cells can be used in combination with mesenchymal cells to stimulate chondrogenesis, this may prove advantageous since the mesenchymal stem cells are likely to have good availability from different adult tissues including bone marrow (BM-MSC) and adipose tissue (AT-MSC). Using MSCs however remains controversial because of the apparent instability of the chondrocyte phenotype, (Pelttari et al., 2006) and inferior ECM production when compared with chondrocytes. (Kim, Mauck et al. 2011).
The time needed to produce the required number of cells is calculated. The average cartilage defect is estimated to be roughly 3cm x 2cm x 4mm, which considering cell densities in articular cartilage gives 50 million chondrocyte cells. 5 million cells can be harvested from the patient site, ie the ear, without long term cosmetic damage. Taking the doubling time for the chondrocyte cell in the specified conditions the overall proliferation time is calculated. Cell doubling time varies with conditions, growth method, and other aspects related to the donor such as age and harvest site. However most available literature sources agree that in monolayer culture the doubling time is around 3 ï¿½ 0.5 days, in suspension culture this jumps to around 10 ï¿½ 1 days. (Strobel, S. 2007) Hence a compromise must be struck between a shorter proliferation time on one hand, and higher expenses along with higher likelihood of dedifferentiation on the other. However differentiation and proliferation are known to be somewhat of a contrasting phenomenon, so less emphasis is placed on differentiation at this stage, until the bioreactor stage. (Potten, C.S. 1982) Monolayer cultivation also requires large surface areas, so at the later stages of proliferation suspension cultures are useful to reduce working space as cell numbers increase.
The process is split into 2 stages. 5 million cells are cultivated in monolayer for a period of 6 days. Given a doubling time of 3 days, the end result is 20 million cells. The medium is refreshed once during this time. After 6 days the cells are moved into suspension culture with a doubling time of 10 days.
The cells are cultivated for 13 days in suspension, giving a total cell count of 50 million, cultured for a total of 19 ï¿½ 2.3 days.
The chondrogenic growth medium used will be DMEM containing 4500mg/L glucose, 584 mg/L glutamine, 10% FBS (fetal bovine serum) (Freed, L.E. 1993) insulin growth factor, fibroblast growth factor, growth hormone, ascorbic acid-2-phosphate and transferrin as described by (Kim, Mauck et al. 2011)The cells are exposed to the growth factors, for example bFGF- 2, not only to enhance de-differentiation but because research shows it can improve the cells capacity to re-establish the differentiated phenotype at a later stage of the process such as in the bioreactor.(Barbero et al 2003) Cell incubation takes place at 37C and 5% CO2, 20% O2.
Conventionally 20% oxygen has been used, however studies carried out under low oxygen tension (5% O2) have shown chondrocytes to have a higher re-differentiation capacity. (Murphy and Polak 2004) This may be because it is closer physiological concentration of oxygen. Studies also show low oxygen tension to have a particularly beneficial effect on chondrocytes harvested from patients over 40 years of age, whose tissues can often experience pronounced dedifferentiation and lack of proliferation. (Malda et al 2003) Further tests should be carried out to find the optimum oxygen concentration for the process.
Hydrogel Mass Transport
The hydrogel is a 3D scaffold which holds the cells in place and at the same time facilitates the exchange of waste and carbon dioxide for fresh nutrients and oxygen allowing the cells to survive. The hydrogel will be 1% hyaluronic acid which has shown excellent structural and diffusive properties. (Leach, J. B. 2005).The scaffold is used to mimic the environmental and mechanical properties that the native cartilage is subject to.(Nicodemus and Bryant 2008)
The hydrogel represents an alternative to the, until recently, most commonly used scaffold architecture consisting of non-woven fibre meshes and porous sponges. (Putnam and Mooney 1996) After the proliferation period the cells are seeded into the hydrogel at densities between 10-20 million cells/ ml. Dynamic loading (seeding) of cells into the hydrogel is achieved in a stirred-flask bioreactor which leads to a relatively uniform cell distribution, although the cells still tend to preferentially locate on the periphery instead of the core regions. (Wendt et al 2003).
The higher the density of cells in the hydrogel, the better the mass transfer parameters need to be in order that cell necrosis does not take place. The hydrogels are also optimised for ECM production and connectivity which forms as the hydrogel degrades, leaving the ECM in its place under the correct conditions. After a period of time in the bioreactor the hydrogel is completely replaced by the formation of the ECM which surrounds the seeded cells, this forms the cartilage which is put back into the patient at the site of arthritis. Before this can happen, however, the diffusion of nutrients into the hydrogel limit the size of the cartilage which can be grown, therefore calculations have been carried out the determine the maximum size that can be obtained before cell necrosis occurs. The substance which diffuses most slowly through the hydrogel will be the one that limits the maximum size of the tissue.
There is debate in the literature as to whether oxygen or glucose is the diffusion limiting substance in the hydrogel. In these calculations, oxygen has been considered as the limiting substance given research done by Zhou et al (2008) and because data is available for the calculations.
The cell density in the gel is 2×106 cells / ml, the bulk concentration of oxygen (z-0) is 2×10-4 mol/L, the diffusion rate is 10-5mm2/s
Input – Output + Accumulation – Reaction =0. FA represents the flux of glucose (mol per unit area per hour)
Dividing by dz and taking the limit as dz ï¿½0,
for dilute concentrations we can leave out UCA and combine the two above equations. The kinetics are assumed to be zero-order.
Put the equations into a dimensionless form using
We call the second term the Thiele modulus since it is just a ratio of a reaction rate to a diffusion rate for a zero order reaction. The equation is integrated.
At l=0, y=1, CA = CA0
At l=1, . Symmetry condition
No diffusion across the mid-plane so the gradient is 0 at l=1. Integrating twice and using the boundary conditions the dimensionless concentration profile is
Which is only valid for values of the Thiele modulus less than or equal to 1
Evaluating the parameters, k
(a) Concentration reaches zero by the time oxygen reaches the centre of the gel, y=0 at l=1 solving
, to find .
Therefore L=3.8 mm thickness before the oxygen concentration drops to zero
(b) Considering minimum concentration at the centre to be 0.1mol/L, which is half of that a the surface, ie y=0.5 at l=1
Therefore L = 2.7mm thickness before the concentration drops below half of that at the surface.
Derivation and calculations (Fogler and Gurmen 2008)
DAB is a function of the hydrogel composition, matrix formation and other environmental conditions and therefore changes with different hydrogel types. Through research and development this diffusion parameter can be optimised. CA0 is a function of the glucose concentration and the rate of matrix formation etc. K is a function of cells per unit volume multiplied by the moles of glucose used per cell per hour. This affected by how closely the cells are seeded into the hydrogel. The cells are seeded at a density of 20 million cells per ml. If the cells are seeded less densely then there is less oxygen demand in the hydrogel, hence the maximum thickness will increase, although the ECM might take longer to form. (Zhou et al 2008)
Clinically relevant size is around 4mm thickness, (Malda et al 2004) by 3cm long and 2cm wide. At 20 million cells per ml, this gives 48 million cells. Just under the amount produced in the proliferation stage. In fact the length and width are not important for the diffusion of glucose, only a greater thickness would be important. It is interesting to note in the body nutrients and oxygen are provided to the avascular articular cartilage in a curious way. During compression of the cartilage in the knee joint, for example, waste and carbon dioxide are repelled from the medium under pressure, as the force is removed fresh glucose and oxygen are drawn back into the cartilage, thereby providing the cells with adequate nutrition.
This is often called intra-tissue fluid flow generated by dynamic loading.(Kim, Mauck et al. 2011) This could be used as a useful method to increase the diffusivity of the hydrogel, consequently increasing the maximum cartilage thickness. To make it possible to maintain the growth up-to 4mm thickness it is recommended to seed the cells at a lower density such as 10 million cells per ml, and use dynamic forces to refresh the nutrient and oxygen in the hydrogel. Given the current maximum thickness it is estimated halving the cell density and adding dynamic nutrient refreshment that the required thickness can be achieved.
The basic process for chondrocyte cultivation is well established. There must be the correct pH, adequate oxygen and carbon dioxide available and the culture must allow for the uptake of nutrients and removal of waste. There are other factors as well, incubation of the scaffolds with BMP-2 and insulin is known to induce re-differentiation and thereby enhance ECM synthesis. (Yamaoka, Asato et al. 2006) Nevertheless, another aspect of the biophysical environment has been shown to enhance chondrogenesis and reduce dedifferentiation. This is the in vitro mechanical stimulation of the culture as the ECM forms in place of the hydrogel.(Kim, Mauck et al. 2011) De-differentiation means that the chondrocytes lose their ability express articular cartilage-specific extracellular matrices (ECM) (Kim, M. et al 2011)
Meaning the chondrocytes will express collagen type 1 protein instead of the desired type 2 collagen which forms the cartilage structure.(Pï¿½rtner, Nagel-Heyer et al. 2005) The mechanical stimulation also helps to modulate cellular alignment and matrix connectivity thereby creating a structure which has been biologically adjusted to its predetermined environment and showing cellular alignment similar to that of figure 1.
The bioreactor will provide the physical stimulus in a controlled manner so that the optimum conditions can be recreated and optimised each time. The physical stimulation comes in four forms which are direct compression; hydrostatic pressure, high shear and low shear fluid environments (Darling, E.M. et al 2003).These forces can be induced by rotating-wall vessel, concentric cylinders and perfusion bioreactors. Optimum chondrogenesis is believed to be obtained when combining a number of physical stimulations at one time such as the three dimensional strain field bioreactor described by Meyer et al (2006).
The mechanical stimulus is designed to mimic the actual conditions found in the knee thereby creating a degree of ECM similarity. Figure 2 shows the geometry of the culture chamber in which the hydrogel is subjected the shear forces and pressure, and the inlet and outlet sections for perfusion of the medium. This bioreactor typically consists of four integrated subsystems: the culture chamber (as shown in the diagram) the electronic circuit, pneumatic circuit and hydraulic circuit.
Figure 2. Diagram of the perfusion bioreactor culture chamber. Lagana, K., M. Moretti, et al. (2008)
In addition to the mechanical stimulus of the bioreactor, there are other aspects to consider. These include maintaining the incubation conditions described in the proliferation section. The bioreactor must also be easy to sterilise, because if it isn’t either it will take too long and cost too much or it will not be appropriately free from contaminants for the next tissue. The ease of cleaning is built into the design. Also built into the design is a simple nutrient refreshment technique such as perfusion.
Mass transfer parameters in the hydrogel limit the thickness of the cartilage that can be produced without cell necrosis occurring. However preliminary calculations would suggest that clinically relevant thicknesses of 4mm would possible to maintain with lower cell densities allowing the cells to live long enough to produce the required collagen type 2 ECM. Dynamic compressive and shear forces have also been shown to enhance mass transport parameters, in similarity to the phenomenon occurring in the native joint cartilage. The bioreactor is used, as successfully shown in literature, to allow for the formation of the ECM is such a manner that it closely represents the tissue found in the joint, thereby providing maximum benefit to the patient. Our position as a manufacturer of hydrogels and specialists in the biotech industry would suggest that the outlined process could be successfully implemented for the benefit of the patient in both joint longevity and mobility.