A PubMed search was performed under the search terms (Osteoarthritis AND Treatment) OR (Bioreactor AND Cartilage) OR (Bioreactor AND Tissue Engineering) OR (Cartilage AND iPSCs) and a combination of those terms, with results limited to the English language and published in the last ten years. Other articles used in this review were taken as a result of cross-referencing. A complementary PubMed search under (Bioreactor AND Cartilage AND Tissue Engineering) was performed, resulting in 287 articles. Articles that did not address hyaline cartilage or osteoarthritis were not considered. For this review, the article choice was based on the journal impact factors, translatability to newer and novel technologies, the potential impact of the study, and opinions from experts in cartilage tissue engineering. Thus, in total, 85 articles were included in this systematic review.

2.1 Bioreactors for in vitro tissue-engineered articular cartilage optimized development

Tissue-engineered cartilage constructs can be formed from multiple stem cell types and numerous techniques [19]. A standard medical implant testing method is in vitro experiments, followed by in vivo animal testing before clinical trials. In vitro, research investigates the applicability, toxicity, and biocompatibility of tissue-engineered implants to reduce possible complications of in vivo applications. This screening is necessary because 90% of biomedical research studies fail to enter routine clinical use even after long testing periods (~ 20 years) [27].

The use of monolayer cell cultures often is the starting point for investigating the potential of TE strategies. However, these cultures do not resemble the dynamic complex environment of the tissue in vivo. Therefore, 3D cell culture models have recently been used to analyze microenvironmental cues that can regulate biochemical responses for chondrocyte gene expression and cell differentiation [28].

For assessing and testing the interplay between tissue, biological, chemical, and mechanical stimuli, a bioreactor is a desirable device to mimic multiple characteristics of the in vivo milieu [29]. In this way, it can be found whether the obtained cartilage tissue has the potential to be implanted or not [21]. E.g., When chondrocytes cannot withstand the bioreactor's environment, they usually start to express genes and cell markers that cause differentiation to other cell phenotypes or cause cell death. If this happens, the cartilage-engineered constructs have a low chance of surviving in vivo and will not make it into further stages to reach clinical trials [30].

Using a bioreactor for screening tissue-engineered cartilage constructs can accelerate the translational process for more clinically relevant research [31]. However, a bioreactor design is not trivial, and its validation can be challenging due to multiple parameters that can be considered (Table 1) to resemble a big part of the in vivo environment for cartilage differentiation, regeneration, and maintenance.

Table 1 Summary of representative stimulus to consider for the design of a bioreactor for tissue-engineered articular cartilage

In 1959, Russel and Burch proposed improving animal welfare through an initiative called the 3Rs: "replacement, reduction, and refinement." The 3Rs aimed to reduce the number of animal studies by using other testing methods [49]. Thus, another significant advantage of bioreactors’ in vitro screening of tissue-engineered constructs is reducing the number of animals required for in vivo trials. If these articular cartilage constructs fail or do not respond positively to the bioreactor stimuli, the constructs need to be neglected for animal testing [29]. In humans, the articular cartilage is a mechanically active tissue that must withstand approximately three times the individual’s body weight [50].

Without bioreactors, the constructed tissues should be tested in animals, which in this case, represent the tissues' in vivo environment. These tests are mainly performed on small animals, like mice, rats, and rabbits [29]. By doing this, the tissue response related to the biological factors can be investigated. However, these small animals cannot expose the tissue to the same forces and other stimuli as humans [20]. Therefore, large animal studies are preferred for in vivo translations.

According to Peroglio et al. [29], a bioreactor, including a control unit and software, can be built for the same amount of money as performing a study with ten large animals. They calculated that testing on a large animal model costs around nine thousand dollars per animal. Thus, ten large animals can be used with a budget of about 100,000 dollars. However, for the same money, a sophisticated bioreactor can be built [29]. Furthermore, it is a system that can be used multiple times and has a broader range of applications. In this manner, bioreactors can considerably reduce the amount of animal research and consequently reduce research costs [30], specifically for articular cartilage TE on OA research.

Another advantage of reducing the number of animals for research is that it saves experimental time. Before performing an animal experiment, a researcher must not conflict with any animal ethics related to the experiments. Therefore, an application to the respective institution´s animal ethical committee must be sent, including details about the project and the intended animal use [51]. The time between the application and the start of the experiment can include a long waiting time, which could be saved using bioreactors for in vitro testing [52]. In that way, it is possible to test faster multiple cartilage constructs created out of different cell lines and techniques compared to animal studies. Therefore, by using a bioreactor, tissue-engineered cartilage development can be accelerated [31].

Three basic bioreactors have been commonly used to engineer multiple tissues in the lab, including hyaline cartilage. A schematic overview of those bioreactors can be found in Fig. 2. In this review, it is emphasized the use of iPSCs cell-seeded scaffolds in combination with bioreactors. However, other 3D systems, such as embryoid bodies [53] or iPSCs-derived cartilage organoids [54], are encouraged to be used for in vitro cartilage fabrication and OA research.

Fig. 2

Schematic representation of A cell-seeded scaffold, B Rotating wall vessel bioreactor, C Spinner flask bioreactor, and D Flow perfusion bioreactor

Scaffolds play a significant role in cartilage TE by creating an ideal cellular environment to enhance tissue integration. Essential features, including scaffold architecture, material biocompatibility, biodegradability, and chemical and mechanical properties of the scaffold, can be achieved by multiple manufacturing methods. Essential information and description of these requirements for cartilage TE can be found elsewhere [55].

The rotating wall vessel (RWV) (Fig. 2B) bioreactor consists of two concentric cylinders with space in between, suitable for the cell-seeded porous scaffold (Fig. 2A) location. The wall rotation induces a laminar fluid flow, facilitating oxygen diffusion and nutrient supply through the scaffold [56]. Studies have demonstrated higher glycosaminoglycans (GAG) and collagen II expressions using RWV on cartilage constructs [57, 58].

The spinner flask is a simple bioreactor that uses a magnetic stir bar to gently agitate the medium where needle-extended cell-seeded scaffolds are cultured (Fig. 2C). Medium agitation helps nutrient and oxygen supply throughout the cell-seeded scaffolds (Fig. 2C), allowing an increment of cell proliferation, GAG content, and collagen II compared to static cultures for cartilage TE [59]. In both RWV and spinner flask bioreactors, to reduce cell waste, 50% of the medium must be replaced every other day during the culture time [25]. Therefore, the flow perfusion bioreactor (Fig. 2D) is one alternative for cartilage TE constructs. This system uses a controlled pump that allows the medium to be perfused through the cell-seeded scaffold, favoring nutrient supply and waste removal [60]. Flow rates between 0.2 and 1 mL/min have demonstrated good cell viability and ideal cell waste removal. Furthermore, This bioreactor has shown a significant increment in DNA and GAG content on Poly-L-lactic acid scaffolds for cartilage formation [32].

Furthermore, bioreactors can also be used for OA drug research. For this application, it is essential to identify drug candidates for potential OA treatments and harmful compounds that can further promote the progression of the pathology. Currently, for drug testing, cell-loaded scaffolds or cell organoids are used in combination with flow perfusion bioreactors, where cell responses to specific drugs can be monitored. E.g., Nichols et al. developed a 3D-printed bioreactor that allows the multidirectional flow of medium containing potential screening drugs through the cell-loaded scaffold [61]. This bioreactor uses a complex microfluidic system that optimizes the drug-medium supply, resembling in vivo conditions. Furthermore, it has optical access for drug response monitoring, which can be used for comparison prior to or in parallel with animal studies to increase the profile safety of the screening candidate drugs [14, 61].

2.2 Biochemical and Physical stimuli necessary for tissue-engineered articular cartilage

The increased interest in using iPSCs for TE approaches, including cartilage regeneration, brings to play multiple biochemical and physiological stimuli that can promote chondrogenesis in the cell's pluripotent state. In the embryo, the formation of cartilaginous anlage happens from the condensation of mesenchymal progenitors. This condensation requires hypoxic environments, making oxygen concentration an essential parameter in a bioreactor environment. Studies using hypoxic conditions (O2 5%) have demonstrated enhanced chondrogenic differentiation [36, 38, 62].

Other conditions, like temperature and pH level, can be easily incorporated into bioreactors simultaneously. Electrical-resistor heat dissipation systems with feedback control are typically used to keep TE constructs under constant temperature. The human body is a system that maintains an average temperature of approximately 37 °C. However, in healthy and osteoarthritic conditions, articulations like the knee can experience a temperature variation [63]. Therefore, TE applications for this particular case should target in vitro temperature culture conditions using bioreactors.

Previous research on articular pH showed that synovial fluid (SF) is the major interactor of the articular cavities. The SF, which keeps an average pH of 7.4, delivers by diffusion nutrients and oxygen necessary to the correct function of articular cartilage [33,34,35]. For iPSCs culturing and differentiation towards chondrocytes, a defined chondrogenic medium (CDM) can supply certain nutrients and growth factors frequently used to stimulate the growth and preservation of tissue [39,40,41]. An advantage of using commercially available basal mediums is maintaining physiological pH levels under 5–10% CO2 conditions [60]. Table 1 summarizes the most important biochemical and physical stimulus for in vitro cartilage regeneration.

2.3 Mechanical stimulation for tissue-engineered articular cartilage

Another advantage of bioreactors for in vitro testing is developing an excellent rehabilitation program after cartilage implantation. By studying the effect of particular stimuli in the bioreactor, conclusions can be made about which stimuli positively or negatively impact specific diseases, as is OA [21].

Mechanical loading of the joints is an essential factor to study. Besides the growth factors and cytokines, the chondrocytes respond to loading by appropriately synthesizing the required molecules, such as type II collagen and proteoglycans [64]. The chondrocyte's mechanical loading mitigates joint destruction by downregulating matrix metalloproteinases (MMPs) and a disintegrin and metalloproteinase with thrombospondin motifs (ADAMTS) [65]. Besides the downregulation of MMPs and ADAMTS, mechanical stimulation increases aggrecan expression, the most abundant glycoprotein in the ECM [66]. The balance in ECM production in healthy articular cartilage tissue depends on a stable phenotype of the chondrocytes. Cartilage under cyclic (0.002–0.01 Hz) compressive strain amplitudes of 10% [26, 43] (corresponding to normal stresses of 0.5 MPa) and intermittent hydrostatic pressures of 1–10 MPa enhanced protein and proteoglycan biosynthesis [67]. Cartilage under constant 10% compressive strain and cyclic shear strain of 1–3% (corresponding to 10–20 kPa shear stress) at a frequency of 0.01 to 1 Hz has doubled protein synthesis, and 25% increased proteoglycan synthesis [26, 42]. A shear strain of around 1% promotes regenerative pathways for cartilage regeneration by promoting chondrocytes-ECM homeostasis [68]. In contrast, increased friction at the cartilage surface may cause much higher shear strains leading to chondrocyte catabolic activity.

Chondrocytes are phenotypically unstable. Therefore, a lack of mechanical loading can result in differentiation to other cell phenotypes, which induces different ECM proteins' production, leading to low-grade tissue mechanical properties. This phenotypic instability is a significant problem in OA development and a characteristic considered in TE. Appropriate mechanical stress on chondrocytes can give the cells valid signals for producing the components necessary for the load-bearing of the ECM [65].

Although a balance between too much and too little mechanical loading is relevant for creating TE solutions combined with bioreactors, excessive compressive and shear load over the hyaline tissue can activate biological pathways that promote articular cartilage degeneration. However, cartilage degeneration accelerates without mechanical stimulation due to softening and thinning of the articular cartilage and decreased glycosaminoglycan content in the joint [69].

For mechanical stimuli (Fig. 3), the bioreactor should mimic the forces generated by the joints' multiaxial movements, simulating the kinematics and mechanical loading in vivo [29]. A continuously changing stress is applied to the articular cartilage for lower limb articulations. Thus, the amount of stress directly depends on the person's weight and other gait factors. People at risk of developing OA are most likely people suffering from obesity. Therefore, exposing the in vitro cartilage to high mechanical loads can be beneficial in investigating whether the iPSCs-engineered cartilage can withstand the ultimate mechanical load and identify which biological routes can be addressed for further research [70].

Fig. 3

Schematic representation of mechanical stimulation for tissue engineering constructs

Both static and dynamic compression have shown a chondrogenic gene expression increment compared to non-stimulated constructs. However, dynamic compression has the advantage of increasing chondrogenic differentiation markers in stem cells, as it resembles the function of articular cartilage during motion [71,72,73]. In dynamic compression, a uniaxial or multiaxial mechanical stimulus can be applied. Making the correct assumptions for these stimuli to mimic human articulations is challenging. Therefore, analyzing these parameters with models and computational tools is helpful instead of using a trial-and-error method [24].

An advantage of using the multiaxial is that it can better replicate the physiological conditions of cartilage, allowing a more accurate simulation of its natural growth and function. The multiaxial stimulus exhibits a higher expression of chondrogenic marker genes [26]. Furthermore, it also improves tissue functionality by promoting a broader range of mechanotransduction pathways, enhancing cell differentiation, ECM deposition, and tissue maturation compared to the uniaxial stimulus [74]. The enhancement of ECM matrix deposition is translated into an increase in collagen type II and proteoglycan (GAGs) deposition; both components can enhance the mechanical properties of the engineered constructs and facilitate tissue integration in case of implantation [75].

Mechanical forces play a significant role in forming joint cavities in embryonic skeletal development. Absence or restricted fetal movement and mechanical force exertion can result in multiple syndromes such as dystrophy and muscle atrophy [76]. Incorporating iPSCs into articular cartilage TE combined with bioreactors is a potential research field to identify mechanobiology routes expressed by the cells in the pluripotent state [23, 77].

2.4 Hydrostatic pressure for tissue-engineered articular cartilage

Hydrostatic pressure (HP) is a physical stimulus present in daily motion articulation activities due to a relationship between the proteoglycan aggregates and interstitial fluid. This relationship provides cartilage compressive resistance through negative electrostatic forces. The proteoglycan fixed charge determines the ECM ion composition. As this ion concentration is higher than in the synovial fluid, the difference results in EMC fluid intake creating HP [78]. In OA patients, the articulations are affected by changes in the quantity and quality of the synovial fluid. A mismatch of the fluid with daily activities can affect the native HP of the joints and, ultimately, the tissue homeostasis. Using controlled HP generated by bioreactors has increased in vitro cartilage ECM synthesis on engineered constructs. The presence of a 7–10 MPa HP showed faster ECM accumulation, prolonged cell survival, and increased chondrogenic gene expression[37, 44,45,46]. Furthermore, HP has been proven to increment the Collagen II and glycosaminoglycan (GAG) content on cartilage TE approaches [65]. Studies using HP bioreactors on PSCs lines have to consider that this physical stimulus can affect the oxygen solubility, thus, affecting the oxygen tension and ultimately forming cartilaginous anlage under hypoxic conditions [37, 46].

2.5 Electrical stimulation for tissue-engineered articular cartilage

As hyaline Cartilage is an avascular tissue, the synovial fluid is responsible for chondrocyte nutrition and maintenance. The cartilage's ECM has abundantly fixed ionized macromolecules that interact with the synovial fluid to create diffusion and strain electric potentials, resulting in essential signals for tissue homeostasis [47]. In OA, unbalanced mechanical forces are present in the articulation due to cartilage loss and synovium inflammation, and thus, the force mismatch harms the SF interaction over the articular cavity. The ECM disruption present in this pathology together with inflammatory proceses create an ionic imbalance of the cartilage environment, with the presumption of electrical field disruption affecting tissue maintenance and changes in the chondrogenic phenotype of the cells [47, 79].

Regenerative pathways for cartilage regeneration on iPSCs engineered constructs can be stimulated by endogenous electrical stimulation (ES, Fig. 4), which induces in vitro chondrogenesis and in vivo cartilage repair, which is demontrated by the results described below.

Fig. 4

Schematic representation of electrical stimulation with an interval and a continuous step signal input for tissue engineering constructs. Figure partially created with Biorender.com

After incorporating ES on stem cells, this stimulus generates ATP oscillations driven by Ca2+ oscillations. These ATP oscillations are crucial for prechondrogenic condensation and posterior expression of chondrogenic markers such as collagen II, aggrecan, and Sox9 [79, 80]. Furthermore, when applied in pulses, ES can modulate the cyclic adenosine monophosphate (cAMP) levels by activating adenosine receptors that promote chondrocyte proliferation and anti-inflammatory pathways [48].

Kwon et al. demonstrated that ES also drives ATP oscillations by cAMP modulation, leading to chondrogenic differentiation in the absence of exogenous growth factors [79]. To introduce ES in vitro, the bioreactor needs to mimic the electrical environment of the articulations. Although this stimulus is present permanently in the cartilage tissue, on in vitro cartilage regeneration or development using iPSCs, the stimulation must be controlled by frequency, duration, amplitude, and type of stimulation (Table 1). Additionally, the bioreactor's components to deliver the ES need to be specifically designed to avoid electrolysis, altering the pH of the medium and modifying cell differentiation. The design must consider the proper electrode placement to ensure the minimal contact with the iPSCs to enhance cell viability. Ideally, the electrodes must be made of carbon as it is widely known that this material hardly corrodes, minimizing the chances of electrolysis. Furthermore, any ES bioreactor intended for iPSCs applications must contain a proper grounding and an electrical current control system to prevent and rectify the system in case of electrolysis.