Keywords

1 Introduction

Biotechnological intercession has triggered tremendous interest in utilizing in vitro culture systems for the production of a variety of promising phytochemicals [1,2,3,4]. Majority of phytochemicals are plant secondary metabolites which are commercially important in the form of pharmaceuticals, nutraceuticals, flavors, essential oils, food additives, feed stocks, and antimicrobials. Till date, several biotechnological strategies have been adopted to produce these bioactive secondary metabolites by establishing in vitro cell and organ cultures [5,6,7,8]. However, in general, when compared to the intact plants, the production of these secondary metabolites remains at lower side in usual lab-scale cell and organ cultures. Alternatively, scale-up of an in vitro culture by replicating a lab-scale bioprocess in larger culture vessels as closely as possible to produce larger amounts of product may provide a solution for large-scale production of desired compounds [4, 9]. In vitro cell and organ cultures growing in liquid medium have a unique blend of physical and chemical culture environment and thus easily extend their use in up-scaling in bioreactors. In the same context, a bioreactor is a mechanical device that can simulate particular physio-chemical environment required for incessant growth and physiological activities of cultured cells and tissues. Besides increased working volumes, the semi or fully automated control system and round the year production efficiency that are independent of seasonal or climatic barriers are the additional advantages of up-scaling of an in vitro culture using bioreactors [10, 11]. However, the growth behavior of a biological entity not only depends upon their morphological and physiological status but also influenced by their physical environment. Therefore, successful operation of these bioreactors involve skillful implementation of engineering principles of designing the culture vessels in a way that can significantly influence the growth of cultured tissue at larger volumes [12]. The prime focus during the designing of bioreactor should be on adequate mixing of culture medium with minimized shear stress and optimized mass transfer with reduced hydrodynamic pressure.

Besides microbial cultures, the use of bioreactors has largely been focused on plant cell suspensions and hairy root cultures (HRCs) for secondary metabolite production [13]. Various reports on large-scale culture of cell suspension and hairy roots in various types of bioreactors are available [14,15,16,17]. However, genetic instability of cells and their higher sensitivity toward variability in culture conditions limit the broader use of cell suspension cultures for desired metabolite production. Further, many a times, the metabolite production is restricted to differentiated cell/tissues that too at their certain physiological age, and thus, in most cases, culture aging leads to consequent reduced production of desired metabolites. Therefore, keeping in mind the vast potential of HRCs for secondary metabolite production, the up-scaling of these cultures has shown striking opportunities in commercializing a bioprocess for desired metabolite production [18,19,20]. In up-scaling of hairy root-based production process, there are various factors which greatly influence the production. These include selection of reactor, optimization of culture conditions, automation and interpretation of synergistic effects of various conditions, etc., which needs judicious scientific attention. The upcoming text provides a descriptive account on bioreactor technology for up-scaling of HRCs for secondary metabolite production. The chapter also discuss the hitherto reports on up-scaling of HRCs of various plant species in terms of modifications in designing of bioreactors for incessant tissue growth concomitantly with metabolite production.

2 Hairy Roots as Potential Secondary Metabolite Production System

In the past three decades, hairy roots and secondary metabolite production became synonymous, and plentiful reports have came into existence which reveals various aspects of growth and production potential of HRCs [3, 20,21,22,23,24]. Hairy roots are disease expression of plants that are infected by soil bacterium Agrobacterium rhizogenes . The etiology behind the disease symptoms describes the stable insertion of bacterial T-DNA from extrachromosomal root-inducing (Ri) plasmid into the host plant genome and a consequent possible interruption in host secondary metabolism [25, 26].

Hairy roots being an easy “establish and explore” low-cost culture obligation have progressively materialized into full-fledged global technology for plant-based secondary metabolite production. Miscellaneous beneficial properties like multienzyme biosynthetic paraphernalia and close physiological and biochemical similarity with parent plant contribute to the preferential lead for choosing the genetic and biochemically stable HRCs over other in vitro systems for the production of variety of phytochemicals that is commercially important [21, 27]. In addition to such exclusive properties, hairy roots are also unique as in some cases they produce compounds that are not known in normal roots. The examples of production of glycoside conjugates of flavonoids in HRCs of Scutellaria baicalensis Georgi and sarpagine alkaloids from Rauwolfia serpentina are relevant to mention here [28]. Normal roots of S. baicalensis are known to produce glucose conjugates only. Additionally, sometimes HRCs are also known to produce compounds in higher quantities as compared to those that are present in normal intact roots [28].

A primary cause of the lack of in commercial production of secondary compounds using HR culture systems is their low yield. Several strategies, like elicitation, metabolic engineering, etc., that manipulate inherent property of hairy root tissues to enhance their production potential have been adopted to increase the production of bioactive compounds utilizing HRCs [20, 29,30,31]. Out of these strategies, the bioreactor up-scaling deals with the principle of large-scale metabolite production in proportion of high biomass growth. Succinctly, higher biomass means higher metabolite production [32]. Thus, keeping in mind high biomass production as the main objective, various types of bioreactors have been designed for the cultivation of hairy roots [32, 33]. A continuous culture in bioreactor can sort out the problems related to the manual handling of cultures and thus rescue the growing tissues from possible microbial infection that occur during regular subculturing. All kinds of bioreactors have heterogenous systems comprising of two or more phases of liquid, solid, and/or gaseous phase. The major confront during the operation is to maintain a homogenous condition which assures optimized mass and heat transfer among the different phases [34]. To deal with the situation, a typical bioreactor consists of various probes and devices to monitor and maintain different physical (temperature, flow rates and turbidity etc.) and chemical (dissolve oxygen, pH) conditions (Fig. 1). The following text discusses the challenges during the maintenance of optimized culture conditions for hairy root growth.

Fig. 1
figure 1

Schematic presentation of bioreactor configuration

Full size image

3 Major Challenges in Up-Scaling of Hairy Roots in Bioreactors

Although, HRCs have shown tremendous possibilities in secondary metabolite production at large scale; at the same time, there are certain challenges in using bioreactors for their up-scaling [34, 35]. Normally, a typical scale-up that starts in laboratory includes jars or shake flasks of 50–250 ml which further moves to 500 ml–10 L small-scale bioreactor. However, for large or industrial scale, stainless steel vessels of varying sizes (>10 L) are required (Fig. 2). It is well known that before going for up-scaling of any culture, optimization of process at bench scale is a prerequisite. However, at this stage, it is relevant to state that the results of bench-scale optimization cannot be directly transferable to higher scales. This is because a large vessel bioreactor may provide totally different culture conditions, and with the increase of vessel size, the homogenous culture ambience remains no longer effective. At this stage various physical factors like gas and liquid flow rates, mass transfer rate, concentration gradients, etc., start affecting the tissue growth simultaneously. These challenges are scientifically handled with the use of proper bioreactor technology that deals with the utilization of engineering principles and mathematical formulations to optimize a bioprocess [20, 36]. At laboratory bench scale, one can only check the practical feasibility of up-scaling of bioprocess in terms of desired productivity and production cost at industrial scale [35]. Various physical and chemical aspects during up-scaling of HRCs that are needed to be judiciously attended are described in upcoming text.

Fig. 2
figure 2

Gradual process scale-up of culturing hairy roots

Full size image

3.1 Aeration, Agitation, and Mixing

The O2 supplementation in reactor vessel is determine by the air flow supply. The basic objective of continuous O2 supplementation is to perform all biological activities by growing root tissue, while agitation, on the other hand, assures the homogeneity of liquid medium [37]. The aeration system consists of sparger, while for mass/heat transfer and uniform air distribution, agitator or impellors are required. Keeping in mind the shear sensitivity and bubble tolerance of growing tissue, normally, stainless steel spargers are used to aerate the bioreactor vessel [38, 39]. However, the bubble size and distribution of bubbles within in the liquid significantly influence the gas holdup and unit time for gaseous exchange or mass transfer by the cells in contact. Thus, a judicious attention to the bubble size and distribution improves the understanding of cell behavior under hydrodynamic pressure and mass transfer phenomenon [40]. For this purpose different types of spargers (porous, orifice, etc.) in combination with different types of impellers (turbine or marine blade, propellers, etc.) are used [41]. This designing of reactor vessel depends upon the morphology and growth pattern of hairy roots to be grown. One more point which needs attention during the selection of vessel design is vortex formation. To prevent vortex formation due to continuous mixing, baffles or metal strips attached on radial sides of vessel are used (Fig. 1). In an effort to develop a bioreactor for enhancing the biomass yield of plant organs, the HRCs of Hyoscyamus muticus were grown in a 5 L capacity bioreactor of which culture vessel was provided with an elongated baffle assembly [42]. The baffle not only prevented vortex formation but also facilitated rotary transport of radial flow of aqueous medium in the vessel. Similar reactor was later on successfully used to enhance the biomass yield of HRCs of Picrorhiza kurroa [43] HRCs, and multiple shoot cultures of Glycyrrhiza glabra [44, 45].

3.2 Root Morphology and Clump Formation

Hairy root cultures of different plant species exhibit differential morphology. Furthermore, the interclonal morphological difference is also significant in hairy root clones of a plant [46]. The randomness of insertion sites of T-DNA into host genome and the number of copies inserted collectively contribute to these interclonal morphological differences. The unique feature of hairy root growth is that they readily form clumps having densely packed rigid central core surrounded by loose highly entangled peripheral root mass [14, 47,48,49]. The dense mass of root clumps endorses bubble coalescing and channeling, poor liquid mixing, and ultimately the localized liquid stagnation. This hampers the continuous air and nutrient flow to the tissues growing toward the center by the resistance provided by peripheral root mass. This creates a phase difference between liquid (medium) and solid (tissue clump) components of growth matrix. The poor localized intraclump penetration of liquid flow greatly impose nutritional and concentration gradients near and within the clump [50, 51]. Such deficiency may lead to the cell necrosis and death at the clump center. These kinds of limitations can be overcome by applying controlled agitation rate and aeration with appropriate impellers and spargers, respectively [52]. For example, porous polypropylene membrane tubing as a supplementary aeration device was used to directly deliver the O2 to Atropa belladonna hairy root clumps. This was found helpful in overcoming mass transfer resistance related with insufficient intraclump penetration of liquid current and improved the biomass production by 32–65 % higher than sparging only of air [53]. Earlier, in similar context of analyzing the effects of oxygen limitations on Atropa belladonna hairy root growth in 250 mL shake flask cultures, oxygen limitations are likely to affect biomass production and kinetic measurements in shake flask cultures of hairy roots [54]. These kinds of optimization and investigation studies at flask level paved the way for acquiring novel strategies for site-directed oxygen delivery into the zones of highest root density in large culture vessels of reactor.

The presence of root hairs also plays an important role in increasing mass flow resistance. Studies reveal that root hairs which improve nutrient uptake under natural conditions are observed unfavorable in liquid medium as they produce resistance to fluid flow and limit the O2 supply to entangled root masses. In bioreactor cultivation of H. muticus transformed root cultures, the root hairs were found as substantial contributor of high fluid flow resistance [55]. To deal with such morphological constraints, some studies proposed the idea of growing hairless roots. In such studies either a root hair mutant line was selected or the root hairs in a normal hairy root clone were chemically removed. In an experimental system, hairy and hairless transformed roots of H. muticus were developed to assess the inhibitory role of root hairs in mass transport. Hairless lines of the test plant were initiated using pyrene butyric acid (PBA) and phosphate. The mixing was significantly enhanced in hairless root cultures grown in a 15 L bubble column reactor. The growth rate of the hairless culture in the bioreactor was as much as 2.4 times greater than growth of the hairy culture under similar conditions. The improved reactor performance was reflected in greater biomass accumulation and respiratory activity. These results show that the root hairs – which facilitate nutrient uptake in a static soil environment – are detrimental to growth in a liquid environment due to their involvement in stagnating fluid flow and limiting oxygen [56, 57].

3.3 Shear Stress and Hydrodynamic Pressure

Effects of hydrodynamic stress on suspended cells and tissues have been the matter of investigation since the beginning of reactor technology. Shear stress is the condition which reflects the extra burden on cell caused by continuous agitation of medium, distribution and fragmentation of gas bubbles due to the stirrer, and bubble rupture at the liquid surface. Hairy roots are also sensitive to hydrodynamic shear. However, no generalized strategy to minimize the shear stress has been concluded as shear stress on hairy roots of a plant species is supposed to be collectively governed by certain factors like tissue morphology, age of culture, aeration and agitation speed, viscosity of liquid medium, etc. Therefore, not only the designing of reactor vessel but also the selection of suitable root line and composition of medium may reduce the shear stress during culture. Although various practical strategies to minimize the shear stress and improve shear tolerance have been discussed for up-scaling of animal cells, for HRCs this aspect has been less explored [58, 59]. Furthermore, mathematical formulations have also been explored to calculate and optimize the shear rate in a continuously operating vessel. Nevertheless, for HRCs and their up-scaling in bioreactors, such studies are sparse and yet to be explored [60].

3.4 Mass Transfer

In multiphase system of bioreactor vessel, mass transfer takes place over phase boundaries. Computational analysis of this mass flux enables to assess the mass distribution in different phases over time and space in the system. The purpose of such computational analysis is to understand the growth behavior of roots and further design as well as control the process [61,62,63]. The efficient mass transfer between roots and their growth environment may define the growth characteristics of roots. In a case study of Tagetes patula hairy roots, methodical testing was performed in order to determine the relative significance of gas-liquid and solid-liquid mass transfer [64]. In shake flask cultures, the biomass growth largely depends upon the volume of medium and shaker speed. Taking an insight, this dependency of root growth indicates the importance of mass transfer in relation to different volumes of culture medium and shaker speed. Such calculation for a specific hairy root line also makes the basis for the selection of suitable bioreactor for up-scaling of that root line.

The gradient in chemical composition of phase boundaries, which occur due to continuous utilization of O2 by rapidly growing tissue, is responsible for the driving force for mass transfer. Several studies have been performed to investigate the O2 requirement and mass transfer phenomenon in HRCs [21, 54, 65]. Taking A. belladonna as model hairy root culture, convection was found to be the dominant mechanism for mass transfer in dense areas of root growth [65]. The mass transfer phenomenon was observed through local dissolve O2 level and rate of O2 uptake at the vessel areas where roots are loosely and/or densely packed in clumps. Specific growth rate and an exponential increase in root growth in terms of root length and number of growing root tips were observed at oxygen tensions between 70 % and 100 % air saturation. As the O2 supplementation decreases by 50 %, the root growth got negatively affected specially in the areas where root tissues have formed dense clumps. Due to the dense entangled mass, the air saturation at such points decreased by half or even low when compared to the areas where roots are loosely packed [65]. Such results indicated mass transfer resistances near the gas-liquid and liquid-solid boundary layer which dominantly affect the oxygen delivery to the growing hairy roots.

In the past few years, computational and mathematical simulations have been used to model the mass transfer behavior in bioreactors for hairy root growth [66, 67]. In a study, a computational fluid dynamics (CFD) model was developed to simulate the hydrodynamics and oxygen mass transfer in hairy roots growing in bioreactor. The CFD model predicted increase in mass transfer rates in root clumps that are stimulated by ultrasound. The model predictions were validated through experimental results in which increased O2 transfer was observed. This increased O2 was correlated with increased membrane permeability of root tissues by ultrasound stress [67]. Similarly, in another study a mathematical model was developed to investigate O2 transfer kinetics in Azadirachta indica hairy roots [68]. The model simulates and predicted the onset of O2 transfer limitation in dense intraclonal areas which eventually indicated the need of increased O2 supply to prevent the subsequent inhibition in growth of the hairy root biomass due to oxygen transfer (diffusional) limitation. Thus, computational/mathematical simulations of mass transfer phenomenon in a hairy root-based bioprocess may help in monitoring and controlling the process.

3.5 Optimization of Process: Medium Components, Temperature, Light, pH, and Inoculation

Chemical composition of nutrient medium directly influences the biomass yield and productivity of metabolites in HRCs. Therefore, studies on the effect of preferred key medium components on tissue growth, as well as product accumulation for incessant productivity, are a primary requisite. Further, this is also crucial to estimate secondary metabolite production as conditions suitable for growth may sometimes adversely affect the product formation and vice versa. Besides the conventional method of growth medium optimization which includes the alteration in various components individually and observing their effect on tissue growth, nowadays for the optimization of various growth conditions, the computational modeling is observed as a flexible strategy [69, 70]. Recently, the use of artificial neural networks (ANN) alone or in combination with other mathematical concepts was proposed to optimize various physical and chemical culture conditions for desired productivity in HRCs [70]. These studies pave the way for comparisons and simulations of similar bioprocess at larger levels involving bioreactors. With particular reference to the scale-up cultures using large reactor vessels, the computational simulations of growth under varying chemical conditions provide results mostly near to accuracy. Besides, this also helps to overcome the cost and time overflow of the conventional optimization. Spatial variations in culture temperature and medium pH lead to variations in physiological performance of the cultured roots. Continuous effect of light and temperature variation on growth and metabolite production has been observed in hairy roots of Catharanthus roseus, Artemisia annua, Echinacea purpurea, and Panax ginseng [71,72,73,74,75]. In P. ginseng hairy roots growing in 5 L bioreactor setup, biomass accumulation and ginsenoside production were found highest under red light fluorescence [75]. Therefore, optimization of the physical parameters for best results of hairy root up-scaling in bioreactors is as important as other factors. Though rather much explored in microalgae culture, the present hype in hairy root biotechnology has shown the viewpoint of using photobioreactors for large-scale culture of light-sensitive roots [76]. In case of light-sensitive roots, formation of clump leads to uneven distribution of light. Therefore, the basic principles for developing a reactor for such root cultures must include efficient and quantitative understanding of both mass and light transfer. On the other hand, the temperature of the liquid medium can easily be adjusted by circulating water in a chiller jacket outside the glass vessel.

Another aspect of process optimization that requires proper attention is the inoculation of tissue for scale-up. This not only includes optimization of inoculum density-media volume ratio and the selection of explant material of desired physiological age but also the aseptic procedure through which transfer of tissue from smaller culture container to reactor culture vessel is done. The former is done by subculturing the hairy roots in shake flasks and allowing them to grow till their logarithmic/exponential phase begins. At the onset of exponential phase, the root tissues that are well adapted to fast growth are aseptically transferred to the reactor vessel through the inoculation port of the bioreactor. Expert handling and incessant maintenance of aseptic conditions are the prerequisites of the inoculation procedure which is done under sterile laminar hood. Sometimes, as an extra precautionary measure during inoculation, the inoculation port is optimally flamed with the sterile cotton swabs soaked in alcohol [43]. Further, in an experiment, alginate-encapsulated R. serpentina hairy root tips were used as inoculum for the reactor vessel (unpublished results). This made the inoculation procedure a bit easier as simple pouring of small amount of liquid medium containing beads was required rather than transferring the whole root tissue using sterile forceps. However, pragmatic results of a detailed study on growth performance of alginate-encapsulated hairy root tips of R. serpentina were behind the idea of using encapsulated tips as inoculum of bioreactor [46]. Thus, a prior knowledge of growth performance of alginate hairy root tips of other plants is required before using them as explants for inoculating the bioreactor.

4 Overcoming the Challenges: Designing of Bioreactor for Hairy Root Cultures

By definition, a bioreactor is a culture vessel used for any biological conversion in terms of cellular growth and related productivity. These conversions are either in the form of cells/tissues/organisms cultured in a set of defined conditions or chemical compounds that are converted or transformed through specific metabolic reactions by mediation of a biological entity. The only difference between bioreactors and conventional chemical reactors is that the former is specifically used to culture biological entities. Another term that is used parallel to bioreactor is “fermenter” which is strictly used for anaerobic processes. Bioreactor designing is an engineering practice which inroads to a field that is known only for biology-based phenomenon. Thus, bioreactor designing characterizes an amalgamation of engineering principles of designing and analysis into growth and production processes of biological entities. The key issue in designing and operation of bioreactor is to control a biochemical phenomenon for a defined period in a consistent optimized way to get incessant maximum productivity. This optimization can be done on two scales; first, the biological entity and its products which include physiological basis of cell line/clone selection , metabolite synthesis, and accumulation, etc. The second scale of optimization includes physical parameters of culture vessel like temperature, pH, air supply, medium continuity, product removal, etc. The second scale basically includes designing a reactor vessel in such a way that maximum physical parameters of culture remain optimized and function in equilibrium to result in desired and consistent productivity. Several reports on cultivating hairy roots in large culture vessels endorse the idea of designing to facilitate adequate O2/nutrient supply during the entire culture duration [32, 33, 77].

On the basis of mode of operation, the bioreactor culture can be batch culture, continuous culture, and semi-continuous culture. Selection of bioreactor operation for hairy root cultivation depends upon various objectives. With particular reference to the HRCs, the bioreactor technology follows the concept of proportionate production of biomass and respective metabolites and, thus, can also be considered as a potential yield enhancement strategy [20]. Initially, Rhodes et al. [47] published a successful report on Nicotiana rustica HRCs in a bioreactor for nicotine production. However, during the past two decades, the bioreactor technology for HRCs has significantly uplifted from laboratory bench scale to industrial scale as various companies are now ready to adapt modified versions of technology to cultivate hairy root biomass for metabolite production [16; ROOTec bioactives Ltd, Switzerland; http:// www.rootec.com; CBN Biotech, South Korea]. Nevertheless, the key challenge in commercialization of bioreactor technology for HRCs is the low productivity of cultures and overall cost of technology [16, 48]. Persistent inflow of ideas can be observed in various reactor configurations that have been advised time to time to overcome limitations of low productivity like shear stress, heterogeneity, mass transfer, and nutrient uptake [32]. A detailed account on various types of bioreactors designed to cultivate HRCs has been given in Table 1 and Fig. 3. Such designing of reactors vessels is made to achieve the major objectives like its suitability to the morphology and physiology of tissue to be grown, adequate optimized culture conditions for maximum productivity, and finally the cost of the procedure. The reactors used to cultivate hairy roots can be divided into liquid phase, gas phase, and hybrid reactors.

Table 1 Comparative account on various major types of bioreactor used for up-scaling of hairy root cultures
Full size table
Fig. 3
figure 3

Schematic presentation of various types of major liquid (red background) and gas phase (blue background) bioreactors used for up-scaling of hairy root cultures

Full size image

4.1 Liquid Phase/Submerged Reactors

In liquid-phase reactors, HRs are allowed to grow throughout in liquid ambience comprised of growth medium. As the roots always grow under submerged conditions, liquid-phase reactors are also known as submerged reactors. Stirred tank (STR) and air lift (ALR) are the simple submerged reactors which are used to grow hairy roots. The STRs consist of impeller to facilitate mass transfer to the growing root tissue. In an earlier study, STR was used to grow HRCs of Datura stramonium [78]. However, the major challenge in these reactors is shearing and wounding of root tissues due to continuous rotation of impeller blades. Besides, in compactly grown root culture, poor liquid circulation and chemical gradients also contribute to the insufficiency of these reactors [79]. To deal with such challenges, the use of steel mesh or any porous substrate made up of nonreactive material was proposed [78]. HRCs of Glycyrrhiza glabra, Rauwolfia serpentina, and Picrorhiza kurroa were successfully grown in these modified STRs where a nylon mesh (pore size 200 μ) was provided to avoid submergence of inoculated tissue and its damage from the impeller [43, 44, 46]. Further, the idea of using of reactors without any mechanical agitation came into existence, out of which air lift and bubble column reactors are the most common examples. In air lift reactors, at the bottom of the vessel, aerators are provided that supply compressed air moisten with liquid medium at slow rates. The air lift reactors were used for HRCs of Lithospermum erythrorhizon, horse radish, and carrot [48, 81]. Further, Solanum chrysotrichum HRCs were also grown separately in a 2L air lift reactor with basic design and a novel modified mesh-draught reactor [82]. The tissue growth and distribution within the vessel was found more desired in later as in continuous operation, the insufficient mixing and nonuniform nutrient supply to the rapidly growing root tissue mass in air lift reactors led to the limiting factors for growth.

The much similar to air lift, bubble column reactors (BCRs) are designed to provide air in the form of bubbles to the liquid medium where the roots are submerged. In these reactors the rate of gaseous supply increased gradually with the tissue growth. Several studies have been performed on growing hairy roots using bubble column reactors with little attachments and modifications. For example, in the case of Tagetes patula and Lithospermum erythrorhizon HRCs, the division of bubble column into two segments with sparger in each segment was done to increase the mass transfer as a whole [83, 84]. Successful use of bubble column reactor is also reported for betalain production from HRCs of Beta vulgaris challenged with different elicitors [85]. However, the betalain productivity was hampered because of its inhibitory action on biomass. In an extension of this, cetyl trimethyl ammonium bromide (CTAB) was used for pigment release which resulted in optimization of concentration for better efflux of betalains without showing any inhibitory effect on hairy root viability. These studies on product enhancement and online extraction of pigment are useful for developing a bioreactor system for specific hairy root cultures. In particular the use of elicitors and efflux studies provide an insight for integrating unit operations and developing a process for continuous operation and higher production of phytochemicals.

Acquiring the concept of mass transfer based on convective fluid flow, convective flow reactors (CFR) were designed for HRCs of Hyoscyamus muticus [55]. The reactor consisted of a stirred tank along with a separate tubular culture chamber for root growth. The medium circulate through the two chambers at a velocity sufficient to overcome the flow resistance of the dense root bed. In a comparison with BCR, the H. muticus hairy root growth in CFR exhibited enhanced biomass production [55]. Proportionate to tissue growth, the sufficient medium supplementation at optimized velocity contributed enhanced convective mass transfer subsequently led to the enhanced growth. In other studies, to assess the effect of liquid flow velocity on the growth, oxygen uptake and productivity of hairy roots single column CFR were developed for Beta vulgaris [86]. In this study, the optimal range of velocity for good growth was observed 15 mh−1 and the increased flow velocity up to 28 mh−1 though enhanced the root elongation rate, but the viability of root tips was reduced due to damage from shear stress.

Based on this result, a radial flow reactor was developed to minimize the pressure on root tissue surface created by the fluid flowing in at a specific velocity. A substantial increase in biomass was observed in this reactor configuration where air-saturated medium flow in at a flow velocity 15 mh−1 through the ports on the sidewall of the reactor and flow out through the ports at the center of the top and bottom plates. Likewise, Atropa belladonna and Solanum aviculare [63] HRCs were also grown in packed bed recirculation reactor to analyze the liquid solid hydrodynamic layer and oxygen uptake. In the experiment, the liquid medium was flown in high velocity to minimize the hydrodynamic boundary layers around the tissue surface. However, it was observed that the rate of oxygen uptake and growth by the roots was dependent on the tissue mass, and very thin or the absence of hydrodynamic boundary layer did not affect the growth. It was proposed that under submerged conditions, roots were covered with thick layers of hydrated mucilage which acts as an additional barrier to oxygen transfer.

Keeping such observations in mind, the idea of cultivating hairy roots in alternate fill and drain cycles of liquid and gas phase has came into existence. To materialize this concept, rotating drum (RDR) and ebb and flow reactors (EFBR) were developed [48, 49, 80]. Rotating drum consists of a drum-shaped vessel mounted on rollers for rotation at low speed of 2–6 rpm. The difficulty observed with the rotating drum reactor was that at the beginning of culture, roots did not adhere to the vessel wall. Consequently, as the roots were rotated above the culture medium, they detached from the wall and were damaged ultimately resulting in low productivity. For the cultivation of carrot hairy roots, Kondo et al. [80] used polyurethane foam to immobilize the roots to overcome this problem. Though such modification had solved the challenge up to much extent, the RDRs were found unsuitable for industrial scale as they require high energy consumption [87]. On the other hand, the EFBR derived its name from the process behavior of liquid medium in reactor which follows repetitive cycles of ebbing and flowing (filling and draining) between the reactor vessel and medium reservoir [51, 88, 89]. Designed for H. muticus HRCs, the EFBRs demonstrated successful scale up to 50× of a 50 ml flask. In a comparative study with bubble column reactors, the EFBRs were found superior in terms of productivity, and liquid holds up profiles by dense entangled root mass. The study concluded that it was back and forth convective flow of EFBRs that made it superior to bubble columns which have negligible fluid convection.

Based on the concept of temporary contact between the cultured tissue and liquid medium, temporary immersion systems (TIS) were also used for the cultivation of hairy roots. These systems allow the cultured tissue to be immersed in liquid medium for a defined duration (flooding) followed by standby stage. Basically, TIS are designed for in vitro plant propagation; however certain advantages of these systems, like ease of medium changes, limited shear damage, less chances of hyperhydration of tissue, etc., have attracted researchers to use them for hairy roots [90]. In a study of Beta vulgaris HRCs, RITA® (Recipient for Automated Temporary Immersion System) apparatus was used to grow hairy roots for betalains biosysnthesis [91]. The maximum accumulation of betalain pigment was observed with 15 min immersed/60 min standby cycles, whereas optimal biomass in terms of intensively branched and morphologically healthy roots was obtained with 15 min immersed/75 min standby cycles. Thus, this was observed as an advantage of TIS in root growth in comparison to the ALR, STR, and bubble column reactors where clump formation and shear damage followed by nutrient and oxygen limitations in root tissues at central core and callus formation, respectively, are associated with loss of productivity. Forced and repeated air replenishment in the system and lesser hydrodynamic pressure to the growing tissue are supposed to be the reasons behind optimal growth of the root tissues in these systems. However, the success of using temporary immersions depends upon (1) the ratio of inoculum density and medium volume that too in relation with the size of container and (2) the optimization of length and frequency of medium immersions. In TIS, the time and frequency of the immersion are the most decisive parameter for optimized productivity, as they influence nutrient and water uptake. Thus, a prior optimization of such parameters is a must requisite. Another point of limitation that would be in need of attention is that, normally, TIS are smaller in size with small interior space which may prove insufficient for fast-growing root tissues.

4.2 Gas-Phase Reactors

In gas-phase reactors, the roots are immobilized in culture vessels with the help of horizontal sheets or rings made up of inert material and exposed to air or a gas mixture. The liquid medium is provided in the form of spray or mist of micron-sized droplets. In such reactors liquid is the dispersed phase and gas is the continuous phase in which roots are exposed. Such setup diminishes many of the limitations associated with liquid-phase reactors. Out of various gas-phase rectors, trickle-bed reactors (TBR), nutrient mist rectors (NMR) , and droplet phase are the common reactors. Since the medium is supplied in the form of droplet, there is a considerable variation in the size of droplets according to the requirement, and various attachments, like ultrasonic transducers, spray nozzles, etc., are used to create perfect sized droplets ranging from 0.5 to 50 μm [62, 92, 93]. In trickle-bed reactors, the medium trickles over roots from the top of the vessel. The used up medium is drained from the bottom of the vessel to a pool and is recirculated at a specific rate. The flow of liquid depends upon gravity and the distribution of liquid depends upon the mechanism of liquid delivery from the top. The growth of H. muticus hairy roots in gas-dispersed reactor exhibited excellent performance and accumulated tissue mass in submerged air-sparged reactors was 31 % of liquid-dispersed controls [49]. A study concluded that the distribution of the roots becomes a key factor in controlling the rate of growth. Noticeable results for hairy root growth, fluid dynamics, and oxygen mass transfer in a trickle-bed reactor were also obtained from the study conducted by Ramakrishnan and Curtis 2004 [94]. These results demonstrate that trickle-bed reactor systems can sustain tissue concentrations, growth rates, and volumetric biomass productivities substantially higher than other reported bioreactor configurations. Mass transfer and fluid dynamics are characterized in trickle-bed root reactors to identify appropriate operating conditions and scale-up criteria. Bioreactor characterization is sufficient to carry out preliminary design calculations that indicate scale-up feasibility to at least 10,000 L. On the other hand, in mist reactors, mists are more water efficient, thus, eliminating the need for extensive recirculation equipment. In comparison to large droplets of spray, provision of mists facilitates nutrient and gaseous exchange by reducing the thickness of the liquid film deposition on the surface of the root tissue [62, 95, 96]. Irrespective of spray or mist, the liquid that forms thin layer on the tissue surface acts as a barrier for efficient mass transfer [96, 97]. On the other hand, roots in a mist reactor are often too sparsely packed to capture mist particles efficiently and cannot, therefore, meet the nutrient demands required to maintain high growth rates. Keeping this in mind, an aerosol model was proposed for Artemisia annua hairy roots growing in NMR [98]. Growth rate was increased when mist medium containing high sucrose concentration was used in which sucrose acted as aerosol particles. In another recent study, an ON/OFF strategy was analyzed for optimizing the operating conditions of a mist reactor for the growth of hairy roots [99, 100]. A mathematical model was developed to optimize the ON/OFF mist duty cycle for the specified growth of hairy roots. Considering the availability and rate of transport of nutrients to the roots as vital parameters for growth, the ON/OFF cycles were proposed as the thin liquid film which continuously builds up during the ON cycle over the root surface is a key limiting factor mass transfer. However, the same film can also act as a finite reservoir of nutrients in the absence of any replenishment during the OFF cycle.

Since the continuous phase is gas, the roots are required to be immobilized in the reactor. For the immobilization of roots, various equipments like horizontal or vertical mesh sheets or rings made up of nylon or stainless steel [63, 101,102,103,104,105] were used. Ramakrishnan and Curtis [61] used Intalox metal process packing elements to immobilize roots in gas-phase reactors.

4.3 Hybrid Reactors

Similar to liquid-phase reactors, in gas-phase reactors the major problem includes liquid channeling and holdup within the root bed [61]. The stagnant holdup of liquid may have different nutrient levels than the bulk fluid [106] or make the roots within effectively submerged and depleted of oxygen [107]. To overcome these limitations, hybrid reactors for hairy root cultures are designed which follow the concept of growing roots under submerged conditions initially so that roots get adapted to the process and start growing. After an initial growth when submerged condition no longer remains effective due to dense growth and poor mass transfer rate, the reactor operation switches over to gas phase. Ramakrishnan and Curtis [61] proposed a combination of bubble column and trickle bed for Hyoscyamus muticus roots. In the study, bubble column was initially inoculated with roots to allow them to evenly distribute and attach to the anchorage provided in the reactor. Following 2 weeks of growth, when roots started forming clumps, the reactor was switched to a trickle-bed operation for another 4 weeks, thus exposing roots to a gas environment. In another study, Datura stramonium hairy roots were grown initially under submerged phase for 21 days followed by droplet phase for 40 days [93]. A hybrid reactor system made up of bubble column and nutrient mist bioreactor was used to study the transient growth characteristics and nutrient utilization rates of Artemisia annua hairy roots [108].

5 Conclusion

Hairy root culture technology has offered perspectives for in vitro large-scale production of valuable plant secondary metabolites. Together it also provides many challenges during large-scale cultivation as with the increase in medium volume, vessel size, and culture density, various physical factors, like gas and liquid flow rates, mass transfer rate, concentration gradients, etc., start affecting the tissue growth simultaneously. Besides, the unusual rheological properties and growth patterns of HRCs also contribute to challenges in their up-scaling. This has laid the background of acquiring and investigating novel approaches of bioreactor designing and process optimization. Optimized mass transfer concomitant with low shear stress and hydrodynamic pressure on growing root tissue is the utmost objective of designing a reactor for desired growth and productivity. The selection of right configuration of reactor vessel also depends upon the morphological and physiological properties of hairy roots. Further, a major issue of the commercial use of a bioreactor setup for hairy root cultivation deals with low capital cost. Keeping this in mind nowadays, bioreactor technology utilizes computational and mathematical simulations for incessant operation. These simulations provide help in meeting the challenges on issues like optimization of physical, biological, and chemical culture conditions, offline and/or online measurement of growth, mass transfer behavior, synergistic effects of various physical and chemical parameters on growth, downstream processing (intracellular/extracellular), product recovery, etc. However, online measurement of biomass growth is a difficult but perennial requirement during bioreactor cultivation. Precise measurement of growth is difficult due to the nonhomogenous nature of cultured tissue. Nevertheless, for offline measurement there are different methods like packed cell volume (PCV), fresh and dry weight estimation, etc. Growth estimation with these methods is relatively time consuming and also require frequent sampling from the running culture at a threat of contamination. Additionally, withdrawing large sample volumes may become problematic while measuring overall productivity against medium volume and inoculum density. To avoid these problems, online measurement is rather observed as an easy and noninvasive method of growth. These include utilization of artificial neural network and response surface methodology-based computational scheming of growth based on variables in culture conditions and predictions of final productivity [109]. Further, measurement of medium conductivity and osmolarity are the other methods that are being used for the growth measurement [110]. Moreover, online fluorescence measurement of metabolic activities of cultured tissues and online monitoring of cell growth by conductometry are the methods that cannot only predict the final biomass at the end of culture but also become helpful in optimizing various parameters during the culture duration [111, 112]. Although such measurement studies have been much explored in cell suspension and microbial cultures, an insight and proper retrieval of information from these studies may provide help in online growth monitoring and up-scaling of hairy root cultures in bioreactor. This may fairly help in filling the gap between capital cost and the benefits of technology at industrial scale.