Electronic Activation and Inhibition of Natural Rubber Biosynthesis Catalyzed by a Complex Heterologous Membrane-Bound Complex

Abstract

Natural rubber biosynthesis is catalyzed by a unilamella membrane-bound heterologous complex with multiple different subunits (rubber transferase, RTase). Two substrates and divalent metal cation activators are required, and their concentrations affect biosynthetic rate and polymer molecular weight. Rate, molecular weight, and complex stability are highly sensitive to Mg²⁺ and Mn²⁺ concentration, but studies are challenging because methods to control ion concentration may dislodge the elongating rubber polymers from the RTase complexes, halting synthesis and producing low-molecular-weight polymer. Here, programmable chemical actuators (PCAs) are used to electrochemically control rubber biosynthetic rate and subsequent molecular weight in enzymatically active rubber particles purified from Ficus elastica (Indian rubber tree). RTase activity was assayed using ³H-FPP (initiator) and ¹⁴C-IPP (monomer). Since only one FPP molecular is needed to initiate a new rubber polymer, the ratio of incorporated ³H-FPP to ¹⁴C-IPP was used to calculate the mean molecular weight of newly synthesized polymers. PCAs exchange ions in solution through REDOX reactions which we show control cation concentration without dislodging the elongating rubber polymers from the RTase. PCAs demonstrated highly tunable control over monomer incorporation and molecular weight in both Mg²⁺ and Mn²⁺ cations. REDOX cycling PCAs did not irreversibly inhibit the rubber transferase complex, and no indication of enzymatic damage was observed. Precise PCA control of RTase activity may pave the way for rubber eventually to be produced in bioreactors.

1. Introduction

Natural rubber (cis-1,4-polyisoprene) is a commercially and strategically important commodity produced from Hevea brasiliensis trees grown as clones in tropical climates [1]. It is distinguished from synthetic rubbers by its source, high elasticity, impact resistance, efficient heat dispersion, ability to form crystallites under strain, and malleability in cold temperatures [1,2,3,4]. Natural rubber’s unique physical properties are tuned by the molecular weight of the polymer, but the mechanisms determining molecular weight in natural rubber synthesis are incompletely understood [2,3,4,5].

Natural rubber is synthesized at the surface of unilamella membrane-bound organelles known as rubber particles [6,7,8,9] and is compartmentalized to the particle interior [9]. All rubber particles have specialized rubber transferase protein complexes (RTase, EC 2.5.1.20) that catalyze the initiation and elongation of cis-1,4-polyisoprene [5] but which have species-specific differences [2,3,4,5,6,10,11]. Also, the composition of the rubber particle membrane is species-specific [6,7,10]. Regardless of species, RTase spans the rubber particle membrane and adds isopentenyl pyrophosphate (IPP) monomers from the cytosol outside the particle to the growing rubber polymer compartmentalized inside the particle via sequential condensation reactions [2,3,4,12].

RTase requires an allylic pyrophosphate (APP) initiator, IPP monomer, and a divalent metal cation activator for rubber synthesis [2,11,13,14]. RTases exhibit substrate affinity for a wide variety of APP initiators, but farnesyl pyrophosphate (FPP) is thought to be the principal initiator of rubber molecules in vivo because of its cytosolic availability [13]. In contrast, RTases exhibit no substrate affinity for the IPP monomer until it is complexed with a divalent metal ion in solution to form IPP-C²⁺ [14]. Due to its dependence on a metal ion for enzyme activity, RTase is classified as a metal-activated enzyme. The effects of Mg²⁺, Mn²⁺, Ca²⁺, and Cu²⁺ on IPP incorporation by Ficus elastica enzymatically active rubber particles [14,15,16] showed that 100 mM magnesium and 10 mM manganese maximized the incorporation of IPP by rubber transferase [14], whereas Ca²⁺ and Cu²⁺ were inactive. Magnesium and manganese cations also were the only divalent activators of Parthenium argentatum and Hevea brasiliensis RTases [17,18,19]. Magnesium ions also stabilized both RTase [18,19] and the rubber particles themselves [20].

When all RTases have bound an initiator (under nonlimiting concentration), incorporation of IPP (also under nonlimiting concentration) continues until monomers are no longer able to bind to the RTase for some reason (e.g., the lack of C²⁺) or the polymer is displaced by a new APP [2,12,13,14,15,16,17,18,19,20]. However, when all IPPs in solution are complexed with a divalent metal cation, RTase activity is inhibited by increasing [C²⁺] because excess metal ions displace IPP-C²⁺ bound to the RTase or block its binding [14,16,17,18,19]. The most likely rubber elongation mechanism is given by equation [19].

C²⁺_aq + IPP_aq ↔ (IPP∙C²⁺)_aq + E_rp ↔ (IPP∙C²⁺∙E)_rp ↔ (IPP∙E)_rp + C²⁺_aq

(1)

where C²⁺_aq is an activating divalent cation in aqueous solution, IPP_aq is an isopentyl pyrophosphate monomer in aqueous solution, (IPP∙C²⁺)_aq is a cation monomer complexed to an IPP monomer, E_rp is rubber transferase on a rubber particle, (IPP∙C²⁺∙E)_rp represents the cation complexed to an IPP monomer and the RTase all associated with the rubber particle, and (IPP∙E)_rp is the RTase complexed to the IPP substrate following cation ejection during the condensation reaction. The active elongating end of the rubber molecule is in the form of an allylic pyrophosphate.

Equation (1) indicates that the divalent cations complexed with IPP are concomitantly released into solution with the condensation reaction.

Enzyme activity, in general, is strongly influenced by the electrochemical conditions of the surrounding medium [21,22,23]. Traditional ways of adjusting electrochemical gradients—such as adding or removing ionic solutions with pipettes, pumps, or microfluidic devices—are labor intensive, difficult to scale, and challenging to integrate with other experimental setups because they demand precise, continuous control [24]. Optical strategies that use photoacids or photobases to generate and monitor pH gradients offer an appealing alternative [25], yet they are not universally applicable; many systems require unobstructed optical access or specific modifications to substrates or proteins to function effectively [22,24,26].

Electrochemical control of enzymatic reactions often relies on attaching the enzyme directly to an electrode surface, but this approach is unsuitable for many proteins, which may lose activity, undergo kinetic changes, or become completely inactivated upon immobilization [27]. The specific cation or cations required to support metal-requiring enzyme activity must also be considered [28]. Another strategy involves using redox mediators to shuttle charge between an electrode and the active site of a freely dissolved enzyme, though this technique is primarily employed in sensing applications rather than detailed kinetic studies [28].

Recent work demonstrated that the kinetics of soluble glucose oxidase can be probed using chronoamperometry in a three-electrode electrochemical cell equipped with an unmodified platinum working electrode, a platinum counter electrode, and a calomel reference electrode [29]. While this configuration is highly sensitive at low enzyme concentrations, it offers limited control over mass transport. Additional studies have shown that solvent choice can alter reaction driven actuation mechanisms, even reversing ion exchange behavior from anion- to cation-dominated transport [29]. It should also be noted that different metal-requiring enzymes often require different metals due to differences in their specific size, charge, and/or coordination chemistry [30]. We have developed a programmable chemical actuator (PCA), an electrochemical platform to spatiotemporally manipulate ion concentration without fluid exchange [31].

PCAs exchange free divalent cations in solution by reducing and oxidizing a dodecylbenzenesulfonate–doped polypyrrole working electrode. The conducting polymer is in the oxidized state when disconnected from an electrical source [32,33,34,35]. When reduced by an electric field, cations are drawn into the polymer to offset the charge imbalance caused by the additional electrons [36,37,38]. Subsequent oxidation of the polymer reverses the charge imbalance and drives ions back into solution according to Equation (2) [36,37,38,39].

$${nC}_{aq}^{2+}+{\left[PP{y^n}^{+}{\left(DB{S}^{-}\right)}_{2n}\right]}_{WE}+{2ne}^{-} \leftrightarrow {\left[\left(PPy\right){\left({DBS}_2^{-}{C}^{+}\right)}_n\right]}_{WE}-{2ne}^{-}$$

(2)

where $C_{aq}^{2+}$ represents activating divalent cations in solution, n is a stoichiometric coefficient, ${\left[PP{y^n}^{+}{\left(DB{S}^{-}\right)}_{2n}\right]}_{WE}$ is polypyrrole doped with dodecyl benzene sulfonate in the oxidized state, e− is an electron, and ${\left[\left(PPy\right){\left({DBS}_2^{-}{C}^{+}\right)}_n\right]}_{WE}$ is PPy(DBS) in the reduced state.

In this paper, we demonstrate electronic activation and inhibition of F. elastica RTase using Mg²⁺ and Mn²⁺ PCAs schematically depicted in Figure 1. The dependence of natural rubber initiation, elongation, and molecular weight on [Mg²⁺] and [Mn²⁺] was determined. These results informed the design of PCAs used to enact Mg²⁺ and Mn²⁺ ion concentration changes. The ability of Mg²⁺ and Mn²⁺ PCAs to control molecular weight was compared to static control mechanisms.

2. Materials and Methods

Unless otherwise specified, all materials and reagents that have been used in the work were purchased from MilliporeSigma (Burlington, MA, USA).

2.1. Latex Collection from Ficus elastica

Latex was collected from F. elastica using published methods [7,8,9,10,15,16]. Briefly, F. elastica plants were purchased from a local nursery and grown in a greenhouse. Leaves were removed at the apical portions of the petioles at a 30-degree angle to the base of the petiole using a scalpel. Latex was collected in 10 mL of chilled collection buffer (10 mM Tris-HCl at pH 7.5, 12% glycerol, and 2.5 mM MgSO₄) on ice in 50 mL centrifuge tubes.

2.2. Ficus elastica Washed Rubber Particle Preparation

Enzymatically active washed rubber particles were purified from collected F. elastica latex as previously described [7,15,16,40]. Collection tubes, filled to 1 cm from the top, were centrifuged at 2500× g for 10 min at 4 °C. The buoyant fraction of rubber particles was transferred to SS-34 centrifuge tubes (50 mL capacity) containing 40 mL Spin Buffer (10 mM Tris-HCl at pH 7.5 and 2.5 mM MgSO₄) and the sedimented fraction was discarded. Tubes containing the buoyant fraction were centrifuged at 2500× g for 10 min at 4 °C and the pelletized rubber particles were washed by discarding the supernatant, resuspending the pellet in Rinse Buffer (10 mM Tris-HCl at pH 7.5, and 2.5 mM MgSO₄), and then respinning at 2500× g for 10 min at 4 °C. Washing was repeated twice and the final pellets were suspended in 1 mL Rinse Buffer. Rubber particle concentrations were determined by drying small aliquots and then weighing them.

2.3. Rubber Particle Preservation

Glycerol was added to stock solutions to achieve a final concentration of 10% and the solution was gently mixed by inversion. Uniformly sized droplets of the solution were flash-frozen in liquid nitrogen and collected into precooled (−196 °C) cryovials [41].

2.4. Siliconization of Reaction Tubes

To begin, 1.5 mL microtubes (Fisher Scientific, Pittsburg, PA, USA) were placed into a tube holder and filled with Sigmacote. Tubes were incubated for 25 h with caps off in a chemical fume hood. Excess Sigmacote was decanted into a clean reagent bottle, and the tubes were inverted and left to dry for an additional 25 h with caps off in a chemical fume hood. Prepared tubes are now referred to as “siliconized tubes” and stored in the dark, at room temperature, with the caps on.

2.5. Rubber Transferase Reactions Under Static Metal Ion Control

RTase activity was determined by tracking the incorporation of radiolabeled initiators and monomers into newly synthesized rubber polymers [16,20]. In this work, frozen beads were thawed on ice, and 25 µg (dry weight basis) of rubber particles were assayed in siliconized 1.5 mL microcentrifuge tubes with 100 µL of reaction mixture containing 10 mM TRIS-Cl (pH 7.5), 20 µM ³H-FPP (American Radiolabeled Chemicals, St. Louis, MO, USA) 200 µM ¹⁴C-IPP FPP (American Radiolabeled Chemicals, St. Louis, MO, USA), and the concentrations of Mg²⁺ or Mn²⁺ indicated in the figures. Reactions were started by adding rubber particles to premixed solutions and incubating for 4 h at 27 °C. Reactions were stopped by adding 50 µL of 500 mM EDTA to the reaction tubes. Rubber particles were harvested by filtration, washed with 1 M HCl to protonate and solubilize any unincorporated IPP, followed with three washes with 100% ethanol to remove any protonated, unincorporated substrate, and the incorporated radioactivity was determined by liquid scintillation spectroscopy in EcoScint-A scintillation cocktail. The relationships between ¹⁴C-IPP incorporation and ³H-FPP incorporation and substrate concentration were linear (Supplementary Figure S1).

2.6. Electrode Fabrication

Reference electrodes were fabricated from 250 μm diameter Ag wire. The Ag wire was trimmed to 5 cm, straightened, and chlorinated for 12 h in a 15% solution of sodium hypochlorite. Following chlorination, the silver–silver chloride wire was rinsed with distilled water and stored in a 0.1 M solution of sodium chloride at 4 °C in the dark until use. Working and counter electrodes were fabricated from 500 µm diameter Au wire. The Au surface was mechanically cleaned using isopropyl alcohol and dried. The Au surface was electrochemically cleaned by performing a potential sweep from −200 mV to +1200 mV in 50 mM NaOH, rinsed with DI water, rinsed with IPA, and then dried under N₂. Following cleaning, the Au electrodes were stored covered, in air, at room temperature, in the dark, until use.

2.7. Programmable Chemical Actuator Fabrication

PCAs were fabricated to fit the small reaction volume used for RTase radiometric assays. A Ag/AgCl reference electrode was threaded through the middle of a borosilicate capillary (Sutter, BF150-86-10; 1.5 mm O.D.; and 0.86 mm I.D. (Sutter Instrument Company, Novato, CA, USA)) and secured using marine epoxy. The Au working and counter electrodes were secured to the capillary using PTFE heat shrink (SM2T-17, Component Supply Company, Sparta, TN, USA).

2.8. Polypyrrole Polymerization

Electro-oxidative polymerization of polypyrrole-doped dodecyl benzene sulfonate was performed in 200 mM dopant solution with 200 mM pyrrole monomer [31]. The polymerization potential was determined by the pyrrole monomer’s oxidation peak in the electrolyte. The monomeric oxidation peak was found potentiodynamically using cyclic voltammetry on an Ivium PocketStat (pocketSTAT1, Ivium Technologies, De Zaale, The Netherlands). The potential was swept from 0 V vs. Ag/AgCl to −0.8 V vs. Ag/AgCl to +0.8 V vs. Ag/AgCl at a scan rate of 50 mV/s. Potentiostatic control during polymerization was carried out using chronoamperometry until the total charge passed during chronoamperometry reached 167 °C for Mg²⁺ actuators and 100 °C for Mn²⁺ actuators, equating to an arial charge density of 0.5 C·cm⁻² and 0.3 C·cm⁻², respectively.

2.9. Polypyrrole Equilibration

Following polymerization, the electrodes were rinsed using distilled water and dried using nitrogen gas and equilibrated [33,38,42]. The electrolyte was exchanged for 500 mM equilibration solution. Mg²⁺ actuators were equilibrated in 500 mM MgSO₄ and Mn²⁺ actuators in 500 mM MnSO₄. The potentiodynamic control strategy for equilibration was carried out via cyclic voltammetry, with sweeping from 0 V vs. Ag/AgCl to −0.4 V vs. Ag/AgCl to +0.4 V vs. Ag/AgCl vs. Ag/AgCl at a scan rate of 50 mV·s⁻¹ until the charge moving into the polymer was within 10% of the charge moving out of the polymer, usually around 25 cycles.

2.10. PCA Operation

PCAs are operated under duplex pulse amperometry [31,33,35,39]. During the reduction phase, the applied potential is pulsed between partially reduced (−0.2 V vs. Ag/AgCl) and fully reduced (−0.4 V vs. Ag/AgCl) states. Similarly, the oxidation phase consists of pulsing between partially (+0.2 V vs. Ag/AgCl) and fully (+0.4 V vs. Ag/AgCl) oxidized states. The effects of adding EDTA [43] were assessed between chronoamperometry experiments by removing the PCA from the reaction volume, adding the indicated amount of EDTA, mixing by pipetting, replacing the PCA, and repeating the chronoamperometry experiment.

2.11. PCA-Controlled Rubber Transferase Reactions

RTase activity under PCA control was determined by the same methods as static metal ion reactions with two exceptions: a PCA was inserted into the reaction volume to control metal ion concentration, and the manufacturer’s cap was replaced with a rubber O-ring to prevent evaporation during incubation while allowing the PCA to access the reaction volume. For oxidation reactions, prior to the addition of rubber particles, reduction pulses were used to remove cations from the reaction medium. For sequential reduction and oxidation cycling experiments, the reaction was halted at the indicated time to assess the effect of the REDOX cycling. All other reactions were allowed to proceed for the full experiment duration before being halted.

2.12. Scintillation Counting

Liquid scintillation counts were performed on a Packard Tri-Carb 2100TR Liquid Scintillation Counter (Packard instrument company, Meriden CT, USA). Before use, the machine was calibrated with ³H, ¹⁴C, and ³²P standards. Samples were incubated with 10 mL EcoScintA (Life Science Products, Frederick, CO, USA) until the filter dissolved. Samples were then loaded into counting racks and separate ¹⁴C and ³H counting protocols were run back-to-back.

2.13. Molecular Weight Determination

The rate of ¹⁴C-IPP incorporation to ³H-FPP incorporation reflects the mean molecular weight of newly synthesized polymer [10,11,12,13,14]. The molecular weight of natural rubber, MWNR, was calculated according to

$${\mathrm{M}\mathrm{W}}_{\mathrm{N}\mathrm{R}}={\displaystyle \frac{(R_{14C·\mathrm{I}\mathrm{P}\mathrm{P}}+\alpha ·(R_{3H·\mathrm{F}\mathrm{P}\mathrm{P}}\left)\right)·{\mathrm{M}\mathrm{W}}_{\mathrm{i}\mathrm{s}\mathrm{o}}}{R_{3H·\mathrm{F}\mathrm{P}\mathrm{P}}}}+{\mathrm{M}\mathrm{W}}_{\mathrm{D}\mathrm{P}}$$

where R_14C·IPP is the incorporation ¹⁴C-IPP in nmol·g dry rubber⁻¹·4 h⁻¹, α is an initiator-specific constant (2 for GPP and 3 for FPP), MWiso is the molecular weight of isoprene, R_3H·FPP is the incorporation of ³H-FPP in nmol·g dry rubber⁻¹·4 h⁻¹, and MWDP is the molecular weight of the pyrophosphate end group on the terminal IPP molecule, which ranges from 174 to 176 Da depending on the ionization state; 176 Da was used for our calculations.

2.14. Statistics

Kolmogorov–Smirnov tests, normal distribution fitting, means, standard deviations, F-tests, and analyses of variances were calculated in MATLAB 9.9 using the statistics toolbox. Unless otherwise indicated all statistical tests were performed at the p = 0.05 significance level.

3. Results and Discussion

3.1. Rubber Biosynthesis Under Static Control

3.1.1. Rubber Initiation Is Unaffected by Divalent Metal Cation Under Static Control

The process of purifying rubber particles dislodges the rubber polymers initiated in vivo if the particles have flexible membranes like P. argentatum and H. brasiliensis and, consequently, all FPP incorporated by washed rubber particles in vitro is in newly initiated polymer [15,16,17,18,19,20]. F. elastica particles have stiff, waxy membranes [7,8,9] and some partially formed polymers are able to remain attached to the RTases [16] unless EDTA is added to chelate all divalent cations [16]. Under saturating conditions, FPP incorporation by F. elastica RTase was not dependent on [Mg²⁺] or [Mn²⁺] (Figure 2) as has been reported previously for H. brasiliensis and P. argentatum [13,18,19]. In this report, Mg²⁺-activated F. elastica RTase had a mean FPP incorporation rate of 3.8 ± 0.2 nmol·gdw⁻¹·4 h⁻¹ (Figure 2) similar to rates reported for H. brasiliensis and P. argentatum RTases [30,31,32]. Mn²⁺-activated RTases exhibited a significantly higher FPP incorporation rate of 5.9 ± 0.4 nmol·gdw⁻¹·4 h⁻¹ (p = 0.05) (Figure 2). Rubber biosynthesis was initiated more efficiently by Mn²⁺ than Mg²⁺ possibly due to the larger hydrodynamic radius of the Mn²⁺ ion, 83 pm, than the Mg²⁺ ion, 72 pm [28]. However, Ca²⁺ (hydrodynamic radius of 114 pm) and Cu²⁺ (hydrodynamic radius of 73 pm) failed to activate RTase activity [12]. The solvated radius of Cu²⁺ is similar to Mg²⁺, but Cu²⁺ exhibits multiple oxidation states in solution which may contribute to its poor performance as an activator of rubber biosynthesis. Divalent cations activate certain enzymes, but not others, when their specific size, charge, and/or coordination chemistry allows them to induce critical conformational changes, anchor substrates within the catalytic site, or/and stabilize charged intermediates in transition states. Selectivity is governed by complex interactions of enzyme, substrate, cations, and other ligands [30]. Thus, initiation is affected both by cation identity and cation concentration.

3.1.2. Rubber Elongation Is Unaffected by Divalent Metal Cation Under Static Control

Consistent with previous reports [10,13,14,18,19], we found that IPP incorporation into rubber by RTase was dependent upon [Mg²⁺] and [Mn²⁺] (Figure 3). IPP incorporation was significantly (p = 0.05) higher in Mg²⁺-supported reactions than in Mn²⁺ reactions, but the size of the standard deviations did not differ significantly between the two cations. The [C²⁺] that activates the highest IPP incorporation rate is referred to as [C²⁺]_MAX. In F. elastica, [Mg²⁺]_MAX = 100 mM and [Mn²⁺]_MAX = 10 mM (Figure 3). The IPP incorporation rate at [Mg²⁺]_MAX was 2.1 ± 0.4 µmol·gdw⁻¹·4 h⁻¹. Increasing the [Mg²⁺] concentration beyond 100 mM inhibited IPP incorporation, likely due to Mg²⁺ displacement of IPP-Mg in the RTAse binding site [14,16]. Similarly, the maximum IPP incorporation rate of 1.2 ± 0.9 nm·gdw⁻¹·4 h⁻¹ occurred at [Mn²⁺]_MAX and was inhibited by increasing [Mn²⁺] beyond 10 mM. It has been shown that C²⁺ can bind to the RTase without first being complexed to the APP initiators [14,16] supporting this polymer displacement action hypothesis. However, direct proof is not yet possible because the structure of the RTase enzyme complex is still only partially understood.

When [Mg²⁺] was between 0.25·[C²⁺]_MAX and [C²⁺]_MAX, Mg²⁺ activated RTase and led to increased IPP incorporation. At 7.5 mM and 10 mM, IPP incorporation rose with increasing [C²⁺]. At [C²⁺] > 1.25·[C²⁺]_MAX, Mg²⁺ caused a greater degree of inhibition than Mn²⁺. At 1.75·[C²⁺]_MAX, IPP incorporation was 16.4% of IPP incorporation at [C²⁺]_MAX for Mg²⁺-activated reactions and 45.8% of the IPP incorporation at [C²⁺]_MAX in Mn²⁺-activated reactions. However, there is a trend of progressive inhibition as [Mn²⁺] increased beyond 1.25·[C²⁺]_MAX.

As discussed in the previous section, the noteworthy different effects of the two cations are related to their specific interactions with the RTase active site, the two substrates, and the allylic pyrophosphate active end of the growing rubber polymer. The specific causes of these differences will likely only be elucidated once crystal structures of the RTase are possible.

3.1.3. Rubber Molecular Weight Is Dependent upon Divalent Metal Cation Under Static Control

Manipulation of the [Mg²⁺] in washed, enzymatically active, F. elastica rubber particles, in nonlimited APP and IPP, resulted in RTase activation and production of rubber polymers. Maximum molecular weight rubber was produced at [Mg²⁺] between 100 mM and 125 mM (Figure 4). The highest mean molecular weight of 37.7 ± 3.1 Mg·mol⁻¹ was synthesized at 100 mM Mg²⁺, but this did not differ significantly from the mean molecular weight of polymer synthesized at 125 mM, 35.1 ± 3.9 Mg·mol⁻¹ (note: Mg is the abbreviation for Mega gram, or one million grams). However, both concentrations of Mg²⁺ produced significantly higher molecular weight rubber than that produced at [Mn²⁺]_MAX (13.8 ± 1.1 Mg·mol⁻¹). The higher initiation rate and lower IPP incorporation rate observed in Mn²⁺-activated reactions produced lower-molecular-weight rubber polymers, in general.

The F tests show that the standard deviations of mean molecular weights differ significantly (p = 0.05) between Mg²⁺ and Mn²⁺ treatments. Yet the standard deviations of [C²⁺]_MAX for Mg²⁺ and Mn²⁺ treatments do not come from different distributions. This is likely caused by the underlying variance of FPP incorporation during biosynthesis, as the standard deviations from IPP incorporation did not differ significantly.

3.2. Programable Chemical Actuator Control of Rubber Biosynthesis

3.2.1. PCAs Change Ion Concentration Without Solution Exchange

Programmable chemical actuators were fabricated with maximum ionic capacitances of [Mg²⁺]_MAX and [Mn²⁺]_MAX. When DBS-doped PPy is electro-oxidatively polymerized, the charge consumed during polymerization correlates to the capacitance of the polymer [32,37,38,39]. A total of 167 µC PPy(DBS) was deposited for Mg²⁺ PCAs and 100 µC PPy(DBS) was deposited for Mn²⁺ PCAs corresponding to aerial charge densities of 0.5 C·cm⁻² and 0.3 C·cm⁻², respectively. Following polymerization, actuators were equilibrated to establish ion-selective transport pathways in the polymer and identify the REDOX potentials for amperometry using established methods [38,42].

PCAs are operated under duplex pulse amperometry because pulse parameters influence the mass transport and kinetics of solvated ions [44,45,46,47]. Figure 5a,b show the potential waveforms for the duplex pulsed potential control strategy used to control [Mg²⁺] in solution. Reduction potential was switched between a partially reduced state and a fully reduced state at a 50% duty cycle (Figure 5a). Pulse widths were 5 s each and pulse amplitudes were −0.2 V vs. Ag/AgCl and −0.4 V vs. Ag/AgCl, respectively. Oxidation pulses (Figure 5b) used the same duty cycle and pulse widths, with pulse amplitudes of +0.2 V vs. Ag/AgCl for partial oxidation and +0.4 V vs. Ag/AgCl for full oxidation.

The current response during reduction pulsing (Figure 5c) quantifies the flow rate of charge into and out of the polymer [47,48,49]. The ion-selective nature of PPy(DBS) ensures that only the target ion contributes to the polymer’s current response [33,36,37,38,50]. The current response to a reduction pulse (Figure 5c) is characterized by an initial spike followed by exponential decay. The spike is attributable to capacitive discharge and uncompensated solution resistance [48,49]. The exponential region is a combination of double layer formation and Faradaic charge [51,52]. The Faradaic charge is assumed to be equal to ½ of the ion ingress for a divalent ion [37,52,53]. The current response to an oxidation pulse (Figure 5d) mirrors that of a reduction pulse where the Faradaic charge is equal to ½ of the ion egress.

The PCA’s charge response to reduction pulsing (Figure 5e) correlates to the total ion ingress into the polymer [37,51,53,54,55,56]. Charge accumulated with continuing reduction pulsing. The first partial reduction pulse and the first full reduction pulse resulted in the highest current responses (Figure 5c) and the most charge ingress (Figure 5e). As time progressed and ions ingressed into the polymer, the double layer charged and slowed ion ingress [44,46,57]. Switching from full reduction to partial reduction discharged the double layer, and the reduced ion motive force allowed ions to migrate to the depleted areas in the electrolyte [44,46,57]. When the next full reduction pulse was applied, the charge response indicated that more ions were available for ingress into the polymer. The charge response to an oxidation pulse (Figure 5d) mirrors that of a reduction pulse where the double layer discharge facilitates ion egress.

To ensure no other chemical species in the electrolyte contributed to charge ingress into the polymer, EDTA was added to chelate Mg²⁺ and Mn²⁺. EDTA covalently binds to divalent cations, neutralizing their charge and preventing them from participating in many chemical reactions [55,56,57,58]. The hypothesis that introducing EDTA would prevent the divalent metal ions from participating in REDOX reactions with PPy(DBS) was confirmed by the cessation of charge ingress during reduction following EDTA injection (Figure 5e). Charge egress from the polymer was unaffected by addition of EDTA (Figure 5f).

3.2.2. Rubber Biosynthesis Under Reduction Control

The effect of removing Mg²⁺ or Mn²⁺ at different times during the rubber biosynthesis reaction was assessed (Figure 6). Reduction pulses were programmed prior to the start of the rubber biosynthesis reaction and continued for the duration of the reaction. The waveform of the reduction pulses initiated at 2 h (Figure 6a) shows that neither PCA reduction nor EDTA controls affected FPP incorporation. This is consistent with our findings that FPP incorporation is not dependent upon [Mg²⁺] or [Mn²⁺] under the conditions tested. However, the mean FPP incorporation rate in Mn²⁺ treatments, 3.6 ± 0.33 nm·gdw⁻¹·4 h⁻¹, was significantly higher than the 2.6 ± 0.14 nm·gdw⁻¹·4 h⁻¹ incorporated in Mg²⁺ treatments under reduction control. FPP incorporation did not differ significantly between static [C²⁺] control and PCA [C²⁺] control (see, also, Supplementary Figure S2).

Decreasing the concentration of both metal ion cofactors using reduction pulsing led to a significant decrease in IPP incorporation. Starting reduction pulses at 1 h and continuing every 5 secs for the duration of the experiment decreased IPP incorporation in Mg²⁺-activated experiments from 2.1 ± 0.4 µm·gdw⁻¹·4 h⁻¹ at [Mg²⁺]_MAX under static control to 0.521 ± 0.006 µm·gdw⁻¹·4 h⁻¹. Mn²⁺-activated treatments demonstrated a similar reduction in the IPP incorporation rate under the same control strategy (Figure 6e). When reduction pulses were started 3 h into the reaction, the IPP incorporation rate did not differ significantly between Mg²⁺-activated reactions (1.3 ± 0.9 µm·gdw⁻¹·4 h⁻¹) and Mn²⁺-activated reactions (1.3 ± 0.4 µm·gdw⁻¹·4 h⁻¹).

IPP incorporation in the EDTA treatments did not differ significantly from background ¹⁴C counts, indicating that chelated cations were unable to activate RTase. The dramatic reduction in current and charge responses from the PCA suggests EDTA chelation also prevented ion ingress into the PPy(DBS) working electrode. As expected, rubber polymer molecular weights closely followed the trends observed in IPP incorporation. Mg²⁺-activated treatments demonstrated a much lower IPP incorporation rate in the RED-4 h treatment (0.1365 ± 0.0004 µmol·gdw⁻¹·4 h⁻¹) than in the RED-3 h treatment (1.3 ± 0.4 µmol IPP·gdw⁻¹·4 h⁻¹) but only a slightly lower molecular weight 23.8 ± 0.8 Mg·mol⁻¹ than the 26.8 ± 1.0 Mg·mol⁻¹ in the RED-4 h and RED-3 h treatments, respectively.

3.2.3. Rubber Biosynthesis Under Oxidation Control

To investigate the effect of increasing [Mg²⁺] or [Mn²⁺] on rubber biosynthesis, a reduced conducting polymer was oxidized at different times during the rubber biosynthetic reaction (Figure 7). A representative waveform of the oxidation pulses initiated at 2 h is shown in Figure 7a. The current and charge responses mirrored those observed in the reduction regime for Mg²⁺ and Mn²⁺ PCAs (Figure 6). Consistent with static [C²⁺] control and PCA reduction control, FPP incorporation was unaffected by PCA oxidation and EDTA addition. The mean FPP incorporation rate under oxidation control was 2.7 ± 0.16 nmol·gdw⁻¹·4 h⁻¹ for Mg²⁺-activated reactions and 3.7 ± 0.41 nmol·gdw⁻¹·4 h⁻¹ for Mn²⁺-activated reactions. As indicated by charge egress and IPP incorporation, the capacitance of the polymer limited the [C²⁺] injected into solution by the PCA and prevented excess cations from hindering IPP-C₂ incorporation. During rubber biosynthesis the low initial [C²⁺] limited the amount of IPP-C₂ available to the RTase preventing IPP incorporation. Oxidizing the PCAs released cations from the polymer and increased the [C²⁺] in solution. As the [C²⁺] rose, more IPP-C₂ was formed in solution and IPP was incorporated into new rubber polymers. Time at [C²⁺]_MAX was key to producing high-molecular-weight rubber as demonstrated by 993.1 ± 20.8 nmol IPP incorporated ·gdw⁻¹·4 h⁻¹ for the OX-1 h treatment and 90.3 ± 1.7 nmol IPP ·gdw⁻¹·4 h⁻¹ for the OX-3 h treatment. Similarly, oxidizing the Mn²⁺ PCA increased the [Mn²⁺] correlating to increased [IPP-Mn] and IPP incorporation of 1.0 ± 0.1 µmol IPP ·gdw⁻¹·4 h⁻¹ for the OX-1 h treatment and 95.2 ± 4.8 nmol IPP ·gdw⁻¹·4 h⁻¹ for the OX-3 h treatment. As observed previously, trends in molecular weight correlated to those seen in IPP incorporation. The maximum molecular weight produced under oxidative control initiated at 2 h was 14.1 ± 0.8 Mg·mol⁻¹ for Mg²⁺-activated reactions and 10.1 ± 1.1 Mg·mol⁻¹ for Mn²⁺-activated reactions.

3.2.4. Rubber Biosynthesis Under Mixed Reduction and Oxidation Control

To test the reversibility and biocompatibility of PCA control, alternating reduction and oxidation regimes were carried out during the rubber biosynthesis reaction (Figure 8). The current responses, charge responses, and FPP incorporation followed the trends observed when reduction and oxidation were used independently (Figure 6 and Figure 7). IPP incorporation rates were 532 ± 27 nmol IPP ·gdw⁻¹·4 h⁻¹ and 538 ± 23 nmol IPP ·gdw⁻¹·4 h⁻¹ when the rubber biosynthesis reaction proceeded uncontrolled for 1 h at 100 mM Mg²⁺ and 10 mM Mn²⁺, respectively. Reducing the polymer for 1 h lowered [C²⁺] in solution, inactivating the RTase, demonstrated by an insignificant change in IPP incorporation in both cations tested (p = 0.05).

Reactivating the enzyme with C²⁺ via oxidation cycling resulted in 1 ± 0.1 µmol IPP·gdw⁻¹·4 h⁻¹ IPP incorporation for Mg²⁺-activated reactions and 1 ± 0.04 µmol IPP·gdw⁻¹·4 h⁻¹ for Mn²⁺-activated reactions. Taken together, these results indicate REDOX cycling programmable chemical actuators successfully controlled rubber biosynthesis without irreversible adverse effects on the rubber transferase complex. Thus, the time at [C²⁺]_MAX is the main determinant of molecular weight in vitro under PCA control. Also, PCAs exhibited linear control over rubber molecular weight (Supplementary Figure S4).

4. Conclusions

The molecular mechanisms behind rubber biosynthesis by RTase are incompletely understood. However, the RTase-mediated rubber polymerization reaction has been proven to be reversibly activated and inhibited by divalent metal cations. We found that reducing PCAs removed metal ions from solution and completely halted RTase activity. Under oxidation control, PCAs injected cations into solution and maximized IPP incorporation without surpassing [C²⁺]_MAX. Cycling the PCA between reduced and oxidized states had the desired “off–on” behavior without irreversibly inhibiting the RTase complex or dislodging the elongating rubber polymer from the RTase complex which does occur when EDTA is used to chelate C²⁺. Because PCAs regulate ion concentration without dislodging the elongating rubber molecule, they may provide a valuable tool to model ion regulation in vivo. Of particular interest is the mechanism behind molecular weight regulation, especially since some polymers produced in this study seem much larger than those produced in the latex of living F. elastica plants.

Further, PCAs can provide an experimental platform to advance the study of ionic effects on rubber biosynthesis in alternative rubber crops, microbial systems, and cell-free systems.