Revisiting Escherichia coli as microbial factory for enhanced production of human serum albumin

Human serum albumin (HSA), synthesized in the liver, is the most abundant protein in the human plasma. The human protein is a 66.5 kDa, monomeric, single-chain, non-glycosylated, heart-shaped protein made up of three homologous domains comprising 585 amino acids, including 35 cysteine residues that participate in the formation of 17 disulfide bonds [1–3]. HSA is a prime determinant of plasma colloidal oncotic pressure, a universal transport and depot protein for many endogenous and exogenous metabolites and drugs while also displaying various important enzymatic, anti-inflammatory and antioxidant activities. Clinically, HSA has been employed to treat several diseases including serious burn injuries, hemorrhagic shock, hypoproteinemia, fetal-erythroblastosis and ascites caused by cirrhosis of the liver. HSA is also utilized as an excipient for vaccines, supplement in cell culture medium, the carrier of oxygen, nanodelivery of drugs and in various others biotechnological applications [1–4].

Because of its wide applications and immense therapeutic potential, annual world demand of HSA has exceeded 500 tons [5]. Currently, HSA is obtained primarily from the fractionation of collected human blood plasma [5]. Blood is a limited and an unsafe source for the production of human therapeutics. It possesses the risk of contamination by various blood-derived pathogens, like human immunodeficiency virus (HIV), hepatitis and many not yet identified, and eventually their transmission [5]. The scientific need is to eliminate the potential risk of such contamination. There exists an indispensable requirement to promote non-animal derived low-cost alternative methods for obtaining ample quantities of pathogen-free rHSA as a primary substitute for the plasma-derived form.

The non-glycosylated and single-chain polypeptide features of HSA make it less complex than the other blood-extracted proteins like plasminogen activator and clotting factors [3, 5]. These features have encouraged several investigators to produce HSA by using recombinant DNA technology and exploiting different host systems including microbial hosts, insect systems, transgenic animals and transgenic plants. Unfortunate-ly, none of the mentioned hosts have been successfully adopted for commercial production of rHSA due to their respective limitations. Transgenic animals [6, 7] and plants [8, 9] when exploited, required high capital, time consumption and yield was too low; exploiting mammalian expression systems is expensive resulting in very low yields. Consequently, several recombinant hosts such as different yeast strains like Saccharomyces cerevisiae [10, 11], Pichia pastoris [12, 13] etc. and insect systems [14], have been used for the industrial preparation of rHSA. These also have been associated with several problems including incorrect processing, poor export, hyper-glycosylation and improper folding. Moreover, it requires intense and complicated downstream processing protocols which make it impossible to produce rHSA in a cost-effective manner.

Of all the host systems, E. coli is one of the most convenient hosts which had contributed to the production of more than 30% of the FDA approved recombinant pharmaceuticals [15]. E. coli is a genetically and physiologically well-characterized organism which grows rapidly and reaches high cell density using inexpensive and simple subst-rates. The fermentation batch turnaround number for E. coli culture is 300 per year, which is far greater than any of the host systems available [16]. Therefore, E. coli derived recombinant products have more economic potential as fermentation processes are cheaper compared to the other expression hosts available.

Surprisingly, one of the production methods had remained less explored i.e. rHSA production using E. coli as a host. Despite all the mentioned advantages, E. coli has not been successfully adopted as a host for rHSA production.

The major difficulty in exploiting E. coli as a host for rHSA production is aggregation of this disulfide-rich, multidomain human protein during overexpression in the E. coli cytoplasm. Evidence had been presented that the majority of the expressed recombinant human serum albumin (rHSA) formed aggregates (more than 90% of the total expressed rHSA) leading to inclusion body formation in the reducing environment of the E. coli cytosol [17, 18]. Recovery of functional protein from inclusion bodies in vitro is a difficult, time-consuming, laborious and expensive task for such a multidomain and disulfide bond rich protein. Because of such limitations, E. coli as a host system was neglected for rHSA production. Therefore, designing advanced strategies to obtain protein in soluble form in vivo could bypass the tedious in vitro refolding strategies.

After considering the advantages of E. coli as a host, the present work has targeted E. coli as an alternate host system for rHSA production through resolving the major issue of inclusion body formation associated with it.

In the present communication, we have demonstrated the capabilities of E. coli as a host for the enhanced functional production of HSA. Parameters like intracellular environment, temperature, induction type, duration of induction and cell lysis conditions etc., play an important role in enhancing the level of production of the desired protein in its native form have been optimized. We have studied the effect of different types of exogenously employed chaperone systems on the functional expression of rHSA in the E. coli host system. We have also examined different aspects of cell growth parameters during the production of rHSA in the presence and absence of exogenous molecular chaperones—GroEL–ES and trigger factor (TF).

Here, we are reporting the development of a simple, novel and innovative method for enhanced soluble and functional production of rHSA in E. coli. It has been achieved through modulation of basic yet very crucial parameters like the cellular growth, protein folding, and environmental parameters. Adoption of this optimized strategy led to an appreciable improvement and enhancement in the expression levels as well as the functional and soluble proportion of the total expressed rHSA in the cytosolic fraction of the E. coli host system as compared to the previous reports [17, 18]. The present developed process thus paves the way for exploiting the E. coli host for enhanced production of a therapeutic protein with immense potential i.e. HSA in a cost-effective manner, which as per our knowledge had not been successfully attempted earlier.

A range of hosts was used for functional recombinant protein production. E. coli is one of the preferred hosts of choice for the preparation of therapeutic proteins because of its simplicity, fast growth, and low-cost fermentation process [15, 16]. The soluble expression of functional HSA in E. coli is the major difficulty for investigators. In earlier attempts, rHSA was expressed in E. coli but the majority of the protein expressed is insoluble, in the form of inclusion bodies [17, 18]. It is known that the majority of the overexpressed proteins that have hydrophobic patches, undergo post-translational modifications or have disulfide bonds are primarily at the risk to form such amorphous aggregates [21]. HSA is a large multidomain disulfide-rich protein and it is often difficult for a disulfide bond rich protein to fold in its functional form under the normal reducing physiological environment of the E. coli cells.

In the present study, we have targeted and eventually resolved the prime limitation associated with the negligence of E. coli as a host for rHSA production. We have developed a process to overcome an unbalanced equilibrium between rHSA protein aggregation and solubilization at the cellular level by varying basic yet very crucial cellular, folding and environmental parameters. A process for the enhancement of rHSA production in E.coli cells has been accomplished in two phases. The first phase of the process is comprised of two tiers. The first tier of the process involves optimization of a set of cell growth parameters for enhancing soluble cytoplasmic rHSA fraction when over-expressed in the E. coli host. Whereas the second tier involves controlling the aggregation of soluble rHSA during cell lysis by employing the step of osmolytic assistance. In the second phase, we attempted further enhancement of functional rHSA production through the employment of chaperone-assisted folding of cellular recombinant proteins. This led to significant improvement in the cellular expression and folding of nascently synthesized rHSA polypeptide in the E. coli cytoplasm as discussed in detail in the subsequent sections.

It may be anticipated that the oxidizing intracellular environment of the E. coli host cell is preferred choice for the enhanced soluble expression of rHSA. A threefold enhancement in the solubility and activity (Fig. 2) has been attributed to the oxidizing intracellular host environment only, since for a disulfide-rich protein like HSA native disulfide bond formation is a critical factor for the proper folding of the active protein. Thus, rHSA expression in the reducing environment shows significant aggregation and misfolding as compared to its expression under oxidizing conditions. It could be deciphered from the present observation that the post-induction incubation temperature is another crucial determinant of the rHSA soluble expression in E. coli. The optimal functional expression of rHSA was obtained when the post induced cells were cultured at a lower temperature (Process condition L; Fig. 2). Incubation at a lower temperature renders reduced protein synthesis rates in the cells [22], thereby providing enough time for the nascent rHSA polypeptides to fold without undergoing intermolecular interaction. Also, the inclusion body formation is substantially reduced at a lower temperature and hydrophobic interaction decreases between the nascent polypeptides which bring about a reduction in the rate of aggregation [22]. Whereas at a higher temperature, the rate of aggregation is increased due to the strong temperature dependence of hydrophobic interaction [23] which makes nascent rHSA polypeptides more prone to undergo an aggregation reaction.

The induction condition also played an important role in determining the functional status of the expressed rHSA. Interestingly, about twofold enhancement in the solubility and activity (Fig. 2) was observed when a tightly regulated autoinduction strategy was adopted. It is also considered to be a highly attractive mode of induction because of the reduced requirement of monitoring the samples (no need to monitor OD₆₀₀) and the culture reaches high cell density and helping to achieve high throughput protein expression as compared to the conventional IPTG based induction method [19]. The efforts indulged in this first tier of the process offers an optimized set of conditions which when employed together leads to a significantly enhanced soluble rHSA (~ 50% of the total rHSA expressed) production in E. coli cytoplasm.

It is known that a significant fraction of proteins which are not found in inclusion bodies but rather are expressed as soluble proteins in E. coli have a tendency to aggregate after cell lysis [24]. During extraction of soluble rHSA (~ 50% of the total expressed), further losses may occur (loss of protein due to degradation or misfolding or denaturation, loss due to increased temperature during sonication) [24], which can reduce the efficiency of production of soluble and functionally active rHSA. The presence of osmolytes during cell lysis directly impact upon the stability and solubility of proteins by assisting in protein folding and preventing protein aggregation. Trehalose is known to be one of the osmolytes which induce thermostabilization of protein’s structure and is helpful in retention of its biological activity at higher temperatures. It interacts with the side chain of proteins and contributes to their stability [25–30]. Presence of trehalose in the present case contributed to a notable enhancement of rHSA in the soluble fraction (~ 60% compared to about ~ 50% for control) and activity (~ twofold) (Fig. 3), thereby improving the recovery of the functional protein product. Hence, altogether the implementation of the first phase of the process delivers about 60% of the expressed rHSA in the soluble and functional form.

The second phase of the process targeted for further improvement in the functional production of rHSA for which we have explored assistance through chaperone co-expression. Molecular chaperones lie at the heart of protein quality control, aiding the nascent polypeptide to reach their final structure i.e. native form [31–34]. During overproduction of a recombinant protein, endogenous chaperone concentration is likely to be limited which may result in inclusion body formation. However, if chaperones are overexpressed concomitantly with the recombinant protein in vivo then they can help in the preferred folding of the expressed recombinant protein [35, 36]. Despite numerous publications reporting large quantities of soluble recombinant protein with the assistance of chaperones in E. coli [37–45], no report is found till date on the attempts to fold rHSA in E. coli cytosol with the help of co-expressed chaperones. We exploited the opportunity to introduce molecular chaperones (GroEL–ES and TF) to assess the effect on the functional preparation of rHSA in the second phase of the process.

To our surprise, it was found that the expression of rHSA increased by twofold when co-expressed along with the TF chaperone, whereas it was diminished to a negligible level when co-expressed with GroEL–ES chaperone system as compared to when rHSA was expressed without any exogenous chaperone assistance (Fig. 4). This observation could be attributed to the fact that GroEL–ES chaperone is an ATP dependent chaperone and a cell over-expressing a tetra-decameric GroEL and a heptameric GroES along with the recombinant protein of interest undergoes stress, which can limit the cellular ability for recombinant protein synthesis [46–48]. Since the amount of the total expressed rHSA in presence of GroEL–ES system was too low, it led to diminished activity values when compared to the control. TF, on the other hand, is an ATP independent chaperone which binds to the nascent polypeptide chain as soon as it is released from the ribosome into the cytosol, which reduces the aggregation of the polypeptides as well as toxicity associated with the cellular aggregates [34, 49, 50], thereby it dictates the translation speed of the rHSA and in consequence enhancing its expression yield by twofold as compared to the control. Interestingly, there are instances available in the literature where the expression of a protein increases when a cell is subjected to cold shock, because of the production of the cold shock proteins [50]. The fact that TF is a cold shock protein again strengthens the evidence of its involvement in the enhanced expression of rHSA when co-expressed in the host.

It was also discerned that the post-induction incubation time required for optimum rHSA expression in absence of chaperone was less compared to the time required in presence of the TF system. The probable reason underlying this observation is that the cells are burdened with the expression of more than one exogenously expressed recombinant proteins, leading to a reduction in the rate of protein synthesis and thereby causing longer induction period to express recombinant proteins at optimum level [22].

In the present study, it was observed that during folding of the rHSA co-expressed with chaperone TF, the percentage of soluble fraction is comparable, ~ 60% of the total expressed rHSA, with that of the soluble fraction without chaperone assistance but interestingly a substantial elevation in the functional levels of the soluble fraction by 20–30% was obtained when compared to the control (Fig. 5). This is clearly indicative of the chaperone-assisted folding of the newly translated protein. It is to be noted that TF is not enhancing the level of soluble protein through further prevention of aggregation, as the extent of the rHSA in the soluble fraction remained the same. This strongly suggests that TF has the potential to properly fold and thereby enhance the proportion of functional rHSA units in the soluble fraction of the protein when co-expressed inside the E. coli cytoplasm. The vast data available in the literature, on HSA folding and on TF assisted folding of proteins, when connected together helped in drawing an inference which again has supported the conclusion withdrawn from the present study i.e. TF aids in the folding of a multidomain protein like HSA. HSA has a subunit-like structure which follows a hierarchical framework model, wherein the formation of the tertiary structure occurs through a hierarchical assembly of the local elements of the tertiary structure [51, 52]. The domain folding of HSA occurs independently followed by the interdomain interactions for further folding [53]. For multidomain proteins, co-translational domain formation avoids intramolecular misfolding and is a critical factor in their folding [54]. TF, on the other hand, has been reported to allow or even accelerate productive co-translational folding processes like a domain-wise folding of multidomain proteins by serving as a folding scaffold and even facilitating a co-translational, maybe domain-wise, folding of nascent chains within its protected void [55–58]. We, from the above observations, hypothesize that TF helps in the folding of the domains of rHSA as soon as the nascent polypeptide is being synthesized. Thus, helping in the efficient and smooth folding of subsequent domains sequentially, which further assemble in a modular manner for their cooperative association in a biologically active form.

When the cell is forced to overproduce recombinant proteins, not necessarily required for its survival, it experiences an additional burden in terms of the energy utilization for the maintenance and expression of recombinant proteins [59, 60]. In the present study, we have observed that the specific growth rate of all the transformed cells was less compared to the control. It was also found that under uninduced conditions, strains harboring both the plasmids-chaperone and HSA, showed decreased growth rate as compared to the E. coli cells harboring only HSA plasmid (Fig. 7). However, under induced conditions, the specific growth rate of the E. coli strain carrying both the plasmids is comparable or slightly more than that of the strain carrying only one plasmid for HSA (Fig. 7). It may be anticipated that the availability of the abundant chaperones under induced conditions helped in the correct folding of various proteins in the cell thereby preventing their aggregation. This increases the efficiency of the cell and in turn, enhances its growth rate. Such an increase in growth rate has been reported for E. coli with the production of heat shock proteins [61–63]. Another significant observation was that cells subjected to growth in different culture media showed a significant difference in their growth profile (Fig. 7). The specific growth rate of E. coli cells grown in AI medium is almost twofold more than that of the cells grown in LB medium and induced with IPTG. It had been reported that IPTG causes a burden to the cells as IPTG after its uptake cannot be degraded inside the cell and expressing recombinant proteins through IPTG induction may slow the growth rate of the cells [64]. The AI strategy adopted here in the process takes an extra edge as lactose as an inducer is being used.

Upon succeeding with the enhanced functional rHSA production in the E. coli cytosol as well as the development of an improved method for protein extraction, it has become an urgent need to develop strategies for purification and its characterization, considering the fact that HSA is one of the most needed therapeutic protein with immense biotechnological applications. It is also needed to develop the strategies for the scaling up of the production for its commercial use.

The process described here for the enhanced functional production of rHSA is relatively easier and more importantly is suitable for the large-scale industrial production of the active protein. Employment of the developed strategy has significantly improved the soluble yield of the expressed rHSA in E. coli. Since ~ 60% of the expressed protein is soluble, which is a giant leap in production, the earlier adopted strategy of inclusion body in vitro solubilization [18] can be simply abandoned. The protein can be isolated directly from the soluble fraction of the E. coli cytoplasm by following the innovative steps of the currently developed process. The present method also takes into consideration the cell lysis step of the downstream processing for the enhanced recovery of soluble protein. The preliminary attempt for the purification of the rHSA through the Ni–NTA based Immobilized Metal Affinity Chromatography (IMAC) suggests 95% purity in the preparation (Additional file 1: Figure S1). Furthermore, the reactivity of the protein has been checked against anti-HSA antibody followed by preliminary ligand binding assays to assess the functionality of the purified protein, which provided positive information. Further work on the improvement of the purity level of the E. coli derived rHSA is in progress.

To our knowledge, this is the first report where the enhancement of the soluble and functional expression of rHSA in E. coli cytoplasm has been attempted successfully. The convenient and simple process developed in the present study targets and eventually resolves the major limitation of aggregation associated with the rHSA expression in E. coli. Thus, paving the way for the well-studied, convenient and potentially economical host E. coli to be exploited as an alternative host system for rHSA production.

In summary, we describe a novel and innovative method for the enhanced soluble and functional production of rHSA protein in E. coli. To best of our knowledge, this is the first report of the successful enhancement of the soluble and functional expression of rHSA in E. coli cytoplasm.

The presently developed process has remarkably shortened and simplified the soluble rHSA preparation using E. coli as a host. The substantial reduction of rHSA aggregation in the host system has been accomplished by employment of an optimized set of crucial yet very basic cellular, environmental, folding parameters, and employment of an exogenous molecular chaperone system. The present development also takes into consideration the optimal recovery of the functional protein produced in the E. coli host system during the cell lysis step with cheaper ingredients and thereby providing a strategy to overcome one of the major limitations in earlier attempts of E. coli based rHSA production [17, 18].

The developed process thus paves the way for exploiting the most well studied, potentially low cost, scalable host system available for the enhancement of functional expression of the therapeutic protein with immense potential i.e. HSA in a cost-effective manner, which has not been attempted earlier. The present work has filled in the gap in the literature by exploiting the enhanced capabilities of E. coli system as an alternative host for the production of the immensely important, multi-application oriented recombinant therapeutic protein like HSA which is truly important and immensely demand driven, in a cost-effective manner.