New Insights on Carotenoid Production by *Gordonia alkanivorans* Strain 1B

*Tiago P. Silva, Susana M. Paixão, Ana S. Fernandes, José C. Roseiro and Luís Alves*

### **Abstract**

*Gordonia alkanivorans* strain 1B is a desulfurizing bacterium and a hyper-pigment producer. Most carotenoid optimization studies have been performed with light, but little is still known on how carbon/sulfur-source concentrations influence carotenoid production under darkness. In this work, a surface response methodology based on a two-factor Doehlert distribution (% glucose in a glucose/fructose 10 g/L mixture; sulfate concentration) was used to study carotenoid and biomass production without light. These responses were then compared to those previously obtained under light. Moreover, carbon consumption was also monitored, and different metabolic parameters were further calculated. The results indicate that both light and glucose promote slower growth rates, but stimulate carotenoid production and carbon conversion to carotenoids and biomass. Fructose induces higher growth rates, and greater biomass production at 72 h; however, its presence seems to inhibit carotenoid production. Moreover, although at a much lower yield than under light, results demonstrate that under darkness the highest carotenoid production can be achieved with 100% glucose (10 g/L), ≥27 mg/L sulfate, and high growth time (>216 h). These results give a novel insight into the metabolism of strain 1B, highlighting the importance of culture conditions optimization to increase the process efficiency for carotenoid and/or biomass production.

**Keywords:** Gordonia alkanivorans strain 1B, biomass, carotenoids, dark/light, glucose/fructose

### **1. Introduction**

Carotenoids are bioactive molecules characterized by their intense coloration, which varies between red, yellow, orange, and pink, depending on their molecular structure. These molecules serve several biological purposes, aiding in photosynthesis in autotrophic organisms, protecting cells from excess light, have high antioxidant properties, and help in regulating membrane fluidity [1]. Due to these characteristics, carotenoids have garnered the interest of the health, cosmetic, pharmaceutical, and food industries, amongst others, giving them a high market value [2]. Indeed, the

global carotenoids market was valued to grow from \$1.5 billion in 2019 to \$2.0 billion by 2026, recording an annual growth rate of 4.2% during the forecast period [3].

Carotenoids can be obtained from many sources, they can be chemically synthesized, extracted from plants and animals, or extracted from microorganisms. The latter option is becoming increasingly more appealing since microorganisms can be cultivated to higher densities, do not depend on seasonality, and can still be considered a natural source and a safe alternative to the synthetically derived pigments, which has become a deciding factor for consumers [4, 5].

Many microorganisms depend on light as a fundamental factor for carotenoid production. In the case of photosynthetic microorganisms, such as microalgae, known amongst the highest carotenoid producers, light is necessary for autotrophic growth and subsequent carotenoid production. Through photosynthesis, these microorganisms, consume CO2 in the presence of light and use it to grow and produce byproducts. This dependence on light, however, leads to some constraints, since higher biomass concentrations will result in lower light penetration, which can hinder culture growth and reduce production yields [6]. As well, for heterotrophic growth, in microalgae, yeast, or bacteria, light can be fundamental as a powerful inducer, or necessary factor, for carotenoid production [7]. To overcome this problem, reactor designs had to be adjusted to increase surface area and light exposure, resulting in larger, more complex biorefineries, which demand higher extents of land use and/or greater initial costs [8].

When producing microbial biomass in an industrial setting, operation conditions can greatly impact production costs and process efficiency [9]. Many microbial-based industries struggle to optimize culture conditions to increase process efficiency [10]. In a biorefinery, carotenoid production could be viewed as a high-added-value byproduct and not as the focus of the process. In this perspective, the ability to produce carotenoids without a light source would make the process much easier, allowing the use of more conventional installations, and bioreactors, reducing the need for space, or complex infrastructures, leading to a reduction in installation costs [11].

The genus *Gordonia* is known for its carotenoid producers, such as strains of *Gordonia alkanivorans*, *Gordonia jacobea*, *Gordonia terrae*, *Gordonia ajoucoccus* and *Gordonia amicalis* [7, 12–15]. *Gordonia alkanivorans* strain 1B is a bacterium with high biotechnological interest. It has been extensively studied for its biodesulfurization (BDS) abilities, as a biocatalyst to substitute/complement the conventional fuel desulfurization methods. Using the 4S pathway, strain 1B can remove sulfur from complex organo-sulfur molecules, at ambient temperatures and pressures, without the need for additional treatments, potentially making the process more efficient and less pollutant [16–20]. However, one of the drawbacks that may hinder the BDS scale-up is the lack of economic viability, thus measures to reduce the overall process costs to make BDS with strain 1B economically competitive include the use of cost-effective feedstocks [21–24], culture medium minimization [10] and the exploitation of highadded value byproducts, such as biosurfactants and carotenoids [25, 26].

Carotenoid production is an attribute seldomly valorized in the literature related to biodesulfurizing microorganisms; however, it is commonly found in isolates from oil and oil-contaminated environments [27–31].

*G. alkanivorans* strain 1B was repeatedly described as a good carotenoid producer, presenting different concentrations and production profiles depending on its growth conditions [7, 25, 32]. Of the different carotenoids produced, three have been identified as canthaxanthin, astaxanthin, and lutein, by comparing with their respective standards through HPLC [7, 25]. Several carbon and sulfur sources have been tested as

#### *New Insights on Carotenoid Production by* Gordonia alkanivorans *Strain 1B DOI: http://dx.doi.org/10.5772/intechopen.103919*

inducers, to increase carotenoid production with this strain, and initial studies have revealed that glucose and sulfate in abundance, in the presence of light, promote the highest accumulation of carotenoids [7, 32]. However, *Gordonia alkanivorans* strain 1B is one of the few described fructophilic bacteria [17], meaning it presents higher growth rates with fructose, producing biomass at faster rates, but also fewer carotenoids [32]. Furthermore, the presence of sulfate causes the inhibition of the biodesulfurization pathways. Concentrations as low as 30 mg/L almost completely inhibit desulfurization, even in the presence of organosulfur inducers [23].

Some work has already been performed to better understand how these factors (carbon-source/sulfur-source) correlate to generate the highest biomass and carotenoid productivity [32], however, it was mostly performed under the influence of light. Indeed, up to now, little is still known on how factors, such as carbon source and sulfur source concentrations, influence carotenoid production by *G. alkanivorans* strain 1B without the stimulus of light. Moreover, there is also a need to better understand the correlation between carotenoid accumulation, biomass production, and carbon consumption, with and without light. The correct balance between these responses is fundamental to better understand the metabolism of strain 1B and efficiently drive the process toward the production of either biomass (i.e., biocatalysts for desulfurization) or carotenoids, depending on the purpose of the biorefinery in consideration (bioproduct *versus* bioprocess).

This work initially focuses on the optimization of culture conditions toward carotenoid production by *G. alkanivorans* strain 1B without the stimulus of light. In this context, a surface response methodology (SRM) based on the Doehlert [33] distribution for two factors (% of glucose in a mixture of glucose + fructose (10 g/L total sugars); and sulfate concentration) was performed in the absence of light. Moreover, these SMR results (total biomass; total carotenoid production) were compared with the SRM results previously obtained by Fernandes et al. [32] for the carotenoids production in the presence of light (400 lux). In addition to biomass and carotenoids, specific carotenoid production (μg of carotenoids/g of dry cell weight), carbon consumption, and carotenoid and biomass production per carbon consumed were also evaluated as responses, both in absence/presence of light.

### **2. Materials and methods**

#### **2.1 Chemicals**

Sodium sulfate anhydrous (>99%) was from Merck (New Jersey, USA). Dimethyl sulfoxide (DMSO) (99.9%), acetone (99.9%), ethyl acetate (99.8%), and methanol (99.9%) were obtained from Carlo Erba Reagents (Val de Reuil, France). The remaining reagents were of the highest grade commercially available. Stock solutions of glucose (glu) and fructose (fru) were prepared at 50% (w/v), filter sterilized, and stored for further use as a carbon source (C-source) in culture media. In the same way, a stock solution of 20 g/L Na2SO4 was also prepared and autoclaved (121°C, 1.03 bar, 15 min) to be further used as the sulfur source (S-source).

#### **2.2 Microorganism and culture conditions**

The microorganism used in this study was *G. alkanivorans* strain 1B, a bacterium isolated in our laboratory [16], and kept at a culture collection of microorganisms

(CCM at LNEG, Portugal, Lisbon). The basal salts medium used for cultivation, maintenance, and for all the growth/carotenoid production assays was described in Ref. [32]. The final pH was adjusted to 7.5 prior to sterilization by autoclave (121°C, 1.03 bar, 15 min). Afterward, the C-source (fructose and/or glucose) was added to the culture medium, in aseptic conditions, to an initial concentration of 10 g/L of total sugar(s). Similarly, the stock solution of S-source (Na2SO4) was also added to obtain the desired final concentrations of 9.04 mg/L, 22 mg/L, and 34.99 mg/L, depending on the assay.

The bacterial cultures were performed in 500 mL Erlenmeyer shake-flasks containing 150 mL culture medium, covered in tin foil to avoid light exposure, incubated in an orbital shaker (≈150 rpm) within an acclimatization chamber (Fitoclima 14000E Walk-In, Aralab, Portugal), at 30°C. All the assays were performed at least in duplicate. Sampling was carried out at 72 h and 216 h for immediate biomass determination (DCW = dry cell weight in g/L), while the remainder of each sample was centrifuged (8600 g at 4–5°C, 20 min in a refrigerated Sigma 2–16 K centrifuge) and the respective cells stored at �20°C until further pigment extraction and analysis. The supernatant was evaluated for sugar concentration through HPLC, using a Sugar-Pak 1 column (6.5 � 300 mm, 10 μm, Waters™, MA, USA) [32].

#### **2.3 Experimental design methodology**

A Doehlert distribution for two factors was used as the base for a surface response methodology (SRM) [33] to study carotenoid and biomass production by *G. alkanivorans* strain 1B in the absence of light (L0). The two factors studied were: *X*<sup>1</sup> – % of glucose in mixture glucose + fructose of 10 g/L of total sugars (0–100% glucose in the mix) and *X*<sup>2</sup> – sulfate concentration (7–37 mg/L of sulfate). Fourteen experiments (seven conditions in duplicate) were carried out. The results were evaluated in terms of responses (*Yi*): biomass and total carotenoids production by strain 1B, at 72 h and 216 h. The model used to express the responses was a secondorder polynomial model:

$$Y\_i = \beta\_0 + \beta\_1 X\_1 + \beta\_2 X\_2 + \beta\_{12} X\_{12} + \beta\_{11} X\_{12} + \beta\_{22} X\_{22} \tag{1}$$

where: *Yi* – response from experiment *i*; *β* – parameters of the polynomial model; and *X* – experimental factor level [7, 23, 32].

In addition, specific carotenoid production (μg of carotenoids/g of DCW) was also evaluated as another response, both with and without a light source. The same polynomial model was applied to these results to generate the corresponding response surfaces. Model validation was performed through the Fischer test, for the effectiveness of the factors and the lack of fit, and R<sup>2</sup> (coefficient of multiple determination).

#### **2.4 Carotenoid extraction and analysis**

Carotenoid extraction and further characterization were performed following the procedure described in Ref. [7, 32]. The pigment results are presented as μg of total carotenoid produced (μg carotenoids per 150 mL) or as specific carotenoid production (μg of carotenoids/g of DCW).
