1. Introduction
With an increase of consumer demands for natural origin in day-to-day products, the space for natural colors in the market continues to accelerate in its switch towards close-to-nature ingredients, which are phasing out artificial dyes. Having this focus, pigments produced by natural sources have become a point of interest among manufacturers and researchers for applications in the sectors of food, beverages, cosmetics, pharmaceuticals, dyeing clothes, paintings, inks, industrial coatings, etc. [
1]. In the case of pigments, natural colors do not have the same color intensity as synthetics and some (not all) are less economical on a dosage basis; however, technological advances have reduced this performance gap. Therefore, production of natural colorants from various natural sources continues to rise due to its adequacy to meet consumer expectations, despite of its higher cost of production [
2]. Among all available natural sources, pigment production from micro-organisms is gaining interest over other sources. In particular, filamentous fungi have been recognized as the most promising potential source to produce not only an extraordinary range of pigments but also several bioactive compounds [
3,
4].
A wide range of fungi belonging to the genera
Aspergillus,
Monascus, Penicillium,
Periconia, Paecilomyces, Talaromyces and
Trichoderma are known to produce pigments of various hues and are widely studied [
5,
6]. Amongst all, a few potential candidates from the family of
Trichocomaceae are key ones for industrial-scale production which includes
Aspergillus,
Penicillium and
Talaromyces [
3,
7].
Talaromyces albobiverticillius (
Talaromyces being the teleomorphic genus of
Penicillium sp.) produces
Monascus-like azaphilone pigments (MLAP) [
8,
9]. From the fungi
T. albobiverticillius 30548 strain, a total production of 12 different compounds was described and among them, six different compounds have been identified by LCMS analysis [
10]. Literature studies have also shown that some species of
Talaromyces synthesize yellow- (monascorubrin and rubropunctatin) and red- (monascorubramine and rubropunctamine) colored pigments and among them, few species do not produce any mycotoxin along with these pigments [
8,
11,
12,
13,
14,
15]. This factor allows the biotechnological production of azaphilone pigments using
Talaromyces spp. favorable for large-scale production. The pigments produced by
T. atroroseus,
T. albobiverticillius,
T. purpurogenus,
T. aculeatus and
T. funiculosus are diffusing into the culture medium in submerged fermentation. However, submerged fermentation in small volumes is a first step to assess the critical parameters, resulting in better understanding of the process before scale-up to industrial scale with the goal of increasing production yields [
16].
Scale-up from shake flasks to bioreactor is aimed at producing target compounds in large quantities and towards improving specific yields and product quality, if all the interactive factors are well controlled. As a means of getting higher yield, several operational parameters such as pH, dissolved oxygen, heat and mass transfer, mixing time, shear rate and agitation speed are the key influencers to make the fermentation successful [
17,
18]. However, in filamentous fungi it is challenging to obtain consistent and reproducible data due to the fungal behavior and morphology inside the bioreactor with the interaction of various factors [
19,
20,
21]. While the basic parameters in fermentation remain the same for all end applications, ranges and their requirements should be modified in response to the requirements of fungal type and target metabolites production [
16,
22].
The primary objective of this study was to run the liquid fermentation in a 2 L bioreactor to understand the behavior of fungal growth and pigment production in a highly controlled, closed bioreactor. Previous experiments in shake flasks allowed us to obtain the optimal conditions to produce pigments with a fixed, controlled temperature, agitation rate and composition of the culture media [
23]. In submerged fermentation using a 2 L bioreactor, parameters such as temperature, pH, dissolved oxygen and aeration rate were considered the initial key factors to monitor the biomass growth and pigment production by setting the above optimized factors as constant. This study was an initial approach to identify the interaction of variables, influence of aeration and agitation, pH control strategy and thereafter, additional efforts will be done to improve the cultivation conditions and to enhance the production in a further scale-up.
2. Materials and Methods
2.1. Organism and Maintenance
Talaromyces albobiverticillius 30548 has been isolated from a marine sediment source in Reunion Island, Indian Ocean area, and sampled from Trou d’Eau, on the external slope of the coral reef, at 17 m depth (21°06′22.11″ S, 55°14′15.78″ E) [
24]. The fungus was grown on potato dextrose agar plates (PDA, Sigma-Aldrich, St. Louis, MO, USA) at 24 °C for a period of 7 days and it produced dark-red colored pigments between days 4 to 7. After 7 days of growth period, the agar plates with the solid culture were maintained at 4 °C for conservation as well as to sub-culture at regular intervals.
2.2. Inoculum Preparation
To make liquid seed culture, Potato Dextrose Broth (PDB, Sigma-Aldrich, St. Louis, MO, USA) was used as a growth medium. An amount of 100 mg of mycelial spores were taken from the fully grown agar plates and aseptically inoculated into 250 mL shake flasks (Erlenmeyer) containing 100 mL working volume of PDB medium. After the addition of spores, the culture was kept in a shaking incubator at 24 °C with an agitation of 200 rpm (Multitron Pro, Infors HT, Bottmingen, Switzerland) for a period of 72 h.
The cultivation for pigment production was composed of two stages: the first stage was preparing liquid seed culture that acts as pre-inoculum; the second stage was submerged fermentation in shake flasks (250 mL total volume) and small-scale bioreactor of 2 L total volume using seed culture.
2.3. Fermentation in Shake Flasks
Preliminary fermentation runs were done using 200 mL working volume of PDB medium in 500 mL shake flasks. The media was sterilized at 121 °C for 15 min and once it was cooled down to room temperature, the medium pH was adjusted to 5.0 under sterile conditions using 0.1 M HCl before inoculation. From 72 h seed culture, 1% (v/v) was added to flasks and those were kept in a shaking incubator at 24 °C and 200 rpm for 10 days. During the length of fermentation, the pigment production and biomass growth were monitored every 24 h at regular intervals and data was recorded. All the experiments were performed as triplicates and statistical analysis such as one-way ANOVA was carried out using SigmaPlot ver.10 (Systat Software Inc., San Jose, CA, USA).
2.4. Fermentation in Bioreactor
As a scale-up from shake flasks, fermentation was performed in a 2 L bioreactor (BIOSTAT® A PLUS, Sartorius Stedim Biotech, Goettingen, Germany) sealed with a stainless steel head-plate. This is a compact, autoclavable fermenter with integrated controls and measurement tools which makes it easy to use, as well as to transit from shake flasks for culturing microbes. The working bioreactor is equipped with temperature and pH control, dissolved oxygen probe and flat-bladed impeller for stirring (3-blade segment impeller, BB-8847398, Sartorius Stedim Biotech, Goettingen, Germany). A working volume of 1.3 L PDB medium was in situ sterilized at 121 °C for 15 min and upon cooling, it was inoculated with 5% v/v of pre-inoculum (1.3 g/L wet mycelia). After the inoculation, there was no significant change in the working volume of the medium. The principal culture parameters such as temperature (24 °C) and initial pH (5.0) for this experiment were kept the same as those used for the shake flasks experiments. The pH was controlled automatically to be kept within a range of 4.9–5.1 using a pH control module (LH Fermentation Ltd., Stoke Poges, Reading, UK) equipped with a steam-sterilizable pH electrode (EF—12/120 K8-HM-UniVessel, Sartorius Stedim Biotech, Goettingen, Germany) by the automatic addition of 0.1 M NaOH.
Based on the results of preliminary tests in the bioreactor, agitation speed and air input were fixed from the start of this experiment and at a later stage, that had to be changed after examination of the culture evolution and morphology. During the entire fermentation process, the agitation speed was maintained by a flat-bladed impeller (200–1000 rpm) and sterile air was supplied at 1.3 L/min (100% pO2, dissolved oxygen) with the use of air filters but varied (50, 70 and 90%) in the later runs depending on fungal growth demand.
2.5. Biomass Estimation
Fermentation was carried out for a total of 240 h and in between, an aliquot of sample (5 mL) was drawn aseptically using a sterile single-use syringe once every 24 h throughout the entire length of fermentation. The fermented broth was filtered using 48 μm Nitex filter cloth (Nitex 03-48/31, SEFAR AG, Heiden, Switzerland) to separate the fungal biomass and supernatant. The separated biomass was precisely weighed using an analytical weighing balance (Adventurer Pro AS214 d = 0.0001 g, Ohaus Europe GmbH, Greifensee, Switzerland) and it was considered as the wet fungal biomass. To determine the dry biomass weight, the sample was dried in a hot air oven (SNB 100, Memmert, Schwabach, Germany) at 105 °C for 17 h and afterwards, it was kept in the desiccator for 30 min to get a precise weight [
25].
2.6. Estimation of Pigments Absorbance
The daily collected supernatant considered as extracellular pigments was used to measure the pigment absorbance. The intracellular pigments were extracted from the biomass at the end of fermentation (day 8) using ethanol after washing the biomass twice with deionized water. Since this extraction is intensive for daily measurements and possess pigments with most orangish red hues, the focus was shifted mostly towards extracellular pigments, which is of interest with most red pigments. Hence, extracellular pigments were more favored as they have a mixture of pigments which have been previously studied and published [
10]. The maximum absorption of the extracellular pigments was determined by scanning the colored extracts over the range of 200–700 nm (UV-vis area) using a microplate reader (Infinite
® 200 PRO series, Tecan Life Sciences, Mannedorf, Switzerland) [
21]. The maximum absorbance wavelength of extracellular pigments in the filtrate was estimated at 470 nm (orange pigments) and 500 nm (red pigments). The extracellular supernatant was diluted with distilled water prior to measuring the absorbance, which was done to keep the concentration within an acceptable range. The pigment concentration in the extracellular sample was expressed as absorbance units (AU) by taking into consideration the dilution factor and the volume of sample.
2.7. Color Characteristics
The remaining extracellular supernatant after pigment absorbance was used to measure CIELAB color coordinates using Spectrocolorimeter (CM-3500d Spectrocolorimeter, Konica Minolta, Tokyo, Japan). The CIE L*a*b* colorimetric system was interpreted as follows: the value L* indicates lightness and covers from 0 (black) to 100 (white). The positive to negative a* value indicates red to green colors, whereas positive and negative b* represents yellow or blue colors, respectively. Chroma, denoted by C, gives saturation or purity of color. Hue angles, h° denotes the degree of redness, yellowness, greenness, and blueness by locating at 0, 90, 180 and 270°, respectively. The values of L, a*, b*, C and h° were obtained automatically during analysis from the color data software called SpectraMagic™NX ((version 1.9, Minolta Co., Tokyo, Japan). The standard illuminant D65, now considered as a principal reference illuminant, was used in all colorimetric measurements to display average daylight. Initial calibration was done using control and blank for quantification and color analysis [
26].
Chroma (C*) and hue angle (h°) were calculated from the following equations:
4. Discussion
The scale-up process from laboratory to pilot level to industrial scale presents several inherent challenges. To achieve an effective bioconversion and produce targeted metabolites in a fermenter, some critical parameters such as homogenization of culture media, pH regulation, mass and heat transfer and dispersion of gas should be regulated throughout, until the end of fermentation depending on the organisms. All these parameters could be regulated with the help of agitation by choosing an impeller that meets the needs of the biological process specific to a particular organism [
28]. Besides, agitation plays a crucial role in influencing fungal morphology and, in turn, morphology has an effect on pigment production and its yield, including extraction efficiency [
35].
Differences in fungal morphology, such as growing as pellets or mycelia, in submerged fermentation have been noticed in many different filamentous fungi, i.e.,
Aspergillus,
Rhizopus or
Penicillium strains [
36,
37,
38,
39]. In most cases, the formation of pellets due to low agitation speed and high bubbling coming out from aeration may interfere with O
2 penetration, dissolution or substrate uptake and thus considerably influences the efficiency of target product formation [
40,
41,
42]. However, low agitation speeds may not generate enough turbulent flow to disperse the air bubbles too effectively into the media. In this study with
Taloromyces albobiverticillus 30548, at a low agitation speed of 30 rpm, pigment production was dramatically depressed. In orbital shake culture flasks at 200 rpm using the same PDB as culture media, growth always took place in the form of mycelia, but pellets were observed in slow-stirred tank bioreactors under the same basic culture parameters (pH 5, temperature 24 °C). This could be partly due to lack of proper oxygen transfer in the culture media in the bioreactor and also to the type of shear forces produced by the stirrer. This leads to the growth of cells with different morphologies, either pellets or filaments, and eventually to a decrease in pigment yield in the case of pelleted-structured growth [
43]. The same behavior was observed with
Monascus fungi, in which cell morphology as pellets had an unfavorable effect on final pigment yield, which diminished pigment production [
44]. As a side effect, the 2 L bioreactor has a larger surface area containing liquid media in contact with gas compared to shake flasks. Also, accessories of the fermenter (agitation impeller, foam control probe and pH and temperature control probes) offer static aerated surfaces. Therefore, the free filamentous fungi seemed to attach and grow on the probes and the wall, where pO
2 is maximum and feeding medium is regularly spilled.
While studying the impact of biomass and pigment production in
Talaromyces albobiverticillius 30548, it was observed that at low agitation speed (30 rpm) and variable pO
2 rates (10–100% pO
2), pigment production was inferior and the fungus grew as pellets. Similar observations of fungal growth as pellets based on the influence of agitation rates have been reported on
Neurospora intermedia and
Rhizopus oryzae [
37,
40]. In addition, agitation speed played a major role in determining fungal morphology but when considering pigment production, the effect of dissolved oxygen concentration was greater than that of agitation speed. From observation of current research, at high-speed agitation (>1000 rpm for 8 days), the structure of fungal mycelia was not damaged.
In shake flasks, the pigment production starts after 48 h and is observed as light orange initially. Fermentation ends within 9 days with dark red pigment production. It has been suggested that red pigments are derived from orange precursors by a chemical reaction and are considered the first biosynthetic product [
45,
46]. A similar color shift was observed in this experiment with the absorbance scan of pigmented extracts of
T. albobiverticillius, the first one near 430 nm and the second one at 511 nm; a few days later, the first one moved in the wavelength range from 430 nm to 470 nm.
During scale-up, there was a drastic decrease in pigment yield and biomass at 10% pO
2 compared to shake flask cultures. In those experimental conditions, the pigment yield obtained from the bioreactor was considerably low (4.89 AU 470 nm, 5.09 AU 500 nm, in 1.3 L PDB working volume, 10 pO
2 and 100 rpm agitation speed) compared to shake flasks (22.21 AU 470 nm, 18.61 AU 500 nm, 200 mL working volume and 200 rpm agitation speed). This might be partly due to the low oxygen diffusion which adversely affected growth (4.88 g/L in fermenter vs. 8.10 g/L in shake flasks) and therefore pigment production. Different pH levels influenced the physiology of fungi, conidial development and pigment synthesis. Reducing the pH inhibits the formation of conidia and increases pigment production, suggesting that the pH of the medium might affect the transport of certain media constituents such as glucose and nitrogen sources [
47]. The pH can affect the activity of enzymes involved in the biosynthesis of pigments and research findings by several groups proposed that careful selection of pH influences the production of the predominant color component [
48,
49]. For
T. amestolkiae cultivation in a chemically defined media along with MSG, deep yellow colorants were observed using neutral and basic pH, whereas deep red colors were seen with acidic pH [
46]. This is in conformity with the red colorant production in
T. albobiverticillius 30548.