Optimized pH and Its Control Strategy Lead to Enhanced Itaconic Acid Fermentation by Aspergillus terreus on Glucose Substrate

Abstract

Biological itaconic acid production can by catalyzed by Aspergillus terreus (a filamentous fungi) where the fermentation medium pH is of prominent importance. Therefore, in this work, we investigated what benefits the different pH regulation options might offer in enhancing the process. The batch itaconic acid fermentation data underwent a kinetic analysis and the pH control alternatives were ranked subsequently. It would appear that the pH-shift strategy (initial adjustment of pH to 3 and its maintenance at 2.5 after 48 h) resulted in the most attractive fermentation pattern and could hence be recommended to achieve itaconic acid production with an improved performance using A. terreus from carbohydrate, such as glucose. Under this condition, the itaconic acid titer potential, the maximal itaconic acid (titer) production rate, the length of lag-phase and itaconic acid yield were 87.32 g/L, 0.22 g/L/h, 56.04 h and 0.35 g/g glucose, respectively.

1. Introduction

Microbial fermentation has been demonstrated as an efficient technology to produce a variety of organic acids such as malic acid, succinic acid, propionic acid, itaconic acid, etc. [1,2,3,4]. The latter, itaconic acid, is taken into account as an important compound since it can serve as a platform molecule for the synthesis of industrially-relevant chemicals, such as plastics, etc. [5,6]. Nowadays, itaconic acid is mainly generated through biological pathways by the assistance of filamentous fungi, particularly Aspergillus terreus [7]. The fermentation of itaconic acid can be carried out on numerous feedstocks, including complex agro-industrial wastes such as lignocelluloses as well as simple (monomeric) sugars, e.g., glucose [8]. Certainly, the properties of the actual starting material will influence the achievable itaconic acid formation efficiency [9,10,11], and besides that, process control via the maintenance of adequate environmental conditions will play a key role. As a matter of fact, ensuring suitable aeration, broth composition, mixing, temperature and pH are crucial criteria for the improved formation of itaconic acid by A. terreus [12,13,14,15].

For the recovery of itaconic acid from the fermentation liquor, membrane electrodialysis (MED) has been proven to be a plausible solution [16,17,18]. Furthermore, MED (depending on the trait of the membrane) was found to be an approach that provides additional process benefits, such as in the case of citric acid downstreaming [19]. It was concluded by scientists such as Tongwen and Waihua [20], as well as Pinacci and Radaelli [21], studying the separation of fermentatively-generated citric acid, that MED equipped with a bipolar membrane enabled the production of caustic soda. This chemical, NaOH, can be recycled to the bioreactor unit in order to adjust and keep the pH at the desired level during the fermentation [12].

From the viewpoint of itaconic acid biosynthesis catalyzed by A. terreus strains, the pH is usually set to the acidic range. However, the appropriate pH adjustment strategy leading to a better fermentation efficiency could be worthy for investigation, since in the literature there is no clear justification of whether pH should be controlled or not and what pH value is the most appropriate. Actually, various studies have come to different conclusions regarding this aspect [8,12,22], implying the need for further (case-specific) examination. Therefore, in this work, we aimed to study how different acidic pH values (2.5–4) and the pH regulation strategy (only initial pH setting vs. continuous pH control; one-step vs. stepwise pH setting) might make any difference in governing the itaconic acid fermentation towards a higher efficacy.

In this respect, analogously to the case of citric acid [20,21], the NaOH—obtainable by a bipolar MED process [23]—may be employed for the regulation of pH during itaconic acid production. To evaluate and rank the various pH regulation scenarios, the progress curves of batch itaconic acid fermentations by A. terreus on a glucose substrate were kinetically analyzed to deliver process performance indicators (lag-phase time, itaconic acid production rate and titer potential).

The importance of this work can be explained by the inconsistencies regarding the effect of the pH on itaconic acid production by A. terreus and, hence, the findings could demonstrate an added-value to this segment of the literature.

2. Materials and Methods

2.1. Microbial Catalyst

A. terreus NRRL 1960 strain [13] was employed for itaconic acid fermentation in this study. The fungus was sustained on Petri-dishes at 37 °C using a solid medium comprising of (g/L): glucose–10; NaCl–20, potato dextrose agar–40; pH = 5. For the inoculation of the fermenter (Section 2.2.), liquid cultures of A. terreus (grown under pH = 3 and a 150 rpm agitation rate on (g/L): glucose–10, KH₂PO₄–0.1, NH₄NO₃–3, MgSO₄ × 7 H₂O–1, CaCl₂ × 2 H₂O–5, FeCl₃ × 6 H₂O–1.67 × 10⁻³, ZnSO₄ × 7 H₂O–8 × 10⁻³ and CuSO₄ × 7 H₂O–15 × 10⁻³) were harvested after 72 h.

2.2. Bioreactor System for Itaconic Acid Production

To aerobically produce itaconic acid under batch conditions, a Lambda Minifor bioreactor system (available online: https://www.lambda-instruments.com/fileadmin/user_upload/PDF/MINIFOR/Operation_manual_of_LAMBDA_MINIFOR_laboratory_Fermentor_and_Bioreactor.pdf; accessed on 07.01.2019) was applied. The bioreactor with a 1.8 L working volume was filled with a medium (similar to the one used for preparing the inoculum) and autoclaved before commencing the actual experiment. The fermentation conditions tested in this investigation can be seen in

Table 1. The pH setting strategies tested in this study.

Experimental Setting	pH	Aeration (L (STP)/min)	Agitation (Hz)	Substrate
A	Initial pH set to 3 and left uncontrolled	1.5	2	glucose
B	Initial pH set to 3 and maintained	1.5	2	glucose
C	Initial pH set to 2.5 and maintained	1.5	2	glucose
D	Initial pH set to 4 and maintained	1.5	2	glucose
E	Initial pH set to 3 and, after 48 h, maintained at 2.5	1.5	2	glucose

. The concentration of the glucose substrate was fixed at 120 g/L thoroughly, and the temperature was controlled at 37 °C. The inoculation rate was 5% in all cases. The pH (Table 1) was adjusted using NaOH and HCl solutions. The term “STP” in Table 1 refer to the standard temperature and pressure conditions.

2.3. Analytical Procedure

In this study, the itaconic acid production was monitored by the High Performance Liquid Chromatography (HPLC) technique on a Young Lin Instrument Co., Ltd. (YL9100-type) device. The unit contained a Hamilton × 300 HPLC column (length: 15 cm, inner diameter: 4.6 mm, particle size: 5 µm) as well as a UV/VIS detector. The analytical method employed a gradient elution (2 mL/min flow rate) where the moving phase was comprised of A (0.01 M H₂SO₄) and B (methanol) solutions (2 min–100% A; 5 min–50% A, 50% B; 8 min–20% A, 80% B). The samples taken at various spots of the fermentation were treated by membrane filtration (0.22 µm PVDF) and thereafter diluted 1000× using 0.01 M sulfuric acid. The itaconic acid yield (as seen in the Results and Discussion section) was estimated on the grounds of the substrate that was added initially. Fermentation metabolites that were possibly competitive to itaconic acid (e.g., itatartaric acid, gluconic acid, oxalic acid, etc.) were not assessed.

3. Results and Discussion

The pH is one of the most crucial among the fermentation variables, therefore requiring special attention for submerged fungal cultures producing organic acids, such as itaconic acid, with a sufficient performance. Basically, the impact of the pH is associated with the (i) activity of enzymes taking part in the biosynthesis of itaconic acid, and additionally with (ii) the subsequent transfer mechanism to the extracellular space/out of the cell. As could be deduced from the literature, itaconic acid generation by filamentous fungi such as A. terreus favors lower pH conditions, mostly around pH = 2–3 [8]. It has been argued that besides enabling the appropriate growth of A. terreus [24], such a fermentation environment can be useful for suppressing the formation of by-products that would lower the final itaconic acid yield and productivity [25]. Typical by-products of fungal itaconic acid fermentation can be itatartaric acid, gluconic acid and oxalic acid, depending on the pH conditions, due to mechanisms reviewed by Mondala [10]. The advantages of a low pH can also originate from the (i) limited threat of microbiological contamination, (ii) the avoidance of an extreme mycelial network expansion facilitating itaconic acid conversion because of an improved carbon flux as well as (iii) the proper morphology of the strain, (iv) the increased transfer of oxygen gas and (v) aided downstream [10,25]. Although it seems to be established that an adequate pH adjustment is a key-step, the results obtained by various studies that apply the same strain of A. terreus are still frequently divergent [12,22]. In fact, although the optimum pH is basically a strain-specific feature, various studies suggest that the pH should be optimized by taking into account other process parameters characterizing the particular bioreactor unit. For instance, Riscaldati et al. [26] demonstrated that the pH and stirring rate together govern itaconic acid fermentation, while Vassilev et al. [27] found by a response surface methodology that itaconic acid production was notably influenced by the complex relationship of the pH, substrate concentration and nitrogen source (e.g., ammonium nitrate). These examples and observations imply the need for investigating the impact of the pH under the actual circumstances of a particular study.

Accordingly, as can be seen in Table 1, the effect of various pH setting strategies was sought. The experiments were planned on the grounds of relevant concepts reported in the literature: one common practice considers only the setting of the initial pH, where afterwards it is allowed to decrease automatically [13,27,28] (Table 1A), while others propose a well-controlled pH throughout the fermentation to prevent the depression of the itaconic acid production efficiency [29] (Table 1B–D). Apart from that, researchers such as Hevekerl et al. [12] found potential in the pH-shift approach, where the pH is initially set, runs freely for a certain period of time and is controlled only from a given point of the biological conversion, e.g., when itaconic acid production begins (Table 1E). In accordance with the previous argument, the pH was varied between 2.5 and 4 (Table 1). To characterize the batch itaconic acid fermentation kinetics, under the conditions listed in Table 1, the modified Gompertz-model (Equation (1)) was adopted [30]:

$$\mathit{IA}(t)=P \exp \{-\exp [\frac{R_{m}e}{P}(\mathsf{\lambda }-\mathrm{t})+1]\}$$

(1)

This approach enables the user to determine important process parameters (

Table 2. Experimental itaconic acid production data.

Time (h)	Experimental Setting
	A	B	C	D	E
	Itaconic Acid Titer (g/L)
0	0	0	0	0	0
24	0	1.63	0	0.80	0
48	2.88	3.99		0.94	0
72	5.63	5.07			17.17
96	8.32	6.07	9.20
120			12.18	0.90
144			18.48	1.79	17.84
168	18.69	8.94	20.37	1.98	24.40
192	21.01	12.48	19.81	3.39	29.47
216	24.24	14.12		3.43
240	26.28	13.87			41.4
264		13.99	19.98
288				5.10

) from the evaluation of fermentation time profiles (Figure 1), where IA(t) is the actual itaconic acid titer (g/L) at time t (h); P is the itaconic acid titer potential (g/L); R_m denotes the maximal itaconic acid (titer) production rate (g/L/h), λ is the length of the lag-phase time (h); and e is 2.718. To obtain the best fitting of the model and experimental curves, this work relied on the least-squares regression method using the Solver tool in MS Excel. The basic statistical assessment of the results in Table 2 is shown in

Table 3. Descriptive statistics for the experimental data presented in Table 2.

Statistical Data	Experimental Setting
	A	B	C	D	E
Valid number of data	9	10	8	9	8
Mean	11.89	8.02	12.5	2.04	16.29
Minimum	0	0	0	0	0
Maximum	26.28	14.12	20.37	5.1	41.4
Standard deviation	10.63	5.39	8.71	1.63	15.44

.

The experimental itaconic acid production data as a function of time for each test condition (whose corresponding graphs are displayed in Figure 1A–E) are listed in Table 2, and were subjected to a kinetic analysis using the modified Gompertz-formula (Equation (1)) to determine the lag/adaptation-phase time, itaconic acid titer potential and maximal titer production rate (

Table 4. Results of the kinetic process evaluation based on the data from Table 2.

Kinetic Data	Experimental Setting
	A	B	C	D	E
P (g/L)	32.70	17.17	20.75	12.98	87.32
Rm (g/L/h)	0.15	0.07	0.26	0.02	0.22
λ (h)	41.61	4.51	63.70	72.44	56.04

).

Based on these process factors, a ranking was performed, considering that the shorter λ, higher P and R_m are preferred. Bearing this in mind, points (1–5) were assigned in a parameter (λ, P, R_m)-wise manner to demonstrate how they were affected by the 5 different fermentation conditions (

Table 5. Ranking of the various pH setting strategies.

Experimental Setting	Score			Sum of Scores	Final Rank
	P	Rm	λ
A	4	3	4	11	2
B	2	2	5	9	4
C	3	5	2	10	3
D	1	1	1	3	5
E	5	4	3	12	1

). The final assessment was made by summarizing the given scores. Accordingly, the various pH setting strategies could be ordered, as follows: (E) > (A) > (C) > (B) > (D).

This outcome signifies that the pH-switch strategy (E) was the only one that led to a better itaconic acid formation characteristic than measurement (A), which is the widely-applied approach in the literature and can thus be viewed as the reference setting. In this respect, the findings of Hevekerl et al. [12] are supportive, as it turned out that the best itaconic acid concentration (146 g/L) was attained when the regulation of the pH to 3 began slightly after 2 days of cultivation in the bioreactor. The positive impact was believed to be ascribed to the lower degree of stress on the fungal cells. Actually, this strategy led to a nearly 70% improvement in comparison to the fermentation with the uncontrolled pH [12].

Under the best fermentation condition (E) of this work, the experimental itaconic acid yield—considering the quantity of substrate added—was 0.35 g/g glucose. This seems to relate well with the literature data, where yields in the range of 0.21–0.62 g itaconic acid/g glucose can be found with A. terreus strains [22,26,31].

Furthermore, from settings (B) and (C), on can infer that even if the pH is kept constant during the entire biological transformation, the (initial) pH value plays an important role. Accordingly, pH = 2.5 resulted in preferable fermentation kinetic features than pH = 3. This is in agreement with the superiority of setting (E), where the pH was consistently 2.5 from the second day onwards. Besides, it can be concluded that the experimental setting (D) with a pH maintained at 4 was the least attractive by far. Hence, the use of NaOH—recoverable with MED [20,21,22], as elaborated above—in relatively larger quantities for adjusting the pH to the less acidic range could have an adverse effect. This can make sense in light of the above statement that most studies regarding itaconic acid generation proposed a pH of around 2–3 [8].

4. Conclusions

In this study, itaconic acid fermentation from glucose by A. terreus was investigated in relation to the effect of the pH and its regulation strategy. It was found that the initial pH value played a significant role and, additionally, that it did make a difference if the pH was initially adjusted or controlled. Ranking the various pH setting alternatives based on the analysis of fermentation kinetics showed that (under the conditions of the experiments, e.g., bioreactor type, aeration, stirring rate, and substrate concentration), the initial adjustment of the pH to 3 and its adjustment to 2.5 after 2 days was the most promising alternative and should therefore be applied.