Unveiling the Effect of NCgl0580 Gene Deletion on 5-Aminolevulinic Acid Biosynthesis in Corynebacterium glutamicum

Abstract

5-Aminolevulinic acid (5-ALA) has recently received much attention for its wide applications in medicine and agriculture. In this study, we investigated the effect of NCgl0580 in Corynebacterium glutamicum on 5-ALA biosynthesis as well as its possible mechanism. It was found that the overexpression of NCgl0580 increased 5-ALA production by approximately 53.3%. Interestingly, the knockout of this gene led to an even more significant 2.49-fold increase in 5-ALA production. According to transcriptome analysis and functional validation of phenotype-related targets, the deletion of NCgl0580 brought about considerable changes in the transcript levels of genes involved in central carbon metabolism, leading to fluxes redistribution toward the 5-ALA precursor succinyl-CoA as well as ATP-binding cassette (ABC) transporters affecting 5-ALA biosynthesis. In particular, the positive effects of enhanced sugar transport (by overexpressing NCgl1445 and iolT1), glycolysis (by overexpressing pyk2), iron uptake (by overexpressing afuABC), and phosphate uptake (by overexpressing pstSCAB and ugpQ) on 5-ALA biosynthesis were demonstrated for the first time. Thus, the transcriptional mechanism underlying the effect of NCgl0580 deletion on 5-ALA biosynthesis was elucidated, providing new strategies to regulate the metabolic network of C. glutamicum to achieve a further increase in 5-ALA production.

1. Introduction

5-Aminolevulinic acid (5-ALA) is an endogenous, non-proteinogenic, five-carbon amino acid, which is widely found in microorganisms, plant cells, and animal cells. It is an important precursor for the biosynthesis of heme, chlorophyll, vitamin B12, and other tetrapyrroles [1]. 5-ALA has recently received much attention, primarily due to its great contribution to the treatment of different cancers [2,3,4,5] and coronavirus disease 2019 (COVID-19) [6,7], as well as its wide applications as an efficient growth regulator and herbicide in agriculture [8,9]. It is projected that the global 5-ALA hydrochloride market will reach USD 145.7 million by 2028, a significant increase from USD 101.5 million in 2021 [10]. As the market demand for 5-ALA grows, environmentally friendly, inexpensive, and sustainable microbial production of 5-ALA is becoming increasingly popular among researchers compared to chemical synthesis [11].

There are two known natural 5-ALA biosynthesis pathways. One is the C4 pathway found in yeasts, mammals, and some photosynthetic bacteria [1]. In this pathway, 5-ALA synthetase (ALAS) condenses succinyl-CoA and glycine to directly synthesize 5-ALA [12]. The other is the C5 pathway encoded by the key genes gltX, hemA, and hemL, mainly found in some bacteria and plants, which produces 5-ALA from glutamate [1]. However, the C5 pathway is subject to complex feedback regulation by its downstream product heme [13,14] and shows a lower yield compared to the C4 pathway [15]. Several different microbial hosts were used to produce 5-ALA, including Escherichia coli, Corynebacterium glutamicum, and Saccharomyces cerevisiae [15]. Among them, C. glutamicum may have a higher potential to produce 5-ALA due to the absence of a glycine cleavage system [16] and better intrinsic acid tolerance.

The development of high-yield strains is crucial for the efficient production of 5-ALA, and a number of metabolic engineering strategies have been developed to construct optimal 5-ALA-producing strains, including the engineering of key enzymes, redistribution of central carbon fluxes toward precursors, regulation of downstream pathways, cofactor engineering, and transporter engineering [15]. For the high-level biosynthesis of 5-ALA, the efficient export of target products from the cell can avoid the accumulation of toxic substances, such as reactive oxygen species generated by 5-ALA, as well as reduce feedback inhibition to some extent [11]. The serine/threonine transporter RhtA was demonstrated to facilitate the efflux of 5-ALA in different recombinant strains [17]. Kang et al. [18] found that overexpression of rhtA increased 5-ALA accumulation by 45.9%, firstly identifying this 5-ALA exporter in E. coli. Recently, the introduction of a rhtA expression cassette using the T7 promoter enhanced the 5-ALA titer from 40 mg/L to 1.6 g/L in E. coli [19]. Auto-inducible expression using the two-component system HrrSA in response to extracellular heme was used to dynamically regulate the expression of rhtA in recombinant C. glutamicum A30, which increased 5-ALA production by 32.8%, reaching 3.16 g/L [20]. Thus, 5-ALA transport also plays an important role in 5-ALA synthesis in C. glutamicum. In addition to the introduction of heterologous RhtA as a transport strategy, changing the permeability of the cell wall to facilitate the transport of 5-ALA also yielded excellent results in C. glutamicum [21,22]. However, endogenous 5-ALA transport proteins in C. glutamicum have not been reported. Recent studies showed that the NCgl0580 gene, encoding a permease of the drug/metabolite transporter (DMT) superfamily, has significant effects on the efflux of serine, threonine [23], cysteine [24], and β-alanine [25], implying that it is a transporter with broad substrate specificity and great potential for the export of amino acids and derived substances.

In the present work, we first introduced the C4 pathway into C. glutamicum and obtained the 5-ALA-producing engineered strain JS. Then, the gene NCgl0580, with a positive effect on 5-ALA synthesis in C. glutamicum, was screened by BLAST and gene perturbation experiments. Surprisingly, 5-ALA production was significantly enhanced after the knockout of NCgl0580 (JSD1), resulting in a 2.49-fold increase compared to the control strain JS, implying that NCgl0580 may have new functions. Subsequently, RNA-seq and experimental validation of key targets were performed on the NCgl0580 deletion strain JSD1 to explore the potential function of this gene. This study reveals the transcriptional mechanism through which NCgl0580 knockout affects 5-ALA biosynthesis, contributing to a better understanding of the 5-ALA pathway and providing new strategies for metabolic engineering to increase 5-ALA production.

2. Materials and Methods

2.1. Strains, Media, and Culture Conditions

All strains used in this study are listed in Table S1. E. coli DH5α was used for the construction of plasmids. C. glutamicum J0 was used as the starting strain for engineering to overproduce 5-ALA. For plasmid construction, E. coli DH5α was cultured in Luria-Bertani broth (10 g/L tryptone, 10 g/L NaCl, and 5 g/L yeast extract) or on LB agar plates (2% w/v agar) at 37 °C and 220 rpm. C. glutamicum was grown in brain heart infusion (BHI) broth (74.4 g/L) or on BHI agar plates (2% w/v agar) for routine culture. For 5-ALA production, C. glutamicum was fermented in CGIII medium (21 g/L 3-(4-morpholino) propanesulfonic acid (MOPS), 2.5 g/L NaCl, 10 g/L tryptone, and 10 g/L yeast extract, pH 7.0). For the revival of C. glutamicum engineered strains stored at −80 °C, BHI agar plates were streaked and incubated for approximately 48 h. Then, single colonies formed on the plates were transferred into tubes containing 5 mL of BHI medium and grown at 30 °C and 220 rpm for 24 h, after which 1 mL of the resulting seed culture was used to inoculate 500 mL shake flasks containing 50 mL of CGIII medium with 10 g/L glucose and grown at 30 °C and 220 rpm for 12 h. The resulting secondary seed culture was used to inoculate 500 mL shake flasks containing 50 mL of CGIII medium and 10 g/L glucose to an initial OD₆₀₀ of 0.5 for further fermentation of 5-ALA at 30 °C and 220 rpm. Where indicated, 7.5 g/L of glycine and 0.1 mM IPTG (isopropyl b-D-1-thiogalactopyranoside) were added to the cultures when the OD₆₀₀ reached 5. Chloramphenicol (10 mg/L) and/or kanamycin (25 mg/L) were added where appropriate to promote plasmid retention. All reagents in this study were purchased from Sangon Biotech (Shanghai, China), except for the brain-heart infusion, which was purchased from Hopebio (Qingdao, China).

2.2. Construction of Plasmids

All the plasmids and primers used for plasmid construction are shown in Tables S1 and S2. For the construction of 5-ALA-producing strains of C. glutamicum, the vector pXPS was constructed. The P_sod(C131T) promoter was amplified from the genomic DNA of C. glutamicum strain CGS15 [26] by PCR using the primer pair pXPS-1/2. The codon-optimized coding sequence of hemA_RP was synthesized by GENEWIZ, Inc. (Table S3). The fragment was amplified using the primer pXPS-3/4. Then, the two fragments were fused by PCR. The resulting product was digested with KpnI and EcoRI and ligated between the corresponding sites of pXK. The pXPP, pXPT, and pXPG were constructed analogously. The plasmid pEC-XK99E carrying an IPTG-inducible promoter was used for plasmid-based gene expression. The plasmid pEC-NCgl0580 was constructed to overexpress the NCgl0580 gene. The NCgl0580 gene was amplified from the genome of C. glutamicum using primers NCgl0580-1/NCgl0580-2, and the resulting PCR product was digested with XbaI/SdaI and inserted between the corresponding sites of pEC-XK99E. The plasmids pEC-NCgl2065, pEC-NCgl0581, pEC-pyk2, pEC-Ncgl1445, pEC-iolT1, pEC-afuABC, pEC-pstSCAB, pEC-ugpQ, and pEC-aceA were constructed analogously. The two-step homologous recombination method using the suicide vector pD-sacB was applied to genetically modify C. glutamicum. To knock out NCgl0580, the upstream and downstream fragments of the target gene were amplified from the genome of C. glutamicum using the primers pD-NCgl0580-1/2/3/4, respectively, and fused by PCR. The fused fragment was then digested with SmaI/BamHI and inserted between the corresponding sites of pD-sacB, obtaining the plasmid pD-NCgl0580. The plasmids pD-NCgl2065, pD-NCgl0580TAA, pD-NCgl0581, PD-sdhCAB, and pD-gdhA were obtained by the same method using corresponding primers and restriction enzymes.

2.3. Analytical Methods

The concentration of glucose was quantified using an SBA-40E sensor instrument (Institute of Microbiology, Shandong, China). The concentration of cells was assessed by measuring the optical density at 600 nm (OD₆₀₀) using a conventional UV-visible spectrophotometer (TU-1901, PUXI, Beijing, China). The 5-ALA concentration was determined using a modified Ehrlich’s reagent [18]. The heme concentration was determined using a fluorescence-based assay [27].

2.4. Transcriptome Analysis

Strains JSD1 and JS (control) were cultured in CGIII medium supplemented with 10 g/L glucose and 7.5 g/L glycine at 30 °C and 220 rpm for 12 h to the mid-logarithmic phase. The cells were harvested by centrifuging the fermentation broth for 30 min at 4 °C and 5000 rpm and were washed once with RNase-free water. The collected cells from 6 samples (3 parallel samples for strains JSD1 and JS) were shock-frozen in liquid nitrogen and sent to GENEWIZ (Suzhou, China) for transcriptome sequencing using an Illumina HiSeq platform. The raw image data of the sequencing results were identified by base calling using the software Bcl2fastq (version 2.17.1.14) to obtain the raw sequencing data. The software Cutadapt (version 1.9.1) was used to pre-process the raw data, filter the low-quality data, and remove contamination and splice sequences to obtain the clean data. The annotated genome (GCF_000011325.1) of C. glutamicum ATCC 13032 was used as the reference for comparative analysis of the clean data. Gene expression calculations were performed using the software HTSeq (version 0.6.1), which calculates gene expression using the FPKM (fragments per kilo bases per million reads) method [28]. The analysis of genetic differences was performed using DESeq2 (version 1.6.3) of the Bioconductor package, which is based on a model with a negative binomial distribution. Pathway enrichment analysis was performed using KEGG pathways as a unit, and hypergeometric tests were applied to identify pathways in which the differentially expressed genes are significantly enriched compared to the whole genomic background.

2.5. Real-Time Quantitative PCR (RT-qPCR)

RT-qPCR was further applied to ensure the validity of the RNA-seq data. The cDNA was reverse-transcribed from total RNA obtained from GENEWIZ (Suzhou, China) and synthesized using a TransScript first-strand cDNA synthesis SuperMix (TransGen, Beijing, China) with random primers. The real-time fluorescence quantitative RT-qPCR was performed on a LightCycler^® 480 instrument (Roche, Basel, Switzerland), which was used to amplify and quantify the PCR products using TransStart^® Top Green qPCR SuperMix (TransGen, Beijing, China). The 16S ribosomal RNA of C. glutamicum was adopted as an internal reference. The fold change in transcript levels was calculated by the comparative 2^−∆∆Ct method. All primers used for RT-qPCR are listed in Table S4. The RT-qPCR of the target genes was conducted in parallel with triple replicates.

3. Results and Discussion

3.1. Enhancement of 5-ALA Synthesis by Overexpressing NCgl0580

Transport engineering plays an essential role in strain development by enabling more efficient export of target products to avoid cytotoxicity as well as feedback inhibition from excessive product accumulation inside the cell [15]. Since similar functions of homologs of E. coli exporters have been shown in C. glutamicum as reported before [23], we speculated that homologs of RhtA in C. glutamicum may also be involved in the transport of 5-ALA. In this study, the amino acid sequence of RhtA from E. coli was used to find homologs in C. glutamicum by BLAST, obtaining seven candidate genes. The top two proteins encoded by the genes NCgl0580 and NCgl2065 with the highest similarity were selected as potential 5-ALA transporters in C. glutamicum.

To identify their effects on 5-ALA biosynthesis, a 5-ALA-producing chassis strain of C. glutamicum was constructed. As the key enzyme of the C4 pathway, 5-ALA synthase (ALAS, encoded by hemA) directly affects 5-ALA synthesis in terms of the enzyme’s intracellular expression level, enzyme activity, and feedback inhibition by heme (Figure S1). Here, the ALAS mutant C75A/R365K (hemA_RP) from Rhodopseudomonas palustris [29] was utilized for 5-ALA synthesis because of its naturally higher enzymatic activity and excellent resistance to heme inhibition. Subsequently, the strong promoters P_glyA, P_pgk, P_tuf [30], and P_sod(C131T) [26] were selected to ensure the high-level expression of 5-ALA synthase in C. glutamicum. The codon-optimized coding sequence of hemA_RP was synthesized and expressed in C. glutamicum under the control of the above promoters using the plasmid pXK, resulting in the engineered strains JP, JT, JG, and JS, respectively. The results showed that P_sod(C131T)-driven hemA_RP expression was the most effective for 5-ALA production in C. glutamicum, as the corresponding strain accumulated 0.59 ± 0.01 g/L of 5-ALA, which was significantly higher than in the other strains (Figure 1A), without negative effects on cell growth and sugar consumption (Figure 1B). The engineered strain C. glutamicum JS was used as the chassis for the functional study of the genes NCgl0580 and NCgl2065 by perturbing their expression.

Strains JSC1 and JSC2 were obtained by overexpressing the genes NCgl0580 and NCgl2065 in strain JS, respectively, using the IPTG-inducible medium-copy plasmid pEC-XK99E as the vector backbone. After 48 h of shake-flask fermentation, the 5-ALA titers of strains JSC1 and JSC2 reached 0.69 ± 0.04 g/L and 0.49 ± 0.02 g/L, respectively, representing increases of 53.3% and 8.9% compared to the control strain JSC (0.45 ± 0.01 g/L) harboring the empty plasmid pEC-XK99E (Figure 1C). The 5-ALA titer of strain JSC was lower than that of strain JS, probably due to the effects of antibiotic stress and the cytotoxicity of IPTG [31]. As seen in Figure 1D, cell growth and sugar consumption were not affected by the overexpression of NCgl0580 and NCgl2065. Overexpression of NCgl0580 resulted in a significant increase in the 5-ALA titer compared with overexpression of NCgl2065. The protein encoded by the NCgl0580 gene, annotated as the permease of the DMT family in the Kyoto Encyclopedia of Genes and Genomes (KEGG) [32], was screened in this study as a homolog of RhtA (E. coli) and showed similar functions to RhtA that export serine and threonine in previous studies [17,23]. Thus, based on the results that overexpression of NCgl0580 increases the 5-ALA titer, it is speculated that the protein encoded by NCgl0580 may exhibit the same function as RhtA for the transport of 5-ALA.

3.2. Unexpected Effect of NCgl0580 Deletion on 5-ALA Biosynthesis

To further determine the roles of NCgl0580 and NCgl2065 in 5-ALA transport, these two genes were deleted in C. glutamicum, resulting in strains JSD1 and JSD2, respectively. The results showed that the deletion of NCgl2065 (JSD2) produced a slight increase in the 5-ALA titer. Strikingly, the NCgl0580 knockout strain JSD1 exhibited a 2.49-fold increase in the 5-ALA titer to 2.13 ± 0.03 g/L (Figure 2A). This result was not consistent with the expectation that the knockout of the transport protein would cause a decrease in the production of the target compound. In addition, plasmid-based overexpression of NCgl0580 in the JSD1 strain decreased 5-ALA production, reverting to the original phenotype to some extent (Figure S2). It therefore stands to reason that NCgl0580 may play additional roles in C. glutamicum.

To confirm that the high level of 5-ALA biosynthesis was caused by NCgl0580 knockout, we investigated the functions of genes upstream and downstream of NCgl0580 on the genome. Gene NCgl0581 was observed to be adjacent to NCgl0580 (Figure 2B) and was reported to encode a global transcriptional regulator in C. glutamicum [33]. In addition, a putative promoter P₁ was predicted by Neural Network Promoter Prediction (NNPP) of the Berkeley Drosophila Genome Project (BDGP) [34] (https://www.fruitfly.org/seq_tools/promoter.html (accessed on 30 December 2022)) between the genes NCgl0579 and NCgl0580 (Figure 2B). Thus, we speculated that this promoter may act directly on the latter gene, NCgl0581, after NCgl0580 knockout, leading to changes in the expression intensity of the transcriptional regulator NCgl0581 and consequent changes in 5-ALA biosynthesis. To investigate whether the significant increase in 5-ALA production was caused by the altered expression level of NCgl0581 or the deletion of NCgl0580 itself, three stop codons were inserted at the 5′ end of NCgl0580 in strain JS to terminate its translation, avoiding the effect on NCgl0581 gene expression due to the knockout of NCgl0580. The results showed that 5-ALA production was significantly elevated (Figure 2C), which implied that the substantial increase in 5-ALA biosynthesis after the knockout of NCgl0580 may be related to the inactivation of NCgl0580 itself. To exclude the effect of NCgl0581 on 5-ALA synthesis, the expression of NCgl0581 was directly perturbed to investigate its effect on 5-ALA biosynthesis. The knockout of NCgl0581 did not cause significant changes in 5-ALA biosynthesis in strains JS and JSD1 (Figure 2C). In addition, the overexpression of NCgl0581 from the plasmid pEC-XK99E with the inducible P_trc promoter did not cause significant changes in 5-ALA biosynthesis in strain JS when induced with different IPTG concentrations (Figure 2D). Therefore, it can be concluded that the NCgl0580 knockout caused a significant increase in 5-ALA production. Thus, the mechanism through which the deletion of NCgl0580 affected 5-ALA biosynthesis required further investigation.

3.3. Transcriptomic Analysis of the Effect of NCgl0580 Knockout on 5-ALA Synthesis

As already noted, the deletion of NCgl0580 was found to significantly increase 5-ALA biosynthesis. To better understand the underlying mechanism, whole-transcriptome sequencing was performed to assess the transcriptional response of the high-producing strain JSD1. The complete raw RNA-seq data are shown in Table S5. A total of 952 genes showed significant changes (|log₂ratio| ≥ 1, q-value (fdr and padj) ≤ 0.05) in their transcription levels between JS and JSD1 (505 up- and 447 downregulated), representing over 30.9% of the C. glutamicum genome (Figure 3A). Real-time quantitative polymerase chain reaction (RT-qPCR) was performed to identify changes in the expression of the 28 genes involved in representative processes and functions exhibiting high concordance with the RNA-seq data (Table S6). The results showed that the NCgl0580 knockout affected the expression of many genes. Zakataeva et al. [35] pointed out that cellular metabolites and their derivatives exported by the proteins of the DMT family (RhtA et al.) may play an important role in global regulation as the signal compounds involved in the intracellular communication triggering quorum sensing (QS) systems [36]. Many of these proteins are characterized by wide substrate specificity and are involved in exporting amino acids, purines, and other metabolites, which may play a signaling role [35]. Based on the previous description, we know that the protein encoded by NCgl0580 belongs to the DMT family and has high homology and functional similarity to RhtA. We therefore speculated that the knockout of NCgl0580 affects the normal efflux of signal molecules and interferes with intercellular signaling communication, thereby causing a systematic rearrangement of the cellular metabolism.

KEGG is the main public database for pathway annotation [32]. The metabolic and signal transduction pathways associated with the differentially expressed genes (DEGs) were determined by KEGG pathway enrichment analysis. Figure 3B shows the distribution of genes with significant differences between the experimental group and the control group in pathways (Table S7). They were categorized into five groups: metabolism, human diseases, genetic information processing, environmental information processing, and cellular processes. Metabolism and environmental information processing (ABC transporters) contained the majority of differentially expressed genes and were subjected to further in-depth study.

3.4. Redistribution of Central Carbon Fluxes toward Succinyl-CoA-Enhanced 5-ALA Biosynthesis

Based on the RNA-seq results, many DEGs were involved in different pathways of central carbon metabolism, including glucose transport, glycolysis, the glyoxylate cycle, the TCA cycle, as well as the glutamate biosynthesis pathway. As shown in Figure 4, the collective expression of genes related to glucose transport and glycolysis was differentially upregulated compared to the control strain. The NCgl1445 gene, encoding a putative major facilitator superfamily (MFS) transporter potentially associated with the import of sugar based on PTS [37], was upregulated 4.18-fold. The iolT1 gene was initially reported to encode an inositol transporter [38], and it was later shown that it might encode a PTS-independent glucose uptake system [39]. Notably, iolT1 was upregulated 2.30-fold. Furthermore, glK2 (encoding glucokinase), pfkA (encoding phosphofructokinase-1), pyk and pyk2 (both encoding pyruvate kinase), which are the rate-limiting enzymes involved in glycolysis, were upregulated from 1.10 to 2.42-fold. The resulting enhanced sugar transport and glycolysis will increase the rate of sugar catabolism, which is consistent with the accelerated sugar consumption of JSD1 (Figure S3). To investigate the effect of the above genes on 5-ALA biosynthesis, they were overexpressed using the plasmid pEC-XK99E. Compared with JSC, the 5-ALA titer of the overexpression strain increased by 14.7%, as well as by 8.5% after overexpression of NCgl1445 and pyk2 (JSCZ1 and JSCZ3). Notably, the iolT1 overexpression strain JSCZ2 produced 0.69 ± 0.02 g/L 5-ALA, representing a 47.2% increase (Figure 5A). It was reported that the PTS system is the primary mechanism of glucose uptake in C. glutamicum but may limit the ability of the organism to overproduce compounds with phosphoenolpyruvate (PEP) as a distal precursor [39]. Recent studies reported that the overexpression of the PTS-independent glucose uptake system (encoded by iolT1) in C. glutamicum may overcome these limitations [39]. For example, the biosynthesis of l-arginine [40], l-lysine [41], l-serine [42], the blue pigment indigoidine [43] (from glutamate), and β-alanine [25] were positively affected by overexpressing iolT1. At the same time, it was shown that the carbon fluxes from glycolysis to oxaloacetate may be channeled mainly through ppc-encoded phosphoenolpyruvate carboxylase rather than pyc-encoded pyruvate carboxylase [21] in C. glutamicum when the acetate synthesis pathway is knocked out. This implied that the enhancement of non-PTS glucose import would contribute to sugar uptake and increase the carbon fluxes toward oxaloacetate in C. glutamicum, which could effectively enhance 5-ALA synthesis. Enhancing the carbon fluxes from glycolysis into the TCA cycle is an effective strategy for increasing 5-ALA biosynthesis [21]. Therefore, it was speculated that the simultaneous enhancement of the PTS and non-PTS sugar uptake systems increased the glucose uptake capacity of the strain while saving PEP, and the enhanced glycolysis further increased the carbon fluxes into the TCA cycle, resulting in improved 5-ALA biosynthesis.

In the TCA cycle, the aceA gene involved in the glyoxylate pathway was upregulated 1.9-fold, while sdhCAB was downregulated from 2.53 to 4.18-fold and sucCD was upregulated from 1.19 to 1.71-fold (Figure 4). Notably, aceA encodes isocitrate lyase, which catalyzes the conversion of isocitrate to glyoxylate and succinate, while sdhCAB encodes the three subunits of succinate dehydrogenase, which catalyzes the conversion of succinic acid to fumaric acid. Changes in the transcript levels of these genes may increase the carbon fluxes at the succinate node [44,45]. The carbon fluxes were then directed toward succinyl-CoA via the succinyl-CoA synthase complex (encoded by sucCD). To test this conjecture, aceA and sucCD were overexpressed, while sdhCAB was knocked out. The results showed that 5-ALA production was enhanced by 16.5% and 15.6% in strain JSCZ4 overexpressing aceA and strain JSDZ1 with sdhCAB deletion, compared with the control strain, respectively (Figure 5A,B). Miscevic et al. [45] found that enhancing the glyoxylate cycle and blocking the oxidative branch of the TCA cycle can increase the carbon fluxes at the succinate node and ultimately increase 5-ALA biosynthesis. However, the overexpression of sucCD showed no effect on 5-ALA production. Choi et al. [46] experimentally confirmed that the production of 4-HB, which has the same precursor succinyl-CoA as 5-ALA, was enhanced by overexpression of sucCD under microaerobic conditions in E. coli. Additionally, in our previous work, overexpression of sucCD under microaerobic conditions in C. glutamicum only exhibited a weak effect on 5-ALA synthesis [47]. Therefore, we speculated that the reason why 5-ALA production was not affected by overexpression of sucCD under aerobic conditions in this study may be due to the different genotypes of chassis and production conditions. The glutamate dehydrogenase (GDH)-encoding gene gdhA was significantly downregulated 2.21-fold (Figure 4). GDH, the enzyme which catalyzes the conversion of α-ketoglutarate (α-KG) to glutamate, links the carbon and nitrogen metabolisms [48]. To assess the effect of gdhA downregulation on 5-ALA biosynthesis, gdhA was knocked out in JS, resulting in strain JSDZ2. The 5-AlA titer of JSDZ2 increased by 12.7% compared with JS (Figure 5B). Ge et al. [49] reduced glutamate production by knocking out gdhA and significantly increased 5-ALA production by introducing the C4 pathway. Thus, downregulation of gdhA might reduce carbon consumption for glutamate biosynthesis and enable more carbon to flow from α-KG to the C4 pathway, which improves the synthesis of 5-ALA.

Based on these findings, it can be inferred that the fluxes redistribution of central carbon metabolism induced by NCgl0580 knockout in C. glutamicum facilitated the flow of carbon toward the C4 pathway precursor succinyl-CoA, which contributed to 5-ALA biosynthesis.

3.5. Enhancement of Iron and Phosphate Uptake Improves 5-ALA Synthesis

ATP-binding cassette (ABC) transporters are ubiquitous membrane proteins that couple the transport of diverse substrates across cellular membranes to the hydrolysis of ATP [50]. Many DEGs were enriched in related pathways based on our RNA-seq data. Interestingly, the iron and phosphate transporters were significantly upregulated (Figure 6). In nature, iron is essential for 5-ALA synthesis, and it has a major impact on the aerobic metabolism of cells [51]. Phosphorus (P) is an essential component of living organisms and is closely related to energy and central carbon metabolism [52]. Therefore, we hypothesized that the upregulation of genes related to these processes also affected 5-ALA biosynthesis.

The afuABC operon encoding an ABC-type Fe³⁺ transporter was upregulated significantly (from 2.18 to 4.48-fold) (Figure 6). As previously reported, iron plays an important regulatory role in the biosynthesis of 5-ALA. Zhang et al. [53] optimized the iron addition (7.5 mg/L), which resulted in a significant increase in the 5-ALA titer. To investigate the effect of afuABC upregulation on 5-ALA synthesis, afuABC was overexpressed, resulting in an increase of the 5-ALA titer by 17.8% (Figure 5A). This result suggested that the overexpression of afuABC may promote iron uptake, which in turn enhanced 5-ALA synthesis. To further confirm this conjecture, different concentrations of Fe³⁺ were added to the fermentation medium. The control strain JSC showed the highest 5-ALA production with the exogenous addition of 27 mM Fe³⁺, producing 0.66 ± 0.01 g/L of 5-ALA, which was 40.8% higher than without Fe³⁺ addition. The highest 5-ALA production of the afuABC overexpression strain was measured in the presence of 18 mM Fe³⁺, with a 5-ALA titer of 0.70 ± 0.06 g/L, representing a 26.7% increase compared with no Fe³⁺ addition (Figure 5C). Notably, strain JSCZ5 overexpressing afuABC obtained a higher 5-ALA titer at lower concentrations of iron than strain JSC, indicating that the overexpression of afuABC enhanced the iron uptake capacity, which eliminated the need for large amounts of extracellular iron addition, improving aerobic cellular metabolism, and consequently promoting 5-ALA biosynthesis.

The transcription levels of pstSCAB (encoding an ABC-type Pi uptake system), ugpQ (encoding glycerophosphoryldiester phosphodiesterase), and ugpAEBC (encoding an ABC-type sn-glycerol 3-phosphate uptake system) were increased more than threefold (Figure 6). The upregulation of these genes seems to promote the uptake of phosphate by the cells [52]. To investigate the effect of the above genes on 5-ALA synthesis, they were overexpressed using the pEC-XK99E vector backbone. The pstSCAB and ugpQ overexpression strains JSCZ6 and JSCZ7 exhibited 13.4% and 9.9% increases in 5-ALA production, respectively (Figure 5A). The ABC-type Pi uptake system encoded by pstSCAB facilitates the uptake of inorganic phosphate [52]. The glycerophosphodiester phosphohydrolase encoded by ugpQ catalyzes the hydrolysis of glycerophosphodiester to sn-glycerol 3-phosphate, which can be utilized as an organic phosphate [54]. In addition, P metabolism is closely related to energy and central carbon metabolism [52]. The interaction between P metabolism and C metabolism is particularly important for amino acid production in C. glutamicum because amino acids are derived from intermediates of central carbon metabolism [52]. Here, the overexpression of the transporter may have affected the energy and central carbon metabolism by increasing the uptake of phosphate, thereby enhancing the synthesis of 5-ALA.

3.6. Effect of Multiple Gene Expression Perturbation on 5-ALA Synthesis

As the metabolic pathway in microbial metabolism is massively complicated and interconnected, the interaction between gene up- and downregulation would affect the global metabolic fluxes. To investigate the effect of interactions between effective genes in previous single-gene validation experiments on 5-ALA synthesis, the genes NCgl1445, iolT1, pyk2, aceA, afuABC, pstSCAB, and ugpC were overexpressed based on the sdhCAB knockout strain JSDZ1 and the gdhA knockout strain JSDZ2, respectively. As can be seen from Figure 7, overexpression of the above genes enhanced 5-ALA production to different extents after deletion of sdhCAB or gdhA. In particular, overexpression of iolT1 had the most significant increase in 5-ALA production in strains JSDZ1 and JSDZ2 by 37.2% and 38.4% (0.80 ± 0.01 g/L and 0.77 ± 0.03 g/L, respectively). Compared with JSC, the 5-ALA titer of strains DZ1C2 and DZ2C2 increased by 70.5% and 64.1%. The results indicate that the combined expression perturbations of effective genes further enhanced 5-ALA production. Therefore, it is speculated that the above-mentioned genes, whose expression was changed by the deletion of NCgl0580, interact to promote global metabolic fluxes toward 5-ALA. The mechanism of NCgl0580 knockout affecting 5-ALA synthesis can be investigated in depth by establishing a gene-interaction network in the future.

4. Conclusions

In this study, overexpression of NCgl0580 was demonstrated to facilitate the extracellular accumulation of 5-ALA in C. glutamicum. More importantly, the NCgl0580 knockout resulted in a remarkable 2.49-fold increase in 5-ALA production to 2.13 g/L. Transcriptomic analysis combined with the validation of key DEGs showed that the changes in transcription levels of genes caused by the NCgl0580 deletion redistributed the fluxes of central carbon metabolism to enhance the supply of the direct 5-ALA precursor succinyl-CoA. Genes encoding ABC transporters for iron and phosphate uptake were upregulated, improving aerobic metabolism, energy balance, and 5-ALA biosynthesis. Encouragingly, it was confirmed for the first time that overexpression of NCgl1445, iolT1, afuABC, pstSCAB, and ugpQ could facilitate 5-ALA biosynthesis. This work elucidated the remarkable effect of NCgl0580 deletion on 5-ALA biosynthesis in C. glutamicum, which provides a reference for further study on the transcriptional mechanisms that control 5-ALA biosynthesis. The elucidation of phenotype-related targets also provides a new direction for metabolic engineering to construct high-yield strains.