Skip to main content
Download PDF

D-optimal design model and biosynthetic pathway for gentamicin production by Micromonospora purpureochromogenes NRRL B-16094

BMC Microbiology volume 25, Article number: 309 (2025) Cite this article

Abstract

Background

Micromonospora purpureochromogenes NRRL B-16094, a natural producer of gentamicin (GEN), a 5,6-diglycosylated 2-dexoystreptamine-aminoglycoside antibiotic (2DOS-AGA) broad-spectrum bactericidal activity. In literature, limited studies are concerned with the biosynthetic route and various cultural conditions influencing GEN production.

Methods

Therefore, this study aimed to explore the GEN biosynthesis pathway and compare it to that of fortimicin and kanamycin. In addition, four key environmental conditions influencing GEN production were statistically optimized using response surface D-optimal design (DOD). Herein, the biosynthetic pathway of GEN was proposed based on the biochemistry of the identified genes/proteins within the gene cluster. Comparing the GEN-biosynthetic gene cluster to that of kanamycin and fortimicin suggested that gentamicin biosynthesis could have originated from a combination of biosynthetic pathways of both antibiotics.

Results

For the optimization experiments, culture media 4 (CM4) and 6 (CM6) gave the highest specific productivity at 6.36 and 3.80 µg/mg, respectively. A DOD quadratic model was successfully generated to optimize four key environmental factors. Predicted and experimentally confirmed optimized factors were an initial pH of 7, an incubation temperature of 30˚C, and an agitation of 300 rpm for 10 days. This resulted in a 13.5-fold increase (289.5 µg/mL) over that produced by the basic CM1 production medium (21.4 µg/mL) and 2.4 times (over that obtained by CM4 (123.7 µg/mL) as verified by HPLC analysis.

Conclusion

DOD is an efficient tool for optimizing GEN. Accordingly, the optimized conditions are highly advisable during the scaling up of GEN production by M. purpureochromogenes NRRL B-16094.

Peer Review reports

Background

Aminoglycoside antibiotics (AGAs) are, with very few exceptions, products of higher differentiating actinomycetes. The kanamycins (KANs) and gentamicins (GENs) are members of the most relevant aminoglycoside subclass of the 2-deoxystreptamine aminoglycoside antibiotics (2DOS AGAs) [1,2,3]. Gentamicins and fortimicins (FTMs) are products of Micromonospora sp. while kanamycins are products of Streptomyces sp. On the other hand, GENs and KANs share the key intermediate, the paromomine during their biosynthesis however, FTMs have a fortamine intermediate [3, 4]. In the biosynthetic pathway, GENs were found to share some biosynthetic steps with both fortamine (such as FTMs) and 2-deoxyfortamine-containing AGAs (such as istamycins, ISMs) which contain a different type of aglycone [1, 4, 5]. The biosynthetic gene clusters of the three AGAs have been fully sequenced and deposited in the NCBI GenBank database [1]. Therefore, exploring the biosynthetic route of the respective AGAs will help researchers make certain modifications of the respective biosynthetic pathways which is known as combinatorial biosynthesis to get new members of these antibiotics exerting more potent bactericidal activities, particularly against multidrug-resistant pathogens [3].

GENs have been applied in various sectors, mostly to treat immunocompromised patients’ severe hospital-acquired infections, particularly those caused by MDR Gram-negative and Gram-positive pathogens [6, 7]. Several biosynthetic feeding studies were performed using certain isotope-labeled precursors to confirm the biosynthetic routes of both GENs and FTMs [8, 9]. In their biosynthetic pathways, the two KANs and GENs share the pseudodisaccharide paromamine. However, KANs are different in their chemical structures from GENs by some extra alterations that GENs share with the FTMs [1, 3, 10]. As a result, both KANs and GENs undergo 3”-amination in the neutral sugar unit that is joined to the 2-DOS aminocyclitol’s 6-hydroxyl group in the subsequent phases of modification. Conversely, the GENs and FTMs have similar 3’,4’-didehydroxylations and 6’-C-methylations [1, 7].

Through the optimization of numerous interrelating environmental factors impacting antibiotic production, computer-aided tools employing response surface methodology (RSM) are utilized to achieve maximal production of the target bioactive metabolites [11]. Using up to five levels for each variable with a minimal probability of missing data, the D-optimal design (DOD) is the most common in RSM experiments [12,13,14]. Recently, RSM was used to optimize cultural conditions influencing the production of ISMs, resulting in a 31-fold rise as compared to un-optimized factors [15]. In this study, the proposed biosynthetic pathway of GENs was illustrated in terms of the identified genes/proteins being involved in the biosynthesis and compared to those involved in related AGAs such as FTMs and KANs. In addition, various environmental and cultural media compositions previously known to influence AGA production were optimized using DOD.

Materials and methods

Microorganisms

Micromonospora purpureochromogenes NRRL B-16,094 (GENs producer) was provided by the USA-ARS Culture Collection (NRRL), Peoria, IL, USA. It was cultured in YMG (M65) broth (DSMZ, Braunschweig, Germany) incubated at 28˚C for 5 days, and stored as a lyophilized dried heavy suspension in bovine serum as recommended by the supplier. The standard Bacillus subtilis DSM 618 was used to calculate the antibacterial activity of the produced GENs [16].

Identification of the GEM biosynthetic gene cluster

The GENs biosynthetic gene cluster was sequenced, assembled, and annotated as submitted into the NCBI GenBank database under the accession code, AJ628149.4 (https://www.ncbi.nlm.nih.gov/nuccore/AJ628149.4) [1]. FramePlot 2.3.2 software was used to determine the biosynthetic open reading frames (ORFs), where BLASTn and BLASTp were used to analyze the nucleotide and amino acid percentage identity, respectively. The ORFs were annotated using FramePlot 2.3.2 https://nocardia.nih.go.jp/frameplot/ [17]. BLAST https://blast.ncbi.nlm.nih.gov/Blast.cgi (accessed in October 2024) was used to analyze its associated ORF functions [18]. The pDRAW32 program, which may be downloaded for free, was used to construct and illustrate the corresponding ORFs: https://www.acaclone.com/(accessed October 25, 2024).

The GEN biosynthetic pathway

Based on the biosynthetic biochemical role of each gene or protein that has been previously examined and validated in the literature, as well as the percentage of similarity to proteins or enzymes that are known to catalyze comparable biosynthetic roles, the biosynthetic pathway of GEN was proposed and characterized [5]. To shed light on the genetics and the processes involved in the formation of GENs, an effort was prepared to point out and examine the associated gene clusters. The foundation of this method was the understanding that all the genes involved in the synthesis of an antibiotic in Actinobacteria are usually contained in a single gene cluster. Furthermore, the majority of cases may be analyzed using partial sequence information obtained from the database because resistance and a few GEN biosynthesis genes have already been reported from our lab and others.

The gentamicin (GEN) versus Kanamycin (KAN) and Fortimicin (FTM) biosynthetic pathways

The GEN, KAN, and FTM biosynthetic gene clusters have been sequenced, deposited to the NCBI GenBank database under the accession codes AJ628149, AJ628422, and AJ628421, respectively. The open reading frames (ORFs) present in each gene cluster have been detected and analyzed in terms of the function of each ORF in the biosynthetic pathways as well as the % identity and degree of conservation and location of the DNA segment harboring the respective antibiotic gene clusters. FramePlot 2.3.2 software [17], BLASTn and BLASTp [18], Pairwise sequence alignment(https://www.ebi.ac.uk/jdispatcher/psa) were used in the analysis.

Optimization of GEN production

Culture media

Trypticase soy broth (TSB broth) [19] as the basic culture medium was initially used to produce GEN by M. purpureochromogenes NRRL B-16,094, as was previously documented [20]. However, as shown in Table S1, an additional six different Streptomyces/Micromonospora culture mediums that were previously used for aminoglycoside production were examined to maximize GEN production [15]. The compositions of the tested culture media are displayed in Table S1.

Evaluation of GEN production and cell growth

About 200 µL of the culture supernatant of the tested production medium was centrifuged and sterilized by membrane filtration after incubation (at the specified time of each experiment) as previously reported [20]. This was done in order to measure the antibacterial activity of the culture-free supernatant and determine whether GEN was produced. The resultant culture-free supernatant (150 µL) was assessed against Bacillus subtilis DSM 618 as formerly mentioned [21]. The amount of GEN produced by each culture was measured by measuring the inhibition zone in millimeters (IZ) after a 24-hour incubation period at 37˚C. A reference calibration curve was created using the standard GEN at different concentrations (µg/mL) by plotting known concentrations (represented as log concentration) of the standard GEN (CAS No. 1405-41-0, Merck, Darmstadt, Germany) against average inhibitory zone widths. The culture growth was calculated using the culture dry weight, as was formerly mentioned [15, 20, 22]. The specific productivity in µg/mg was calculated by dividing the observed GEN concentration (µg/mL) by the corresponding dry cell weight (mg/mL) [15].

HPLC analysis for measuring the produced gentamicin

The quantity of GEN generated in the culture supernatant of M. purpureochromogenes NRRL B-16,094 was measured by HPLC analysis. Standard gentamicin (GEN; CAS No. 1405-41-0, Merck, Darmstadt, Germany) solutions at varying concentrations (20–3560 µg/mL) were produced in the employed mobile phase to create the standard curve. The Zorbax ODS (25 cm x 4.6 mm I.D.) chromatographic column (DuPont, Wilmgton, DE, USA) equipped with a C18 (3 cm x 4.6 mm I.D.) was used for the analysis, which was conducted using an HPLC Shimadzu CBM 40 at 30 °C with a UV detector set at 260 nm. For 60 min, 10 µL of a mobile phase consisting of a 1:1 mixture of HPLC-grade acetonitrile and methanol was injected at a flow rate of 1 mL/min. As previously mentioned [23], derivatization was performed using 2,4 dinitrofluorobenzene (DNFB) of 98% purity (Rochester, NY, USA).

Culture media affecting GEN production

A 250 mL baffled Erlenmeyer flask was filled with 25 mL (TSB) of M. purpureochromogenes NRRL B-16,094. The mixture was then incubated at 28 °C at 200 rpm to produce seed culture. About 5% v/v of this seed culture (0.5 × 107) was inoculated after three days of incubation with a total of 25 mL of each of the tested culture media (Table S1), which had been prepared in a 250 mL baffled Erlenmeyer flask and incubated at 28˚C for six days [20]at a shaking incubator (200 rpm) [19]. As previously reported, 200 µL was taken at the end of the 6-day incubation period to measure the amounts of GEN generated and the dry cell weight [15].

Statistical optimization

In these studies, the optimal culture medium and incubation duration were selected from previous trials. For statistical optimization, The D-optimal design (DOD) of the Response Surface Method (RSM) was utilized using the statistical software application Design Expert v.7 [14, 24, 25]. The DOD was used in a quadratic design model to assess four factors: intial pH (coded as variable A), incubation temperature (coded as variable B), and agitation rate (coded as variable C) and incubation time (factor D). A set of 23 runs is produced by the DOD, as seen in Table 1. The analysis of variance (ANOVA) for the suggested model was used to acquire the statistical analysis’s findings [26].

Table 1 Different runs and the tested variable generated by DOD design
Full size table

Experimental confirmation of statistical optimization

As stated in the approach, the unoptimized condition was compared to the ideal circumstances generated by the DOD in the lab based on the model’s significance (determined using ANOVA within the Design expert application).

Statistical analysis

The mean and standard deviation were calculated using Microsoft 365’s Excel software, and each lab experiment was run in triplicate. The statistical analysis for the suggested model was determined using ANOVA [26].

Results

GEN biosynthetic gene cluster versus FTM and KAN gene clusters

The GEN, KAN, and FTM biosynthetic gene clusters have been sequenced, assembled, annotated, and submitted to the NCBI GenBank database under the accession codes, AJ628149, AJ628422, and AJ628421, respectively. The open reading frames (ORFs) present in each gene cluster have been detected and analyzed as depicted in Fig. 1. The following succinctly describes the cluster structures’ immediate interpretation for the three clusters under consideration: (i) comparing the three clusters reveals a surprisingly low overall degree of conservation; (ii) the gene order inside the clusters is likewise not generally conserved, for example, few gene subsets exhibit a somewhat preserved order, such as the gene set forHIJ corresponding to genHJ or the subcluster kanS1CD2M2DS2 related to genS1CD2M2D1S2. The alignment of the DNA segment coded for the forHIJ from the fortimicin biosynthetic gene cluster (NCBI accession code, AJ628421) with the DNA segment coded for genHJ from the gentamicin biosynthetic gene cluster (NCBI accession code, AJ628149) showed almost 95.0% conservation of the nucleotide sequence is displayed in Fig. S1. In addition, the alignment of the DNA segment coded for the kanS1D2M2D1S2 from the kanamycin biosynthetic gene cluster (NCBI accession code, AJ628422) with the DNA segment coded for genS1D2M2D1S2 from the gentamicin biosynthetic gene cluster (NCBI accession code, AJ628149) about 97% conservation of the nucleotide sequence as depicted in Fig. S2. (iii) the adjacent genomic regions are not preserved at all, but they are highly conserved in the full-genome sequenced S. coelicolor A3(2).

Fig. 1
figure 1

Biosynthetic gene cluster of gentamicin (GEN) produced by M. purpureochromogenes NRRL B-16,094 compared to that of fortimicin and kanamycin gene clusters. Open reading frames (ORFs) that are conserved in gentamicin (gen), kanamycin (kan), and fortimicin (for) biosynthetic gene clusters are colored red. ORFs conserved in gen- and kan- clusters are colored in yellow. ORFs conserved in gen and for clusters are colored in green. ORFs reside outside the ACAGA gene clusters (dotted blue)

Full size image

With the exception of the genB1/forB and genQ/forD2 encoded pairs of identical proteins, most of the gene/protein families are conserved in the gene clusters for the majority of the 2DOS-AGAs, which form a paromamine precursor in their biosynthetic routes. The following genes were covered by the pairs of strongly conserved segments between the kan- and gen-clusters: kanE/genE, kanS2/genS2, kmr, kanM1/gmrA, and genM1. The following sets of equivalent gene sets were also covered by the DNA segments of conservation between the for- and gen-clusters: forHIJ/genHJ, forD2/genQ, forPBK/genKB3PB4 (genes B3 and B4 appear to have originated from a recent gene duplication), forT/genI, fosDEFG/genYAFG, fosC/genW, fmrB,fosA/gmrB, and genP, as illustrated in Fig. 1. These findings suggest that GEN biosynthesis could originate from a combination of FTM and KAN biosynthetic pathways, and this is evidenced by the resemblances in the modified sugars and aglycone moieties of the respective AGAs as depicted in Fig. 2.

Fig. 2
figure 2

Chemical structure of gentamicin, fortimicin and kanamycin. Arrows indicate similarities in structure in both sugar and aglycone moieties among three aminoglycoside antibiotics (AGAs)

Full size image

Biosynthetic pathway of GEN

Based on the previously biochemically identified genes and proteins involved in 2-deoxystreptamine (2DOS), basic cyclitol, and specific glycosylation or sugar modification processes, the putative GEN biosynthesis route was outlined in Fig. 3. Based on amino acid sequence and structural resemblance to homologous proteins or related ACAGAs that participate in comparable enzymatic activities, the other genes and proteins that are present in the GEN biosynthesis gene cluster but have not yet been biochemically identified were added to the pathway. However, as illustrated in Fig. 1, the GEN biosynthetic route theory is based on an examination of the gene products that the gen-cluster encodes.

Fig. 3
figure 3

The proposed biosynthetic pathway of gentamicin in its natural producer, M. purpureochromogenes NRRL B-16,094. AT, aminotransferase, GT, glycosyltransferase, GEN, Gentamicin, GEN-C (Gentamicin C-complex), SSM, Sisomycin, VDM, Verdamicin

Full size image

Production optimization of gentamicin (GEN)

Assessment of GEN concentration

A standard curve was created by plotting the GEN concentrations against the Area Under the Curve (AUC) (Fig. S3). The GEN- was computed using the suggested linear equation. The formula for a concentration was y = 0.205 x + 67.86. x is the GEN concentration in µg/mL, and y is the AUC with R2 = 0.9925 (Fig. S3). GEN concentration was also calculated from the following equation: Y (inhibition zone diameter (mm) = 17.467X -2.9147 where X is antilog GEN concentration in µg/mL with R2 = 0.939 as shown in Fig. S4.

Effect of culture media

Table 2 displays the dry cell weight and GEN generated by M. purpureochromogenes NRRL B-16,094. The highest specific productivity was found in CM4 and CM6, with respective values of 6.36 and 3.8 µg/mg and corresponding GEN productivity of 103.9 and 66.23 µg/mL. As a result, CM4 was selected as the ideal culture medium for the next examinations.

Table 2 Corresponding GEN concentration produced by Micromonospora purpureochromogenes NRRL B-16,094 and specific productivity obtained from the tested culture media
Full size table

D-Optimal design (DOD)

Using the DOD, four parameters—initial pH, incubation temperature (˚C), agitation rate (rpm), and incubation period (days)—have been optimized. Both the predicted and observed responses for the 23 runs have been documented, as seen in Table 3. According to ANOVA analysis, the model’s F-value was 11.21, meaning it is significant (Table 4). ANOVA study confirmed the applicability of the model and showed that the generation of GEN was significantly influenced by the initial pH (factor A), agitation (factor c), and incubation duration (factor D). Additionally, CD and A2 were significant (P value < 0.05), according to the data. However, the measured incubation temperature range of 25–37 ˚C showed no discernible effect on GEN production in terms of the recorded inhibition zone, (P value, 0.3001). Eventually, the equation generated by DOD for evaluation of the amount of GEN produced is:

Table 3 Runs of the D-optimal design for optimizing GEN production showing observed and predicted responses
Full size table
Table 4 ANOVA analysis of the GEN production using the D-optimal design quadratic model
Full size table

GEN activity (IZ mm) = --67.25875 + (20.18893 * Initial pH) + (0.052546 *Incubation temperature) + (0.088 * Agitation rate) + (2.83075 * Incubation Time)– (9.58709E-003 * Agitation rate* Incubation Time)– (1.50768 * Initial pH2).

Contour and 3D-dimensional plots

The contour (Fig. S5-S8, supplementary file) and three-dimensional plots of the experienced four variables including, initial pH versus incubation temperature (Fig. 4a), Incubation temperature versus agitation rate (Fig. 4b), agitation rate versus incubation time (Fig. 4c), and incubation time versus initial pH (Fig. 4d) showed that the maximum GEN production of M. purpureochromogenes NRRL B-16,094 was attained at an initial pH of 7.0, an incubation temperature of 30˚C, an agitation rate of 300 rpm and incubation time 10 days using the production medium CM4.

Fig. 4
figure 4

Three-dimensional (3D) surface plots for the effects of: a) initial pH versus incubation temperature; b) incubation temperature versus agitation; c) agitation versus incubation time; d) incubation time versus initial pH on gentamicin (GEN) produced by M. purpureochromogenes NRRL B-16094

Full size image

Validation of the generated model

As seen in Fig. 5a, box-cox plot analysis demonstrated that the model was applicable and significant (lambda current = 1 and the best = 0.6), and no further transformation was needed. Additionally, the residuals vs. run plot (Fig. 5b) showed that the model fits the data, and the plot showed that the actual values were very similar to the predicted ones (Fig. 5c). Additionally, the residuals’ normal plot, which was created by plotting the internally studentized residuals against normal percentage probabilities, demonstrated that the residuals had a normal distribution and that the points drew a straight line, as seen in Fig. 5d.

Fig. 5
figure 5

Model Diagnostics of the D-optimal design (DOD): (a) Box-Cox plot for Power Transforms; (b) Residuals vs. Run plot; c); Predicted vs. actual plot; d) Normal plot of Residuals on GEN produced by M. purpureochromogenes NRRL B-16094

Full size image

Laboratory verification of the optimized factors generated by the DOD model

To achieve maximum GEN production as recommended by DOD design, the optimal levels of the four variables under test—initial pH 7.0, incubation temperature 30˚C, incubation time 10 days, and agitation rate 300 rpm—were selected. This resulted in an average inhibitory zone of 35 mm, proving the model’s validity. This amount was equivalent to 245.5 µg of GEN per milliliter. This led to a 2.4-fold increase in comparison to the optimized culture media (CM4, 7 days; 103.9 µg/mL) and an 11.5-fold rise in comparison to the unoptimized culture conditions (TSB broth for 6 days incubation; 21.4 µg/mL).

Experimental confirmation using HPLC

Figure S7 shows the HPLC chromatogram of standard GEN and the one generated in the M. purpureochromogenes NRRL B-16,094 culture supernatant. Standard GEN’s retention time was 4.5 min. From 4.1 to 4.5 min, the generated GEN’s retention period in the culture supernatant was noted. HPLC analysis was used to verify the generated GEN, and the resulting AUC of the generated GEN was computed using the defined calibration curve as depicted in Fig. S9. As depicted in Fig. 6, the model produced 289.2 µg/mL (corresponding to AUC 1079) of GEN, whereas CM4 produced 123.7 µg/mL (equivalent to AUC 512.8), resulting in a roughly a 13.5-fold increase in GEN production as compared to basic production conditions (21.4 µg/mL, using CM1 production medium, incubation time 6 days, 28 ˚C incubation temperature and 200 rpm agitation).

Fig. 6
figure 6

Graphical presentation of the GEN produced by M. purpureochromogenes NRRL B-16,094 under un-optimized, optimized culture medium (CM4) and optimized condition using the D-optimal model

Full size image

Discussion

The first objective of this study is to explore the proposed biosynthetic pathway of GEN and compare it to those of KAN and FTM antibiotics. The identified genes and proteins involved in the biosynthesis of GENs were used in this work to show the suggested biosynthetic route, and they were compared to those involved in related AGAs like FTMs and KANs. Starting from the expectation that the 2DOS aminoglycoside gentamicin resulted from a combination of gene sets for the production of KAN- and FTM-type, the working hypothesis that the GEN (gen) cluster should contain both strong and alternative relationships to both production gene clusters of the other two gene clusters (kan, for) was tested by isolation and sequence analysis of the three gene clusters and their immediate genomic vicinity. For this purpose, the genomic DNAs harboring the biosynthetic gene clusters of S. kanamyceticus DSM 40,500 (kanamycin-producing, AJ628422), M. purpureochromogenes NRRL B-16,094 (gentamicin-producing, AJ628149), and M. olivasterospora DSM 43,868 (fortimicin-producing, AJ628421), were compared and analyzed.

The explanation of the cluster structures for the three antibiotics is: (I) the overall degree of conservation, when all three clusters are compared, is unexpectedly low; (II) the gene order inside the clusters is also not conserved at large, i.e. only a few subsets of genes show a somewhat retained order, e.g. the subcluster kanS1CD2M2DS2 is related to genS1CD2M2D1S2, or the gene set forHIJ corresponds to genHJ; (III) the adjoining genomic regions are not conserved however, they are enriched with ORF’s that are highly conserved in the completely analyzed streptomycete genomes, S. coelicolor A3 [27] and S. avermitilis [28]. Interestingly, the genomic DNA sequence from M. purpureochromogenes NRRL B-16,094 (M. echinospora strain ATCC 15835) was obtained comprising 43 ORFs as depicted in the results. Two additional groups from M. echinospora strain ATCC 15,835 that have previously been submitted to the NCBI database (NCBI AY524043.1 and AJ575934.2) also studied a subsegment of the sequenced genomic area presented here. The sequences that overlap is nearly identical: The disclosed sequence comprises a portion of an rRNA operon that appears to surround the GEN-production cluster on one side. Because it shares the closest link with the rrnE operon of S. coelicolor A3(2) [27]. Our results allow us to hypothesize that the gen-cluster may have essentially resulted from a merger of the progenitor gene clusters for and kan.

The following is a summary of the findings: (I) The gen-cluster DNA displayed notable stretches of similarity to both other DNAs in mutually exclusive regions, whereas the kan- and for-cluster DNAs did not exhibit any notable similarities. In general, the two flanking segments of the gen-cluster contained segments of strong similarity towards the for-cluster, while the central part of the gen-cluster essentially contained the similarity regions towards the kan-cluster but was broken up by another stretch of for-similarity. This made the distribution of similar regions rather complex. (II) Numerous rearrangements, including insertions of non-conserved genomic material, deletions, inversions, and displacements, could be found within these sequences. (III) The following genes were covered by the pairs of highly conserved regions between the gen- and kan-clusters: kanE/genE, kanS1CD2M2/genS1CD2M2, kanS2/genS2, kmr, kanM1/gmrA,genM1. (IV) The forHIJ/genHJ, forD2/genQ, forPBK/genKB3PB4, forT/genI, fosDEFG/genYAFG, fosC/genW, fmrB,fosA/gmrB, and genP were among the sets of analogous gene sets spanned by segments of conservation between the for- and gen-clusters. (V) According to the data on the gene clusters for the production of neomycin, paromomycin, ribostamycin, lividomycin, tobramycin, apramycin, and hygromycin B, the level of conservation between the gen/for pairs of equivalent genes/proteins is generally much higher (55 to 95%), and the majority of these genes do not occur in any other of the currently analyzed AGA gene clusters [1].

The immediate interpretation of the cluster structures for the three clusters under investigation is summarized as follows: The gene order within the clusters is also not generally conserved; for instance, only a small number of gene subsets show a somewhat preserved order, such as the gene set forHIJ corresponding to genHJ or the subcluster kanS1CD2M2DS2 related to genS1CD2M2D1S2. Comparing the three clusters reveals a surprisingly low overall degree of conservation. The nucleotide sequence conservation of the forHIJ DNA segment from the fortimicin biosynthesis gene cluster (NCBI accession code: AJ628421) was nearly 95.0% when it was aligned with the genHJ DNA segment from the gentamicin biosynthesis gene cluster (NCBI accession code: AJ628149). Furthermore, as shown in the results (supplementary data), there is approximately 97% nucleotide sequence conservation when the DNA segments coding for genS1D2M2D1S2 from the gentamicin biosynthesis gene cluster (NCBI accession code, AJ628149) and kanS1D2M2D1S2 from the kanamycin biosynthesis gene cluster (NCBI accession code, AJ628422) are aligned. The obtained data support our hypothesis that gentamicin biosynthesis could have originated from a combination of biosynthetic pathways of kanamycin and fortimicin antibiotics.

The second objective of this study is to statistically optimize GEN production by its natural producer, M. purpureochromogenes NRRL B-16094. Accordingly, in this study, the production of GEN was evaluated using various actinomyces/antibiotic production media, and the results showed that two media showed maximum specific productivity, namely CM4 which is also named M65 (DSMZ, Braunschweig, Germany), followed by CM6 (also named as APM; [29], respectively. Therefore, the CM4 culture medium was selected for the next experiments. Our results are in accordance with previous studies where the respective AGA production media gave maximum specific productivity of paromomycin [20], istmaycin [15], and fortimicin [30]. As previously reported, environmental culture conditions, for instance, culture media compositions, incubation temperature, agitation rate, initial pH, and incubation time, play an important role in the production of antibiotics by their natural producers [31, 32]. Therefore, in this study, four environmental factors, including the initial pH of the selected production medium (CM4), incubation temperature, agitation rate, and incubation time, have been evaluated and statistically optimized for maximum GEN production. Accordingly, RSM using DOD was selected for accomplishing this purpose where the tested factors were as follows: initial pH (factor A), incubation temperature (factor B), agitation rate (factor C), and incubation time (factor D) each factor was tested at two levels (-1, and + 1). As previously stated, the response surface DOD was used for its appropriateness since four interacting components, each at two levels, must be assessed (mixed models with multi-response) [11, 25].

The D-optimal design was chosen for our model’s construction because it is characterized by having the best subsets of all possible experiments, is a practical model to work with harsh or extreme conditions, and is highly sensitive and predictive with the least percentage bias [24, 33]. The main goal of D-optimal design is to influence optimal system fulfillment by simultaneously improving a large number of elements [25]. It investigates how changes in factor levels affect a selected response and, in turn, how responses can be anticipated [25]. Our study’s DOD produced 23 experiments that, according to ANOVA, produced a significant model (P value < 0.001). With just a 0.01% probability that a “Model F-Value” may occur due to noise, the Model F-value of 11.21 indicates that the model is significant [34]. Furthermore, ANOVA analysis revealed that three of the four factors—the initial pH, agitation rate, and incubation time—had significant effects (p-value < 0.05) on the generation of GEN within each tested range, which was in line with other findings [35, 36]. However, within the tested range (25–37 ˚C) of the incubation temperature (factor B) does not significantly affect the formation of GEN (p-value = 0.633). We regard these results as new evidence that M. purpureochromogenes NRRL B-16,094 produces GEN. The Lack of Fit “F-value” for the DOD model was 2.54, which indicates non-significant behavior in relation to the pure error (p-value = 0.203). In essence, a non-significant lack of fit is considered a model’s confidence marker. The lab results showed a remarkable charge admiration with a low coefficient of variation (CV) of 12.65%. The CV is a valuable guarantee to ascertain the importance of accuracy in assessing the readings. The “Adj R-Squared” of 0.735 and the “Pred R-Squared” of 0.640 are reasonably coordinated. For the signal-to-noise ratio, “Adeq Precision” must be four or above. It’s interesting to note that our model’s ratio of 10.972 indicated an acceptable signal, and it can be used to explore the design space. The optimal values of 7 for the initial pH, 30 ˚C for the incubation temperature, 300 rpm for the agitation, and 10 days for the incubation time were determined by the contour and three-dimensional graphs of the factors of our model that were predicted. Our lab confirmed the expected values collectively, and the results showed a 2.4-fold increase over the optimized production medium (CM4) and a 13.5-fold rise over the unoptimized culture condition as verified by HPLC analysis using basic AGA production CM1 incubated at 28 ˚C, with an agitation of 200 rpm for 6 days.

Conclusion

The biosynthetic pathway of GEN has been elucidated based on an analysis of the genes/proteins that the gentamicin (gen-) gene cluster encodes. By comparing the gen- biosynthetic gene cluster to those of kanamycin and fortimicin, results revealed that gentamicin biosynthesis could have originated from a combination of fortimicin and kanamycin biosynthetic pathways, which is evidenced by the presence of many structural similarities in the modified sugars and aglycone moieties. Six actinomyces/aminoglycoside production media were evaluated for maximum production of gentamicin and compared to the basic gentamicin production medium (TSB). Results showed that CM4 gave the highest specific productivity. The initial pH of the CM4 (pH ranged from 5 to 9), the incubation temperature (25–37 ˚C) and agitation rate (100–300 rpm), incubation time (3–10 days) were statistically optimized using a D-optimal design resulting in a 2.4-rise increase as compared to the optimized culture media and a 13.5-fold increase as compared to the unoptimized culture conditions. The resulting optimized conditions are highly recommended for scaling up the production of GEN produced by M. purpureochromogenes NRRL B-16,094, taking into consideration factors influencing the scale-up process such as mass transfer limitation (inoculum size), metabolic flow redistribution, constant power input per unit volume, mixing time and agitation rate in terms of stirrer speed and aeration rate as previously reported [37,38,39].

Data availability

All data generated or analyzed during this study are included in this published article and supplementary file. The gentamicin, kanamycin and fortimicin biosynthetic gene clusters were deposited in the NCBI GenBank database under the accession code, AJ628149 (https://www.ncbi.nlm.nih.gov/nuccore/AJ628149), AJ628422 (https://www.ncbi.nlm.nih.gov/nuccore/AJ628422), and AJ628421 (https://www.ncbi.nlm.nih.gov/nuccore/AJ628421), respectively.

References

  1. Piepersberg W, Aboshanab KM, Schmidt-Beißner H, Wehmeier UF. The biochemistry and genetics of aminoglycoside producers. Aminoglycoside antibiotics. New York, USA: Wiley; 2007. pp. 15–118.

    Chapter  Google Scholar 

  2. Krause KM, Serio AW, Kane TR, Connolly LE. Aminoglycosides: an overview. Cold Spring Harb Perspect Med. 2016;6:a027029.

    Article  PubMed  PubMed Central  Google Scholar 

  3. Aboshanab KM, Alshahrani MY, Alafeefy A. Combinatorial biosynthesis of novel aminoglycoside antibiotics via pathway engineering. AMB Express. 2024;14:103.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  4. Wachino J-I, Doi Y, Arakawa Y. Aminoglycoside resistance. Infect Dis Clin North Am. 2020;34:887–902.

    Article  PubMed  PubMed Central  Google Scholar 

  5. Wehmeier UF, Piepersberg W. Enzymology of aminoglycoside biosynthesis-deduction from gene clusters. Methods Enzymol. 2009;459:459–91.

    Article  CAS  PubMed  Google Scholar 

  6. Chaves BJ, Tadi P, Gentamicin. London, UK: StatPearls; 2024.

  7. Takemoto JY, Altenberg GA, Poudyal N, Subedi YP, Chang C-WT. Amphiphilic aminoglycosides: modifications that revive old natural product antibiotics. Front Microbiol. 2022;13.

  8. Hasegawa M. A novel, highly efficient gene-cloning system in Micromonospora applied to the genetic analysis of Fortimicin biosynthesis. Gene. 1992;115:85–91.

    Article  CAS  PubMed  Google Scholar 

  9. Odakura Y, Kase H, Nakayama K. Sagamicin and the related aminoglycosides: fermentation and biosynthesis III. Isolation and characterization of Micromonospora sagamiensis mutants blocked in gentamicin C1 pathway. J Antibiot (Tokyo). 1983;36:125–30.

    Article  CAS  PubMed  Google Scholar 

  10. Flatt PM, Mahmud T. Biosynthesis of aminocyclitol-aminoglycoside antibiotics and related compounds. Nat Prod Rep. 2007;24:358–92.

    Article  CAS  PubMed  Google Scholar 

  11. Ibrahim HM, Elkhidir EE. Response surface method as an efficient tool for medium optimisation. Trends Appl Sci Res. 2011;6:121–9.

    Article  Google Scholar 

  12. El-Sayed SE, Abdelaziz NA, El-Housseiny GS, Aboshanab KM. Octadecyl 3-(3, 5-di-tert-butyl-4-hydroxyphenyl) propanoate, an antifungal metabolite of Alcaligenes faecalis strain MT332429 optimized through response surface methodology. Appl Microbiol Biotechnol. 2020;104.

  13. Vehapi M, İnan B, Kayacan-Cakmakoglu S, Sagdic O, Özçimen D. Production of Bacillus subtilis soil isolate as biocontrol agent under bioreactor conditions. Arch Microbiol. 2023;205:52.

    Article  CAS  PubMed  Google Scholar 

  14. Zhao J, Tian G, Qu H. Application of I-Optimal design for modeling and optimizing the operational parameters of ibuprofen granules in continuous Twin-Screw wet granulation. Biomedicines. 2023;11:2030.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  15. Gomaa FAM, Selim HMRM, Alshahrani MY, Aboshanab KM. Central composite design for optimizing Istamycin production by Streptomyces tenjimariensis. World J Microbiol Biotechnol. 2024;40:316.

    Article  CAS  PubMed  Google Scholar 

  16. Miller JH. Experiments in molecular genetics. USA: Cold Spring Harbor Laboratory; 1972.

    Google Scholar 

  17. Bibb MJ, Findlay PR, Johnson MW. The relationship between base composition and codon usage in bacterial genes and its use for the simple and reliable identification of protein-coding sequences. Gene. 1984;30:157–66.

    Article  CAS  PubMed  Google Scholar 

  18. Altschul SF, Gish W, Miller W, Myers EW, Lipman DJ. Basic local alignment search tool. J Mol Biol. 1990;215:403–10.

    Article  CAS  PubMed  Google Scholar 

  19. Kieser T, Bibb M, Buttner M, Chater K, Hopwood D. Practical Streptomyces genetics. Norwich, UK: John Innes Foundation; 2000.

    Google Scholar 

  20. Ibrahim AA, El-Housseiny GS, Aboshanab KM, Yassien MA, Hassouna NA. Paromomycin production from Streptomyces rimosus NRRL 2455: statistical optimization and new synergistic antibiotic combinations against multidrug resistant pathogens. BMC Microbiol. 2019;19:18.

    Article  PubMed  PubMed Central  Google Scholar 

  21. Balouiri M, Sadiki M, Ibnsouda SK. Methods for in vitro evaluating antimicrobial activity: A review. J Pharm Anal. 2016;6:71–9.

    Article  PubMed  Google Scholar 

  22. Chalenko VG, Volodina IK, Rumiantsev SN. Colorimetric dry weight of bacterial cultures. I. Details of the method]. Zh Mikrobiol Epidemiol Immunobiol. 1970;47:75–81.

    CAS  PubMed  Google Scholar 

  23. Elrod L, White LB, Wong CF. Determination of Fortimicin a sulfate by high-performance liquid chromatography after derivatization with 2,4-dinitrofluorobenzene. J Chromatogr A. 1981;208:357–63.

    Article  CAS  Google Scholar 

  24. Schenkendorf R, Kremling A, Mangold M. Optimal experimental design with the Sigma point method. IET Syst Biol. 2009;3:10–23.

    Article  CAS  PubMed  Google Scholar 

  25. Gong Z, Zhu X, Zhang C. D-optimal design of the additive mixture model with multi-response. Math Biosci Eng. 2022;19:4737–48.

    Article  PubMed  Google Scholar 

  26. Thompson HW, Mera R, Prasad C. The analysis of variance (ANOVA). Nutr Neurosci. 1999;2:43–55.

    Article  CAS  PubMed  Google Scholar 

  27. Bentley SD, Chater KF, Cerdeño-Tárraga A-M, Challis GL, Thomson NR, James KD, et al. Complete genome sequence of the model actinomycete Streptomyces coelicolor A3(2). Nature. 2002;417:141–7.

    Article  PubMed  Google Scholar 

  28. Ikeda H, Ishikawa J, Hanamoto A, Shinose M, Kikuchi H, Shiba T, et al. Complete genome sequence and comparative analysis of the industrial microorganism Streptomyces avermitilis. Nat Biotechnol. 2003;21:526–31.

    Article  PubMed  Google Scholar 

  29. Nam DH, Ryu DD. Relationship between Butirosin biosynthesis and sporulation in Bacillus circulans. Antimicrob Agents Chemother. 1985;27:798–801.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  30. Selim HMRM, Gomaa FAM, Alshahrani MY, Aboshanab KM. Response surface D-Optimal design for optimizing fortimicins production by Micromonospora olivasterospora and new synergistic Fortimicin-A-Antibiotic combinations. Curr Microbiol. 2025;82:68.

    Article  CAS  PubMed  Google Scholar 

  31. Yang E, Fan L, Yan J, Jiang Y, Doucette C, Fillmore S, et al. Influence of culture media, pH and temperature on growth and bacteriocin production of bacteriocinogenic lactic acid bacteria. AMB Express. 2018;8:10.

    Article  PubMed  PubMed Central  Google Scholar 

  32. Thakur D, Bora TC, Bordoloi GN, Mazumdar S. Influence of nutrition and culturing conditions for optimum growth and antimicrobial metabolite production by Streptomyces sp. 201. J Mycol Med. 2009;19:161–7.

    Article  Google Scholar 

  33. Jones B, Allen-Moyer K, Goos P. A-optimal versus D-optimal design of screening experiments. J Qual Technol. 2021;53:369–82.

    Article  Google Scholar 

  34. Liang T, Guan R, Shen H, Xia Q, Liu M. Optimization of conditions for Cyanidin-3-OGlucoside (C3G) nanoliposome production by response surface methodology and cellular uptake studies in Caco-2 cells. Molecules. 2017;22:457.

    Article  PubMed  PubMed Central  Google Scholar 

  35. Singh LS, Sharma H, Talukdar NC. Production of potent antimicrobial agent by actinomycete, Streptomyces sannanensis strain SU118 isolated from Phoomdi in Loktak lake of Manipur, India. BMC Microbiol. 2014;14:278.

    Article  PubMed  PubMed Central  Google Scholar 

  36. Khoshayand F, Goodarzi S, Shahverdi AR, Khoshayand MR. Optimization of culture conditions for fermentation of soymilk using Lactobacillus casei by response surface methodology. Probiotics Antimicrob Proteins. 2011;3:159–67.

    Article  CAS  PubMed  Google Scholar 

  37. Melchor-Moncada JJ, García-Barco A, Zuluaga-Vélez A, Veloza LA, Sepúlveda-Arias JC. Scale-Up of the fermentation process for the production and purification of serratiopeptidase using silkworm pupae as a substrate. Methods Protoc. 2024;7:19.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  38. Xia J, Wang G, Fan M, Chen M, Wang Z, Zhuang Y. Understanding the scale-up of fermentation processes from the viewpoint of the flow field in bioreactors and the physiological response of strains. Chin J Chem Eng. 2021;30:178–84.

    Article  CAS  Google Scholar 

  39. Hsu Y-L, Wu W-T. A novel approach for scaling-up a fermentation system. Biochem Eng J. 2002;11:123–30.

    Article  CAS  Google Scholar 

Download references

Acknowledgements

The authors would like to acknowledge the Open access funding provided by The Science, Technology & Innovation Funding Authority (STDF) in cooperation with The Egyptian Knowledge Bank (EKB) and Springer Nature transformative agreement. The authors also acknowledge Microbiology and Immunology Department, Faculty of Pharmacy, Ain Shams University (ASU), Egypt for providing the laboratory facilities for this study. The authors extend their appreciation to the Deanship of Scientific Research at King of Khalid University for funding this work through the Large Groups Project under grant number RGP1-84-44.

Funding

Open access funding provided by The Science, Technology & Innovation Funding Authority (STDF) in cooperation with The Egyptian Knowledge Bank (EKB).

The authors extend their appreciation to the Deanship of Scientific Research at King of Khalid University for funding this work through the Large Groups Project under grant number RGP1-84-44.

Author information

Authors and Affiliations

  1. Department of Clinical Laboratory Sciences, College of Applied Medical Sciences, King Khalid University, Abha 9088, P.O. Box 61413, Saudi Arabia

    Muath Suliman & Mohammad Y. Alshahrani

  2. Department of Microbiology and Immunology, Faculty of Pharmacy , Ain Shams University, Cairo, 11566, Egypt

    Amr S. Bishr & Khaled M. Aboshanab

  3. Department of Microbiology and Immunology, Faculty of Pharmacy (girls), Al-Azhar University, 11651, Cairo, Egypt

    Sally T. K. Tohamy

  4. Central Labs, King Khalid University, AlQura’a,P.O. Box 960, Abha, 9088, Saudi Arabia

    Mohammad Y. Alshahrani

Authors
  1. Muath Suliman
    View author publications

    You can also search for this author inPubMed Google Scholar

  2. Amr S. Bishr
    View author publications

    You can also search for this author inPubMed Google Scholar

  3. Sally T. K. Tohamy
    View author publications

    You can also search for this author inPubMed Google Scholar

  4. Mohammad Y. Alshahrani
    View author publications

    You can also search for this author inPubMed Google Scholar

  5. Khaled M. Aboshanab
    View author publications

    You can also search for this author inPubMed Google Scholar

Contributions

MS, ASB, STT, MYA and KMA, conceptualization. MS, ASB, MYA and KMA, conceived and designed the study; MS, ASB, MYA and KMA made data curation; MS, STT, and ASM were involved in the Methodology, MS, STT, MYA, and ASB; wrote the original draft preparation; KMA; revised, and the manuscript. All authors reviewed the manuscript”.

Corresponding author

Correspondence to Khaled M. Aboshanab.

Ethics declarations

Ethical approval

Not applicable.

Consent to participate

Not applicable.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.

Additional information

Publisher’s note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Electronic supplementary material

Below is the link to the electronic supplementary material.

Supplementary Material 1

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Suliman, M., Bishr, A.S., Tohamy, S.T.K. et al. D-optimal design model and biosynthetic pathway for gentamicin production by Micromonospora purpureochromogenes NRRL B-16094. BMC Microbiol 25, 309 (2025). https://doiorg.publicaciones.saludcastillayleon.es/10.1186/s12866-025-04001-8

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: https://doiorg.publicaciones.saludcastillayleon.es/10.1186/s12866-025-04001-8

Keywords

BMC Microbiology

ISSN: 1471-2180

Contact us