This is an integrative literature review, conducted using the MEDLINE portal and the PubMed database. The PICo strategy (acronym for Population/Problem, Interest, and Context) was used to formulate the research question, enabling the identification of descriptors that aid in locating relevant primary studies in the databases [34]. Table 1 illustrates the formulation of the research question using this strategy.

Table 1. Formulating the research question—PICo strategy.

Thus, the delimited research question is: “What are the main strategies for obtaining, cultivating, and processing bacteria that produce antibiotic substances against Staphylococcus aureusandPseudomonas aeruginosa described in the literature, considering their sources of isolation, taxonomic diversity, growth conditions, and methods of extraction and identification of bioactive molecules?” The keywords were selected from the Health Sciences Descriptors of the BVS and MeSH Database, using Boolean operators “AND” and “OR” to combine the terms: Anti-Bacterial Agents; Peptides; Bacteriocins; Lipopeptides; Enzymes; Biofilms; Bioprospecting; Metabolome; Microbiota; Supernatant; Wound healing; Wound infection; Cell-free supernatant; Cell-free extract; Chromatography, High Pressure Liquid; Liquid Chromatography-Mass Spectrometry; Nuclear Magnetic Resonance, Biomolecular; Proton Magnetic Resonance Spectroscopy; Magnetic Resonance Spectroscopy; Staphylococcus aureus; Pseudomonas aeruginosa.

The sampling method used was sequential, which consists of recruiting all accessible productions that meet the eligibility criteria over a specific time interval [35], which was conveniently limited to the last five years.

The inclusion criteria used are: in vitro studies, adherence to the objective and the proposed theme, articles published in full, in English, with a publication time frame from 2020 to 2025.

Exclusion criteria: Duplicate articles, literature reviews, texts unavailable in full format, isolated case reports, opinion articles, and those that did not address the defined topic.

Data was collected from the PubMed database, maintained by the National Institutes of Health (NIH). This database was selected because of its broad coverage of biomedical and microbiological literature relevant to the scope of this review. The respective descriptors and Boolean operators were used: [(((Anti-Bacterial Agents) OR (Peptides) OR (Bacteriocins) OR (Lipopeptides) OR (Enzymes) OR (Biofilms)) AND ((Bioprospecting) OR (Metabolome) OR (Microbiota) OR (Supernatant) OR (Wound Healing) OR (Wound Infection) OR (cell-free supernatant) OR (cell-free extract) OR (Chromatography, High Pressure Liquid) OR (Liquid Chromatography-Mass Spectrometry) OR (Nuclear Magnetic Resonance, Biomolecular) OR (Proton Magnetic Resonance Spectroscopy) OR (Magnetic Resonance Spectroscopy)) AND ((Staphylococcus aureus) OR (Pseudomonas aeruginosa)) NOT (plants) NOT (herbal) NOT (Oils, Volatile) NOT (oils) NOT (phytochemicals) NOT (intestine) NOT (in vivo) NOT (mice) NOT (Capsules) NOT (gut) NOT (dental plaque) NOT (probiotics) NOT (seaweed) NOT (fungi) NOT (virus) NOT (silver) NOT (zinc) NOT (gold) NOT (metals) NOT (Hydrogels) NOT (Review [Publication Type]))]. Restrictive NOT filters related to probiotics and gut-associated terms were applied to narrow the scope of the review and avoid the overrepresentation of studies involving probiotic microorganisms and intestinal microbiota, which are already extensively described in the literature. The objective of these exclusions was to focus the search on less explored bacterial sources with potential for antibacterial bioprospecting.

After searching the aforementioned database, the articles were refined beforehand by reading the abstract and selected after a complete reading of the text, to ensure they met the inclusion criteria. A flowchart of the search was created, containing the databases and the number of publications located. Search filters were also applied, based on the inclusion and exclusion criteria, to select potential bibliography for this study using the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA), as shown in the flowchart (Figure 1).

Figure 1. Article selection flowchart.

After applying the descriptors and Boolean operators in PubMed/MEDLINE, a total of 3882 articles were identified. Subsequently, filters for articles published in English and within the last five years were applied, resulting in 3689 and 847 articles, respectively. No duplicate articles were identified due to the use of a single database. The titles and abstracts of the 847 articles were screened, and 44 studies were considered eligible and included in the review after full-text assessment. The main reasons for exclusion during the screening process were studies conducted exclusively or predominantly in vivo (n = 79), articles not adherent to the proposed objective and theme of the review (n = 521), literature reviews (n = 194), case reports (n = 7), and opinion articles (n = 2). Study selection and data extraction were performed by a single reviewer according to the established eligibility criteria. Due to the integrative nature of this review and the methodological heterogeneity among the included in vitro studies, no formal methodological quality appraisal tool was applied. The studies were evaluated based on their relevance to the proposed objective and the established eligibility criteria.

The data extracted from the selected studies were organized into a spreadsheet, arranged in manually assigned sequential numbers, and which includes information such as number, authors, year, title, genera or species of antibiotic-producing bacteria and their culture parameters, source of isolation of the producing bacteria, target species of the study, molecule extraction method used, identification method used, and nature or identification of the bioactive molecule.

The methods used to identify the biomolecules present in the supernatants were classified as follows:

Separation techniques only: Thin Layer Chromatography (TLC), Cation Exchange Chromatography (CIEX), C18 chromatographic column, High-performance liquid chromatography (HPLC), Ion Exchange Chromatography (IEX), Reversed Phase High-performance liquid chromatography (RP-HPLC), Reversed-phase Chromatography (RPC) and Size Exclusion Chromatography (SEC);Spectroscopic techniques: Fourier Transform Infrared Spectroscopy (FTIR) and Nuclear Magnetic Resonance (NMR);Spectrometric techniques: Mass Spectrometry (MS), Electrospray Ionization Mass Spectrometry (ESI-MS) and Electrospray Ionization coupled to Mass Spectrometry (ESI-MS/MS);Electrophoretic techniques: Polyacrylamide Gel Electrophoresis (PAGE) and Sodium Dodecyl Sulfate Polyacrylamide Gel Electrophoresis (SDS-PAGE);Hybrid techniques: Gas Chromatography coupled to Mass Spectrometry (GC-MS), High Performance Liquid Chromatography coupled to Mass Spectrometry (HPLC-MS), High Performance Liquid Chromatography coupled to UV-Visible Detector (HPLC-UV), Liquid Chromatography coupled to High-Resolution Mass Spectrometry (LC-HRMS), Liquid chromatography-mass spectrometry (LC-MS), Liquid chromatography-mass spectrometry coupled with mass spectrometry (LC-MS/MS), High-Resolution Mass Spectrometry coupled with ultraviolet detector and Liquid Chromatography (LC-UV-HRMS), Ultra-High Performance Liquid Chromatography coupled to a Diode Array Detector (UHPLC-DAD), Ultra-High Performance Liquid Chromatography coupled to Mass Spectrometry (UHPLC-MS), Ultra-High Performance Liquid Chromatography coupled to a Diode Detector and Mass Spectrometer (UPLC-DAD-MS), Ultra-High Performance Liquid Chromatography with High-Resolution Mass Spectrometry and Tandem Mass Spectrometry (UPLC-HR-MS/MS) and Ultra Performance Liquid Chromatography coupled to UV-Visible Detector (UPLC-UV).

The publications were analyzed separately, and the studies selected for this review were categorized using codes from 1 to 44. The results are presented in tables, primarily aiming to demonstrate the aspects identified as most relevant, according to the objectives of this study.

After reviewing the selected bibliographic material, information was extracted that answered the guiding research question: “What are the main strategies for obtaining, cultivating, and processing bacteria that produce antibiotic substances against Staphylococcus aureusandPseudomonas aeruginosa described in the literature, considering their sources of isolation, taxonomic diversity, growth conditions, and methodologies for extracting and identifying bioactive molecules?” The overview of the articles found, in relation to the species that produce antibiotic substances and their sources of isolation, and target species, is presented in Table S1.

3.1.1. Genera of Antibiotic-Producing Bacteria and Target Pathogens

Among the target species evaluated in the included studies (Table S2), Staphylococcus aureus was the most frequently investigated, appearing in 32 out of 44 studies (72.7%). Some studies simultaneously analyzed Staphylococcus aureusandPseudomonas aeruginosa, with 9 out of 44 studies (20.5%), while only a small number of articles focused exclusively on Pseudomonas aeruginosa, representing 3 out of 44 studies (6.8%). This overview shows that, although both species are relevant pathogens, there is a predominance of studies focused on the potential use of bacterial supernatants in prospecting for molecules with activity against S. aureus, whileP. aeruginosa remains relatively unexplored, representing a gap and, at the same time, a promising opportunity for future investigations.

Regarding the genera of bacteria that produce bioactive substances, there is a predominance of the genus Bacillus, one of the bacterial genera most associated with its ability to produce enzymes and proteins [36], with 13 species belonging to this genus being used in the studies (Table S3). Furthermore, there is a predominance of lactic acid bacteria genera, which includes genera such as Brevibacillus, Enterococcus,Lacticaseibacillus,Lactiplantibacillus, Lactobacillus, Lactococcus,Ligilactobacillus,Loigolactobacillus,Pediococcus,and Vagococcus.

Among the species of the genus Bacillus, the one that stood out the most was Bacillus subtilis, present in 3 studies. B. subtilis is a Gram-positive bacterium that forms heat-resistant spores as part of its life cycle and is reported as a fast-growing bacterium. Its genetic material can also be manipulated due to its ability to acquire external DNA and integrate it into its genome, making this species one of the most studied bacteria for this type of function, along with Escherichia coli [37].

Because it is a non-pathogenic, versatile, and easy-to-cultivate bacterium, B. subtilis can be used for various purposes, such as the production of traditional foods in Asia and the production of vitamins, amino acids, and enzymes [37]. Due to the many applications of B. subtilis being associated with its ability to produce and secrete proteins, research has focused on its secretome, which includes both its protein secretion machinery and the proteins themselves [36].

These findings demonstrate the importance of bioprospecting as a methodology for obtaining bioactive molecules, given the variety of species found that can be used for this purpose. Li and collaborators [38] used species from the genera Lactiplantibacillus, Lactobacillus, and Ligilactobacillus, belonging to the lactic acid bacteria group, from different foods, obtaining bioactive molecules against P.aeruginosa. On the other hand, studies have used species that are not part of the lactic acid bacteria group, such as Staphylococcuspseudointermedius, isolated from the raccoon dog, which has shown the ability to produce molecules against S. aureus.

3.1.2. Sources of Isolation

Of all the reported sources of bacterial isolation, the marine environment was the most frequently cited, appearing in 13 studies (Table S4). This environment, especially bacteria isolated from marine sponges, has been shown in recent research to be a rich source of substances with diverse biological actions, particularly in relation to antibiotic or antibiofilm molecules [39], but also for applications in the energy, food, nutraceutical and biorefinery industries [3][40][41].

Food and soil are also explored sources, with 8 studies each obtaining species from these locations. Food production, for example, is closely linked to the relationship between microorganisms and plants, given that microorganisms present in the soil are able to participate in nutrient cycling and waste processing through the enzymes they produce, and such enzymes can be the target of bioprospecting processes [42]. Garcia-López and collaborators [43] used strains of Lactiplantibacillusparaplantarum and Pediococcusacidilactici isolated from sausages, finding activity against Listeria monocytogenes, Escherichia coli, Clostridium perfringens, andStaphylococcus aureus.

Among the sources investigated in the analyzed studies, the human microbiota remains comparatively underexplored as a source for antimicrobial-producing microorganisms, despite its recognized microbial diversity and biotechnological potential. Notably, only one study included in this review reported the use of intact human skin as a source for bioprospecting. The skin microbiome represents a highly diverse and dynamic ecological niche, shaped by continuous interactions between microorganisms and the host across one of the largest and most exposed surfaces of the human body. This unique ecological context suggests substantial, yet still largely untapped, potential for the discovery of new antibiotic substances. Therefore, expanding bioprospecting efforts toward the human microbiota may represent a promising direction for future research aimed at identifying new antimicrobial-producing microorganisms [44].

The resident skin microbiota is capable of producing antimicrobial peptides that, in the normal functioning of the human body, have the function and capacity to combat the establishment of external species at that site. Among these antimicrobial peptides, bacteriocins demonstrate this ability to inhibit pathogenic microorganisms in in vitroandin vivo studies [45]. Due to the need to discover new compounds with antibacterial activity to control the spread of antibiotic-resistant microorganisms, interest in the protective role played by the skin microbiota has increased. In the study by Pedretti and collaborators [44], a strain of Bacillussiamensis was isolated from intact skin and showed the ability to produce molecules with antimicrobial activity against several pathogenic species, such as Staphylococcus aureus, Enterococcus faecalis, Streptococcus agalactiae,andCandida spp.

Table S5 summarizes the approaches used across studies for the discovery and characterization of antibiotic molecules, including extraction strategies, cultivation conditions, and the general types of analytical methods applied for compound identification. Collectively, this synthesis highlights the methodological diversity employed to obtain and characterize antibiotic substances with activity against antibiotic-resistant bacteria.

3.2.1. Culture Media

The main culture media used were Man, Rogosa and Sharpe (MRS) broth, used in 11 studies, and Luria-Bertani (LB) broth, used in 9 studies (Table S6). MRS broth, considered a rich culture medium, is the standard choice for the isolation of lactic acid bacteria. It uses glucose as a carbon source and various nitrogen sources in its composition, including yeast extract, meat extract, and peptone, and is supplemented with minerals that provide micronutrients [46][47]. LB broth is a complex culture medium containing tryptone, yeast extract, and NaCl [48]. It is noted as one of the main culture media for bacteria, being a good source of energy for the growth of Lactobacillus and other lactic acid bacteria [49].

3.2.2. Incubation Temperature

The most frequent incubation temperatures were 37˚C and 30˚C, with 15 and 12 studies, respectively (Table S7). Each microorganism is adapted to specific temperature conditions for its cultivation, as evidenced by the work of Lei and collaborators [50], in which the species Lactobacillus plantarum isolated from human feces was cultivated at 37˚C, the temperature corresponding to its habitat at the time of collection and isolation. In contrast, the study by Shirazi and collaborators [51], which isolated the species Laceyellasacchari, Thermoactinomyces sp., and Laceyella sp. from the surroundings of volcanoes and other locations, cultivated these species at 50˚C, given that they are thermophilic microorganisms.

3.2.3. Incubation Times

The most frequently used incubation times were 24 and 48 hours, with 14 and 9 studies each, respectively (Table S8). In the work of Levenfors and collaborators [52], the incubation time varied between 4 and 7 days, in an attempt to evaluate the conditions under which the production of bioactive substances was optimized. This practice, along with variations in other cultivation parameters, can help determine the best yields for the production of substances of interest.

3.2.4. Shaking during Bacterial Growth

Regarding the use of shaking during the incubation of the culture medium containing the inoculum, 26 studies used this technique, while 15 did not (Table S9). During their growth, bacterial cultures in broth that are under agitation tend to have a higher growth rate, which can be attributed to the aeration of the medium due to the agitation process [53]. Therefore, this practice should be considered for experiments involving bioprospecting, when applicable, but taking into account the most suitable conditions for obtaining the bioactive substance.

3.2.5. Oxygen Availability

The most frequently observed oxygen availability in the studies was aerobic, with 34 studies using this incubation condition and 7 studies using anaerobic conditions (Table S10). The species cultivated in the absence of oxygen were Bacillus spp., Bacillus subtilis, Lacticaseibacillusrhamnosus,andLactobacillus plantarum. These species are facultative anaerobes and can be cultivated under such conditions, especially if the biotechnological direction of their production of bioactive substances is promoted.

3.2.6. Extraction Methods

Among the methodologies used for the extraction of antibiotic molecules from bacterial culture supernatants, the organic solvent extraction method is the most widely used, being cited in 19 studies (Table S11).

The selection of molecules by different polarities allows the separation of the various components originating from bacterial metabolism [54], enabling phenotypic tests with smaller sets of molecules or even the isolated molecule. This approach is also important for the subsequent identification of molecules, since it makes it possible to know with greater certainty the bioactive molecule among several molecules produced by that species [55].

Another methodology used is ammonium sulfate precipitation, aiming at the separation of proteins present in the cell-free supernatant (CFS). This method was used in 13 studies and is based on the salting-out process, associated with the decrease in protein solubility when the salt concentration in the solution increases, allowing an increase in extraction yield and optimizing the protein purification process [56].

3.2.7. Identification Methods

After extracting the molecules of interest from their original CFS, another important step is the structural identification of that compound. The main methods found were Nuclear Magnetic Resonance (NMR), used in 10 studies, in addition to several methodologies based on chromatographic methods followed by mass spectrometry (Table S12).

In the studies included in this review, nuclear magnetic resonance (NMR) spectroscopy was primarily used for the structural elucidation of purified antibiotic substances obtained after extraction and preliminary fractionation steps. By providing one-dimensional spectra based on hydrogen spin analysis, NMR enabled detailed structural characterization, supporting compound identification with high reliability. In addition to its analytical robustness, NMR was frequently valued for being non-destructive, allowing sample recovery after analysis [57].

Mass spectrometry (MS) was widely applied for the detection and preliminary identification of bioactive molecules in complex biological extracts. In most studies, MS was coupled with separation techniques such as high-performance liquid chromatography (HPLC) or liquid chromatography (LC) to reduce sample complexity prior to ionization, commonly through electrospray ionization (ESI). This combined approach improved the detection of antimicrobial metabolites produced by the studied microorganisms [58]-[62].

Chromatographic techniques, particularly LC and GC, were consistently used as essential preparatory and separation steps before MS or, in some cases, NMR analysis. These methods enabled the fractionation of crude extracts and facilitated the isolation of antibiotic substances. In quantitative applications, LC-MS workflows frequently relied on targeted analyses using standards, although a recurring limitation across studies was the scarcity of appropriate reference compounds for absolute quantification. To address this, some studies employed isotopically labelled standards or metabolic labelling strategies to improve quantification accuracy in complex mixtures [63][64].