Guava is consumed in natura and is also of considerable importance to the food industry. The seeds and peel of this fruit are discarded, however, guava seeds yield oil (~13%) that can be used for the bioproducts synthesis. The use of a by-product as a carbon source is advantageous, as it reduces the environmental impact of possible harmful materials to nature, while adding value to products. In addition, the use of untested substrates can bring new yield and characterization results. Thus, this research sought to study rhamnolipids (RLs) production from guava seed oil, a by-product of the fructorefinery. The experiments were carried out using Pseudomonas aeruginosa LBI 2A1 and experimental design was used to optimize the variables Carbon and Nitrogen concentration. Characterization of RLs produced occurred by LC-MS. In this study, variables in the quadratic forms and the interaction between them influenced the response (p < 0.05). The most significant variable was N concentration. Maximum RLs yield achieved 39.97 g/L, predominantly of mono-RL. Characterization analysis revealed 9 homologues including the presence of RhaC₁₀C_14:2 (m/z 555) whose structure has not previously been observed. This research showed that guava seed oil is an alternative potential carbon source for rhamnolipid production with rare rhamnolipid homologues.

Rhamnolipids (RLs) are versatile biosurfactants produced by Pseudomonas aeruginosa with significant industrial potential. However, high production costs remain a barrier to large-scale use, necessitating genetic strategies to improve yields. While many genes are reported to influence RL production, studies often rely on qualitative phenotypic assays of questionable reliability. We systematically evaluated 29 P. aeruginosa PA14 mutants using traditional assays (Siegmund–Wagner blue plates, swarming motility) and validated the findings using liquid chromatography–mass spectrometry (LC/MS). We found that traditional phenotypic assays have a high misprediction rate (~35% to 38%), primarily due to confounding factors like variable flagellar function, colony spreading, and growth rates. Specifically, LC/MS quantification revealed that rpoN and pvdQ knockouts significantly increased total RL titers, whereas crc , dksA , and dspI knockouts decreased production. Notably, the increased titers in rpoN and pvdQ mutants were linked to enhanced biomass accumulation rather than higher per-cell biosynthetic rates. These findings highlight the critical necessity of using quantitative analytical methods for accurate strain screening and provide a clarified set of genetic targets for metabolic engineering aimed at optimizing RL production. IMPORTANCE: This study addresses a critical methodological flaw in biosurfactant research: the overreliance on qualitative phenotypic assays that too often lead to inaccurate conclusions. By systematically comparing traditional screening methods with liquid chromatography–mass spectrometry quantification across a collection of Pseudomonas aeruginosa mutants, we demonstrate that common assays like swarming motility and blue plates fail to accurately predict rhamnolipid (RL) production in over one-third of cases. These inaccuracies lead to the misidentification of genetic targets and may waste resources in metabolic engineering efforts. Our work provides a reliable framework for strain screening, identifies specific genes that influence RL yields, and clarifies the biological factors—such as flagellar motility and growth dynamics—that bias traditional results. These findings are essential to optimize biosurfactant production and ensure data reproducibility.

The aim of this study was to isolate Weizmannia spp. that produce lactic acid from xylose and use an experimental design to optimize the production of the metabolite. After isolation, the experiments were conducted in xylose-yeast extract-peptone medium. The identification of isolates was performed using the 16S rDNA PCR technique, followed by sequencing. A central composite rotatable design (CCRD) was used to optimize the concentrations of the carbon source (xylose), nitrogen source (yeast extract and peptone), and sodium acetate. Two strains were considered promising for lactic acid production, with W. coagulans BLMI achieving greater lactic acid production under anaerobic conditions (21.93 ± 0.9 g·L⁻¹) and a yield of 69.18%, while the strain W. ginsengihumi BMI was able to produce 19.79 ± 0.8 g·L⁻¹, with a yield of 70.46%. CCRD was used with the W. ginsengihumi strain due to the lack of records in the literature on its use for lactic acid production. The carbon and nitrogen sources influenced the response, but the interactions of the variables were nonsignificant (p < 0.05). The response surface analysis indicated that the optimal concentrations of carbon and nitrogen sources were 32.5 and 3.0 g·L⁻¹, respectively, without the need to add sodium acetate to the culture medium, leading to the production of 20.02 ± 0.19 g·L⁻¹, productivity of 0.55 g/L/h after 36 h of fermentation, and a residual sugar concentration of 12.59 ± 0.51 g·L⁻¹. These results demonstrate the potential of W. ginsengihumi BMI for the production of lactic acid by xylose fermentation since it is carried out at 50 °C, indicating a path for future studies