Two New Bacterial Species Discovered in the Breast Milk of Senegalese Women

Scientists have identified and formally described two previously unknown species of bacteria recovered from human breast milk, expanding the known diversity of a genus that thrives in nearly every environment on Earth. The two organisms, isolated from milk samples donated by nursing mothers in rural Senegal, have been named Pseudomonas senegalensis and Pseudomonas millioni, and their discovery offers a rare culture-based glimpse into a microbial world that has long been dominated by DNA sequencing alone. The work, published in the journal New Microbes and New Infections, was carried out by researchers based in Senegal and at the Institut Hospitalo-Universitaire Méditerranée Infection in Marseille, France, as part of a broader investigation into how the milk microbiota relates to infant nutritional status.

The genus Pseudomonas is one of the most versatile groups of bacteria known to science. First defined in 1894 by the German mycologist Walter Migula, it comprises Gram-negative, aerobic, rod-shaped bacteria equipped with flagella. Over more than a century of microbiological discovery, the genus has grown to encompass 381 validly published species, according to the List of Prokaryotic names with Standing in Nomenclature, with Pseudomonas aeruginosa serving as the type species. Members of the genus colonize soil, rivers, plants and animal hosts, and their roles range from promoting plant growth and cleaning contaminated environments to causing serious opportunistic infections in vulnerable patients. Among the many niches these bacteria occupy is human milk, a biological fluid whose microbial community helps seed the newborn gut and shapes early immune and metabolic development.

Despite the growing interest in the milk microbiome, African populations remain markedly under-studied, and most surveys of milk bacteria rely on culture-independent methods that detect genetic signatures without ever yielding a living isolate. Motivated by this gap, the research team turned to a culture-based, genome-resolved strategy. Breast milk samples were collected from mothers of both malnourished and healthy children in Niakhar, Senegal, with informed consent and approval from the Ethical and Scientific Committee of the Senegalese Ministry of Health and Public Hygiene. Ten to fifteen milliliters of milk from each participant were transported at four degrees Celsius to a laboratory in Dakar, aliquoted, frozen at minus eighty degrees Celsius, and later shipped on dry ice to Marseille for culturing, sequencing and analysis.

Two strains, designated Marseille-QA0332 and Marseille-QA0892, emerged from this effort. The first was obtained by direct plating of milk onto Columbia blood agar during routine colony counting, before any enrichment step. The second was recovered through a culturomics protocol, in which milk was inoculated into a series of liquid enrichment media and subcultured at scheduled intervals over thirty days; strain QA0892 appeared at the day-fifteen subculture from an aerobic blood-culture bottle containing MacConkey broth. Rigorous sterility controls, in which every batch of medium was tested for contamination before use, ensured that any bacterial growth traced back to the milk samples themselves rather than to laboratory reagents.

Initial identification attempts using MALDI-TOF mass spectrometry failed to match either isolate to a known species, prompting whole-genome sequencing. Comparison of 16S ribosomal RNA gene sequences showed that strain QA0332 shared its highest similarity, 97.98 percent, with Pseudomonas ceruminis, while strain QA0892 was 99 percent similar to Pseudomonas matsuisoli. Those figures alone might have suggested close relatives, but whole-genome comparisons told a different story. Average nucleotide identity values, which should exceed roughly 95 to 96 percent between strains of the same species, came in at just 81.1 to 81.4 percent for QA0332 against its nearest neighbors and 75.0 percent for QA0892 against P. matsuisoli. Digital DNA-DNA hybridization values, with a species threshold of 70 percent, were even lower, at 26.7 percent and 21.1 percent respectively. By every genomic yardstick, the two isolates represented lineages distinct from any previously described species.

The phenotypic portraits of the two bacteria are strikingly different. Pseudomonas senegalensis grows optimally at 37 degrees Celsius under aerobic conditions and forms circular, convex, yellowish, mucoid colonies. It is motile, non-spore-forming, oxidase- and catalase-positive, and capable of fermenting a broad range of sugars including glucose, xylose, arabinose and trehalose. It also produces urease and arginine dihydrolase, traits that help distinguish it from close relatives. Pseudomonas millioni, by contrast, prefers cooler temperatures, growing best at room temperature and 28 degrees Celsius in alkaline media, and its colonies mature from translucent punctiform dots into dry, filamentous, opaque mats. Uniquely among its comparisons, it reduces nitrate to nitrite and hydrolyzes esculin. Fatty acid profiling by gas chromatography further separated the two: the cell membranes of P. senegalensis are dominated by hexadecanoic acid and cis-9,10-methylene-hexadecanoic acid, while those of P. millioni feature a more balanced mixture of unsaturated and saturated fatty acids.

Genome architecture reinforced the case for two new species. Pseudomonas senegalensis carries a single circular chromosome of 4,577,173 base pairs with a GC content of 63.65 percent, encoding 4,173 genes, of which 4,079 are protein-coding. Pseudomonas millioni has a slightly larger chromosome of 4,658,126 base pairs, a GC content of 60.22 percent, and 4,327 genes. Neither genome contains plasmids, and functional annotation revealed repertoires dominated by amino acid transport, transcription, energy production and inorganic ion metabolism, consistent with metabolically versatile, free-living organisms. Mining of secondary metabolite biosynthetic gene clusters uncovered eleven candidate regions in P. senegalensis and eight in P. millioni, including a pseudomonine-like metallophore cluster in the former, suggesting a siderophore-based iron-scavenging system well suited to the iron-limited environment of milk. Both genomes also harbor carotenoid, hydrogen cyanide and N-acetylglutaminylglutamine amide clusters, although the researchers caution that these remain computational predictions rather than chemically validated products.

From a clinical standpoint, the news is reassuring. Screening against virulence factor databases found none of the hallmark weapons of Pseudomonas aeruginosa: no type III secretion system, no exotoxin A, no elastase or phospholipases, and no high-risk pathogenicity islands. No acquired resistance genes, such as extended-spectrum beta-lactamases or carbapenemases, were detected in either genome. The only resistance determinants identified were intrinsic chromosomal efflux pumps and regulators typical of the genus. One notable exception emerged in Pseudomonas millioni, which proved resistant to the antibiotic ceftazidime. A targeted search revealed a chromosomal class C beta-lactamase, bearing the three canonical catalytic motifs of Ambler class C enzymes, that accounts for this phenotype. Because the enzyme is intrinsic and not transmissible, the researchers conclude that both species pose a low biosafety risk, though they note that pathogenic potential cannot be formally excluded without infection models.

Perhaps the most intriguing evidence for the ecological identity of these bacteria comes from mining public metagenomic archives through the IMNGS platform. Both species were detected predominantly in human-associated ecosystems rather than environmental ones. Pseudomonas senegalensis appeared in 0.95 percent of human gut metagenomes and 0.69 percent of skin metagenomes, while Pseudomonas millioni was found in 1.53 percent of gut metagenomes and 0.50 percent of blood metagenomes. This pattern of low prevalence but consistent human association supports the idea that the two organisms are genuine members of the human microbiome rather than transient contaminants. The researchers argue that the combination of viable cultured isolates, verified sterile media and expected recovery of Pseudomonas from milk makes a laboratory origin for the strains implausible.

The discovery also carries a cautionary lesson about genome-based inference. Both genomes encode a carotenoid biosynthetic cluster, yet only Pseudomonas senegalensis forms visibly yellow colonies. This dissociation between genotype and phenotype demonstrates that the mere presence of a biosynthetic gene cluster does not guarantee its expression, tempering the functional conclusions that can be drawn from genome mining alone. With type strains deposited in the Collection de Souches de l’Unité des Rickettsies in Marseille and the Colección Española de Cultivos Tipo in Spain, and genome sequences available in GenBank, Pseudomonas senegalensis and Pseudomonas millioni now stand as formally recognized members of the bacterial tree of life, and as reminders of how much microbial diversity still hides in ordinary human fluids waiting to be cultured.

Subject of Research: Discovery and taxonomic description of two novel Pseudomonas species isolated from human breast milk in Senegal

Article Title: Pseudomonas senegalensis sp. nov. and Pseudomonas millioni sp. nov., two novel species isolated from the breast milk of Senegalese women

Article References: Pseudomonas senegalensis sp. nov. and Pseudomonas millioni sp. nov., two novel species isolated from the breast milk of Senegalese women. (n.d.). https://doi.org/10.1016/j.nmni.2026.101846

Image Credits: AI Generated

DOI: 10.1016/j.nmni.2026.101846

Keywords: Pseudomonas senegalensis, Pseudomonas millioni, breast milk microbiota, new bacterial species, taxonogenomics, Senegal, culturomics, whole-genome sequencing, antimicrobial resistance, human microbiome, MALDI-TOF, New Microbes and New Infections