Artificial Sputum Medium Formulation Impacts Pseudomonas aeruginosa Phenotype and Chemotype

Artificial sputum medium (ASM) is a class of in vitro bacterial culture medium intended to mimic the nutritional environment of cystic fibrosis (CF) pulmonary mucus. One of the most commonly studied microbes in ASM is Pseudomonas aeruginosa, a prevalent and dangerous pathogen of the CF pulmonary microbiome. Many ASM formulations have been reported in literature, with differing nutrient concentrations and availability. Here, we show that common formulations of ASM yield different phenotypes and chemotypes of P. aeruginosa. Further, we demonstrate that iron in commercial porcine gastric mucin (PGM) is sufficient to alter production of P. aeruginosa siderophores in the chemically defined ASM, synthetic CF medium 1 (SCFM1). These results highlight that the choice of ASM formulation for in vitro investigations of microbial pathogenicity, physiology, and interactions should be carefully considered. IMPORTANCE In vitro culture media are being developed to resemble the in vivo nutritional environment more closely. These culture media are used to investigate microbial pathogenicity and ecology in environments that are more reflective of disease states. In cystic fibrosis (CF), a number of different artificial sputum media (ASM) formulations have been created to recapitulate the CF lung environment. However, these ASM have different sources and concentrations of nutrients. Here, we cultured Pseudomonas aeruginosa in nine different formulations of ASM. P. aeruginosa is the primary pathogen causing lung infection in CF. We show that different ASM formulations lead to different phenotypes and chemotypes by P. aeruginosa and one component of ASM, mucin, contains high levels of iron, which may affect P. aeruginosa physiology.

experimental results suggested that these metabolites are more available in CF sputum 111 than in SCFM2 (11). We used PGM was used as the mucin source as it is a cost 112 effective and common replacement of .  In ASMDM ASM, P. aeruginosa grew heterogeneously with a layer of growth at the air-119 liquid interface, giving way to turbid growth further down the culture. In SDSU ASM, 120 PAO1 formed a green, floating structure near the center of the well that reached the 121 surface of the medium. PAO1 grew as a turbid culture with bright green coloration in 122 SCFM1; while in SCFM2 and SCFM3, PAO1 growth was more structured with a 123 distinctly blue color. 124 P. aeruginosa produces similar specialized metabolites in all ASM formulations. 125 As metabolism often underlies phenotypic changes, we hypothesized that different ASM   (40,48,56,57), among others (58-60). We 185 quantified four PHZs: 1-hydroxyphenazine (1-HP), pyocyanin (PYO), phenazine-1-186 carboxamide (PCN), and phenazine-1-carboxylic acid (PCA) (Fig. 3A Phenazines). 187 While PHZs were detected in all PAO1 cultures, total PHZ production by PAO1 varied 188 between formulations with lowest production in Soothill ASM and highest production in 189 SCFM2 and SCFM3 (Fig. 3B Phenazines). The low levels of PHZs produced by PAO1 190 in Soothill ASM is likely due to the absence of amino acids in the medium. In addition to 191 variations in total PHZ production, the production of individual PHZs varied. In Soothill 192 ASM, PAO1 produced primarily PCA. PCA has been implicated in inducing P. ASMDM, Cordwell, and SDSU ASM formulations. PYO is produced highest in SCFM2 199 and SCFM3, which contributes to the characteristic blue phenotype of these cultures.

208
Like PHZ production, P. aeruginosa production of individual and total AQs varied 209 depending upon ASM formulation. We quantified forty-three AQs from the PAO1 ASM 70.0% of RHLs produced (Fig. 5C Rhamnolipids). The production of RHLs is 233 dependent on the rhl QS system; specifically the binding of the autoinducer C4-HSL, 234 produced by RhlI, to the transcription factor RhlR (76). The RhlI/RhlR QS system is 235 activated by the LasI/R QS system and directs the regulation of many virulence 236 encoding genes (77, 78). Lower production levels of RHLs by PAO1 may indicate 237 altered QS signaling, which may affect the production of other virulence factors.

238
Strikingly, production of siderophores by P. aeruginosa was almost exclusive to 239 SCFM1 cultures (Fig. 6  family were prioritized for annotation based upon coloration of the pie chart, which 245 indicated high production by PAO1 in SCFM1 (Fig. 6A Siderophores). Although PVD 246 biosynthesis was reported to be upregulated in SCFM1 (7), we were surprised to detect 247 PVDs as the production of this molecular family by PAO1 is below our limit of detection 248 when cultured in most media. The rapid visualization of molecular families and their 249 production levels in FBMN enabled us to identify this molecular family in a non-biased 250 manner, when we were admittedly not expecting to observe its production.

251
Addition of commercial PGM to SCFM1 reduces PAO1 siderophore production 252 due to the presence of iron. As the production of PCH and PVD is triggered by low 253 environmental iron (82), the production of PCH and PVD by P. aeruginosa in SCFM1 254 was surprising as all ASM formulations contain similar concentrations of added iron.

261
To replicate the CF lung environment more closely, SCFM1 was modified by the 262 addition of DNA, DOPC, GlcNAc, and mucin to create SCFM2 (11). As PGM was the 263 most complex additive to our SCFM2 formulation, we hypothesized that commercial sputum ranges from 16-30 mg/mL (1). In the lung, mucin is a source of nutrients for 268 inhabiting microbes (2, 87, 88) and also is involved in microbial adhesion (87), 269 aggregation (89, 90) and dispersion (83).

270
To test if the addition of PGM to SCFM1 altered siderophore production by P. would likely lead to different results. The P. aeruginosa AQ HQNO is known to alter S. 299 aureus susceptibility to antibiotics (96). Therefore, differential production of HQNO by P.  Collection (Clarivate Analytics) was searched for papers with the 'TOPIC' terms: 346 "sputum medium", "sputum media", "cystic fibrosis medium", "cystic fibrosis media", 347 "sputa media", "sputa medium", "CF media", or "CF medium" in the title, abstract, or 348 keywords fields on May 10, 2020. All terms were searched as shown, including 349 quotation marks. The results of this search were analyzed using the Web of Science 350 tools 'Create Citation Report' and 'Analyze Results'. The Analyze Results tool was used 351 to download data corresponding to Web of Science Categories. The citation report was 352 used to select ASM formulations that were commonly cited.

353
Preparation of media: LB broth was prepared according to manufacturer instructions.

358
Antibiotics were not included in any medium. For the Cordwell formulation (5), the 359 reported 0.1% (w/v) agar was excluded. For SCFM2 and SCFM3 (11), bovine 360 submaxillary mucin (BSM) was replaced with PGM at the published concentration, a 361 common substitution (12)(13)(14). Each ASM formulation was brought to the published pH 362 (3-11). All media was stored in the dark at 4°C, checked for sterility prior to use, and 363 used within one month of preparation. 10 minutes, and a secondary mechanical disruption by pipetting using wide-bore pipette 383 tips. Disrupted samples were aliquoted for measurements of growth (CFUs), pH, and 384 metabolomics analysis. Three representative biological replicates from each culture 385 condition were serially diluted, spotted onto LB agar, incubated at 37°C overnight, and 386 counted to determine CFU/mL (Fig. S1 Growth of PAO1 in ASM).