The strains of S. pneumoniae serotype 6A (CMCC 31476), 6B (CMCC 31490), 11A (CMCC 31572), and 33F (CMCC 31847) were purchased from the National Center for Medical Culture Collections. These lyophilized strains were used to establish both master seed lots and working seed lots. Tests were conducted to ascertain the characteristics of the seed lots by the National Institutes for Food and Drug Control of China according to Pharmacopeia, the purity of bacterial cultures was verified by methods of suitable sensitivity. These include inoculation into the solid medium, examination of colony morphology, microscopic examination of Gram-stained smears and culture agglutination with specific antisera obtained from Statens Serum Institute (Copenhagen, Denmark), following the immunochemical method (2.7.1) detailed in the European Pharmacopoeia (11th edition).

A 1 mL aliquot from the working seed lots was used to inoculate solid culture medium and incubated in a CO₂ incubator for 9–15 h (with a CO₂ concentration of 10% and a cultivation temperature of 36°C); subsequently, the strain was transferred to a 3 L flask containing 1 L of liquid medium and incubated at 36°C to achieve an optical density (OD) of 0.4–0.6 at 600 nm. Once the OD600 reached 0.4∼0.6, the culture was inoculated into bioreactors containing fermentation medium at a volume of 5%–10%.

The lab-scale fed-batch fermentations were performed in a 15 L bioreactor (Shanghai Bailun Biotechnology, China) with a working volume of 10 L. The bioreactor cultivation was conducted under a 100% CO₂ atmosphere at a flow rate of 0.1 vvm. The pH was controlled by the addition of 5 M NaOH, maintaining a pH level of 7.2. The temperature was kept constant at 37°C as this culture temperature can potentially increase CPS yield according to previous study (Li et al., 2024), and the agitation was provided by a mechanically driven impeller at 90 rpm. Throughout the fermentation process, a continuous glucose medium was employed to ensure that the glucose concentration remained above 5 g/L within the fermentation system. Samples were collected at various time intervals for the measurement of optical density at 600 nm and glucose concentration analysis. At the final stage of fermentation, inactivation reagents were introduced for cellular lysis and sterilization. The fermentation broth was treated with 0.1% DOC, stirred for 40 min, and then incubated at 4°C for 12 h to achieve sterilization. After cell lysis using DOC, the samples were centrifuged to separate the pellets and supernatants, the polysaccharide content in the supernatant was determined.

Three fermentation replicates were conducted for each type of peptone. In these replicates, the fermentation medium formulation and process parameters were kept identical to assess the process variation in the growth curves. The average and standard deviation values were then plotted on the growth curves using Excel. Due to the fact that optical density of medium containing different peptones is quite low, all falling below 0.1. Therefore, their contribution to the total optical density is relatively minor.

A streamlined purification process was developed, which includes the following steps:

• Ultrafiltration and Diafiltration: After sterilization, the fermentation broth was centrifuged at 8000 rpm for 30 min to separate the supernatant. This supernatant was then subjected to ultrafiltration using a 100 kDa membrane to concentrate the solution. Subsequent diafiltration was performed to eliminate small molecular impurities.

The content of total polysaccharide was determined based on the rate nephelometric method using specific antisera obtained from Statens Serum Institute (Copenhagen, Denmark), following the immunochemical method (2.7.1) detailed in the European Pharmacopoeia (11th edition). The purified CPSs were subjected to a series of tests according to European Pharmacopoeia including: Protein (2.5.16), Nucleic acids (2.5.17), Total nitrogen (2.5.9), Phosphorus (2.5.18), Molecular size (2.2.30), Uronic acids (2.5.22), Hexosamines (2.5.20), Methyl pentoses (2.5.21), O-Acetyl groups (2.5.19).

Molecular-size and molecular-mass distribution were determined by size-exclusion chromatography (2.2.30) combined with an appropriate detection system. The size exclusion chromatography (SEC) liquid chromatography separation was conducted using a Shimadzu high performance liquid chromatography (HPLC) system (Shimadzu, Japan). The separation was conducted in a TOSOH TSKgel G5000PWxL column (7.8 mm × 30 cm, 13 µm particle size, Tosoh Bioscience, Tokyo, Japan). Both multi-angle light scattering (MALS) detector (DAWN HELEOS-II) and differential refractometers (Optilab T-rEX) (Wyatt Technology Corp., Santa Barbara, CA, United States) were connected to the UV detector. Throughout all experiments performed in this study, the detectors were arranged in the following sequence: SEC-UV-MALS-RI.

Analysis of CPS by ¹H NMR spectroscopy (Bruker, Avance III 400 and 600) was conducted at Beijing Center for Physical and Chemical Analysis, Beijing, China. Prior to analysis, the CPS samples were freeze-dried and reconstituted to final concentration of 2–4 mg/mL in deuterium oxide. In ¹H NMR data, splitting patterns are indicated as follows: s for singlet, d for doublet, t for triplet, q for quartet, m for multiplet, and br for broad singlet. NMR chemical shifts (δ) are reported in parts per million (ppm), and coupling constants (J) are reported in Hertz (Hz).

GraphPad Prism was employed to conduct a one-way analysis of variance (ANOVA) to assess the influence of soy peptones on the molecular size and molecular weight of the refined CPSs for serotypes 6A, 6B, 11A, and 33F.

For S. pneumoniae serotype 6A, A3X soy peptone and Casein Tryptone showed similar growth curves, while Difco soy peptone and HY-SOY 4D soy peptone demonstrated consistent growth trends with higher maximum bacterial densities. Notably, HY-SOY 4D soy peptone supported slightly slower growth (Figure 1). Corresponding CPS concentrations were highest with A3X soy peptone (727 mg/L), exceeding Difco soy peptone (655 mg/L), Casein Tryptone (639 mg/L) and HY-SOY 4D soy peptone (625 mg/L) (Table 2).

S. pneumoniae serotype 6B exhibited similar growth curves with A3X soy peptone and Difco soy peptone, reaching stationary phase faster than the other peptones. Casein Tryptone and HY-SOY 4D soy peptone showed similar growth trends with slightly slower growth observed with HY-SOY 4D (Figure 2). Final CPS yields were highest with A3X soy peptone (793 mg/L), followed by HY-SOY 4D soy peptone (742 mg/L), Difco soy peptone (681 mg/L), and Casein Tryptone (695 mg/L) (Table 2).

In serotype 11A, A3X soy peptone showed a slightly higher maximum bacterial density, with HY-SOY 4D soy peptone again showing slower growth (Figure 3). CPS yields were highest with A3X soy peptone (377 mg/L), followed by Difco soy peptone (359 mg/L), Casein Tryptone (328 mg/L), and HY-SOY 4D soy peptone (319 mg/L) (Table 2).

Serotype 33F showed significant differences among nitrogen sources, with A3X soy peptone providing the highest bacterial density, followed by Difco soy peptone (*P < 0.05), HY-SOY 4D soy peptone (**p < 0.01) and Casein Tryptone (***p < 0.001) (Figure 4). CPS productivity mirrored these trends, with A3X soy peptone yielding 1,565 mg/L, higher than Difco soy peptone (1,335 mg/L), HY-SOY 4D soy peptone (1,215 mg/L), and Casein Tryptone (1,242 mg/L) (Table 2).

These results indicate that different soy peptones can moderately affect growth profiles and CPS productivity in S. pneumoniae fermentations. A3X soy peptone from ORGANOTECHINE consistently supported the fast pneumococcal growth and high CPS productivity across the four serotypes, followed by Difco soy peptone, HY-SOY 4D soy peptone and Casein Tryptone. On the contrary, the HY-SOY 4D soy peptone exhibited reduced pneumococcal density and a decelerated growth rate across the majority of serotypes. This characteristic could potentially facilitate a more extended period for the elongation of pneumococcal CPSs.

To determine the effect of nitrogen sources on the quality attributes of pneumococcal CPSs, we conducted three consecutive batches of a uniform fermentation and purification process. The purification process comprised ultrafiltration, acid precipitation, alcohol precipitation, final ultrafiltration and diafiltration, excluding traditional CTAB precipitation, phenol extraction, and costly chromatography steps. This streamlined process significantly reduced costs and time compared to conventional methods. The purified CPSs from each serotype were analyzed to determine the quality attributes of CPSs, including percentage content of components, molecular size, molecular weight, etc.

Analysis of purified CPSs from S. pneumoniae serotypes 6A, 6B, 11A, and 33F revealed that all CPSs met European Pharmacopeia standards regardless of nitrogen source. However, the type of nitrogen source significantly influenced the molecular size, molecular weight and content of functional chemical groups of the pure CPSs (Figures 5, 6). The results from Table 2 indicate that nitrogen source selection significantly influences the molecular size, molecular weight, and methyl pentose content of pneumococcal CPSs for serotypes 6A and 6B, without affecting residual protein and nucleic acid, nitrogen, or phosphorus content. HY-SOY 4D soy peptone supports the synthesis of higher molecular weight CPSs with higher content of methyl pentose, suggesting better immunogenicity. For serotype 11A, nitrogen sources moderately affect molecular size, molecular weight, and O-acetyl group content, with HY-SOY 4D soy peptone promoting higher molecular weight CPSs and higher O-acetyl content. In serotype 33F, nitrogen sources greatly impact molecular size and weight, with HY-SOY 4D soy peptone and Casein tryptone leading to higher molecular weight CPSs.

Overall, HY-SOY 4D soy peptone can support the synthesis of high molecular weight CPSs, with a higher content of functional chemical groups, which is associated with better immunogenicity (Howard et al., 1971; Loetscher, Mottlau, and Hochuli, 1992; Balázs et al., 2002; MacLennan and Vinuesa, 2002; Craxton et al., 2003; Xiao et al., 2006; Zhao, Zhao, and Xie, 2012; Soubal et al., 2013; Qiong et al., 2022). Despite this, HY-SOY 4D soy peptone is preferred over Casein tryptone due to its plant origin, lower risk of contamination (Calix et al., 2011a; Calix et al. 2011b; Calix et al. 2012; Camilli et al., 2014).

Peptones are derived from the partial hydrolysis of proteins from plant, yeast, or animal sources. The protein hydrolysis process can be achieved by using strong acids, bases and proteolytic enzymes, the final complex, nutritionally enriched product can be easily consumed by microbes and cells. The generic composition of a typical peptone includes 25% polypeptides, 30% free amino acids, 20% carbohydrates, 15% salts and trace metals, and 10% of other components (vitamins, organic acids). While peptones generally share similar components, the specific hydrolysis method and protein source can lead to distinct product profiles. Microorganisms require accessible carbon, nitrogen, phosphate, sulfur, trace metals, and vitamins for growth, which are readily provided by peptones. The choice of peptone must match the microorganism’s needs, with some preferring short-chain polypeptides and others long-chain ones.

HY-SOY 4D soy peptone, Casein tryptone, Difco soy peptone, and A3X soy peptone exhibit unique profiles in total nitrogen (TN), amino nitrogen (AN) Table 3. Their molecular weight distributions also vary significantly. For example, HY-SOY 4D has a molecular weight distribution with 48.6% below 500 Da and 20.4% between 500 and 1,000 Da, while A3X has 77.0% below 500 Da and 21.1% between 500 and 1,000 Da. This difference in small polypeptide percentages may affect S. pneumoniae growth and polysaccharide biosynthesis. Additionally, these peptones differ in their total and free amino acid compositions, which can influence microbial uptake and utilization. HY-SOY 4D contains notably higher levels of glutamic acid (11.5 mg/g) and tyrosine (2.1 mg/g) compared to other peptones. A3X stands out with a higher percentage of free amino acids, which can be directly absorbed by microorganisms, potentially impacting CPS synthesis.

The purified CPSs derived from fermentation with HY-SOY 4D were characterized using ¹H NMR spectroscopy, as depicted in Figure 7, and were compared with the structures reported in the literature (Gisch et al., 2013; Geno et al., 2015). The analysis revealed that our prepared CPSs possess a complete structure with a notably low impurity content; specifically, the content of CWPS in all samples is less than 6%. The impurity content within the CPS was determined by integrating the phosphocholine structure on LTA in the ¹H NMR spectrum. The acetylation sites of type 11A CPS are known to be complex (Zartler et al., 2009), and the degree of acetylation varies at each site. In conjunction with the literature, it was established that there are four acetylation sites present. Our capsules exhibit 100% acetylation at the sites on glycerol and galactose, and a partial acetylation on glucose. However, the acetyl resonances could not be definitively assigned to specific sites due to their complexity.