Rapid quantification of pullulan in fermentation broth using UV-visible spectroscopy and partial least squares...

Area of Science:

• Biotechnology
• Analytical Chemistry
• Spectroscopy

Background:

• Traditional exopolysaccharide (EPS) quantification involves lengthy, multi-step sample preparation, hindering real-time bioprocess monitoring.
• Developing non-destructive analytical methods with minimal sample preparation is crucial for efficient bioprocess control.

Purpose of the Study:

• To develop and validate a rapid, non-destructive method for quantifying pullulan in fermentation broth using spectral data.
• To compare different spectral region selection strategies for partial least squares (PLS) regression models.

Main Methods:

• Partial least squares (PLS) regression models were developed using spectral data (204-400 nm) to quantify pullulan.
• Optimization strategies including genetic algorithm, particle swarm optimization, competitive adaptive reweighted sampling, and adaptive bottom-up space exploration were employed for spectral region selection.
• Models were validated using cross-validation, assessing metrics like Root Mean Square Error of Cross-Validation (RMSE_CV) and R-squared (R_CV^2).

Main Results:

• PLS regression models successfully quantified pullulan in both cell-free supernatant and redissolved precipitated forms.
• The full-spectrum model for cell-free supernatant demonstrated high accuracy (RMSE_CV: 0.020 g l^-1, R_CV^2: 0.997).
• Adaptive bottom-up space exploration yielded the best results, achieving low RMSE_CV values with minimal spectral variables, suitable for real-time monitoring.

Conclusions:

• The developed PLS-based spectral method provides a rapid, non-destructive, and accurate alternative for pullulan quantification in fermentation broth.
• This approach significantly reduces sample preparation time, addressing a key bottleneck in bioprocess monitoring.
• The method's adaptability suggests potential for real-time monitoring and control of bioprocesses involving similar biopolymers.