Analysis of chemometric models applied to Raman spectroscopy for monitoring key metabolites of cell culture

The Food and Drug Administration (FDA) initiative of Process Analytical Technology (PAT) encourages the monitoring of biopharmaceutical manufacturing processes by innovative solutions. Raman spectroscopy and the chemometric modeling tool partial least squares (PLS) have been applied to this aim for monitoring cell culture process variables. This study compares the chemometric modeling methods of Support Vector Machine radial (SVMr), Random Forests (RF), and Cubist to the commonly used linear PLS model for predicting cell culture components—glucose, lactate, and ammonia. This research is performed to assess whether the use of PLS as standard practice is justified for chemometric modeling of Raman spectroscopy and cell culture data. Model development data from five small-scale bioreactors (2 × 1 L and 3 × 5 L) using two Chinese hamster ovary (CHO) cell lines were used to predict against a manufacturing scale bioreactor (2,000 L). Analysis demonstrated that Cubist predictive models were better for average performance over PLS, SVMr, and RF for glucose, lactate, and ammonia. The root mean square error of prediction (RMSEP) of Cubist modeling was acceptable for the process concentration ranges of glucose (1.437 mM), lactate (2.0 mM), and ammonia (0.819 mM). Interpretation of variable importance (VI) results theorizes the potential advantages of Cubist modeling in avoiding interference of Raman spectral peaks. Predictors/Raman wavenumbers (cm⁻¹) of interest for individual variables are X1139–X1141 for glucose, X846–X849 for lactate, and X2941–X2943 for ammonia. These results demonstrate that other beneficial chemometric models are available for use in monitoring cell culture with Raman spectroscopy.     

Bioprocess in-line monitoring using Raman spectroscopy and Indirect Hard Modeling (IHM): A simple calibration yields a robust model

To increase the process productivity and product quality of bioprocesses, the in-line monitoring of critical process parameters is highly important. For monitoring substrate, metabolite, and product concentrations, Raman spectroscopy is a commonly used Process Analytical Technology (PAT) tool that can be applied in-situ and non-invasively. However, evaluating bioprocess Raman spectra with a robust state-of-the-art statistical model requires effortful model calibration. In the present study, we in-line monitored a glucose to ethanol fermentation by Saccharomyces cerevisiae (S. cerevisiae) using Raman spectroscopy in combination with the physics-based Indirect Hard Modeling (IHM) and showed successfully that IHM is an alternative to statistical models with significantly lower calibration effort. The IHM prediction model was developed and calibrated with only 16 Raman spectra in total, which did not include any process spectra. Nevertheless, IHM's root mean square errors of prediction (RMSEPs) for glucose (3.68 g/L) and ethanol (1.69 g/L) were comparable to the prediction quality of similar studies that used statistical models calibrated with several calibration batches. Despite our simple calibration, we succeeded in developing a robust model for evaluating bioprocess Raman spectra.     

Integration of Process-based Soil Respiration Models with Whole-Ecosystem CO2 Measurements

We integrated soil models with an established ecosystem process model (SIPNET, simplified photosynthesis and evapotranspiration model) to investigate the influence of soil processes on modelled values of soil CO₂ fluxes (R _Soil). Model parameters were determined from literature values and a data assimilation routine that used a 7-year record of the net ecosystem exchange of CO₂ and environmental variables collected at a high-elevation subalpine forest (the Niwot Ridge AmeriFlux site). These soil models were subsequently evaluated in how they estimated the seasonal contribution of R _Soil to total ecosystem respiration (TER) and the seasonal contribution of root respiration (R _Root) to R _Soil. Additionally, these soil models were compared to data assimilation output of linear models of soil heterotrophic respiration. Explicit modelling of root dynamics led to better agreement with literature values of the contribution of R _Soil to TER. Estimates of R _Soil/TER when root dynamics were considered ranged from 0.3 to 0.6; without modelling root biomass dynamics these values were 0.1–0.3. Hence, we conclude that modelling of root biomass dynamics is critically important to model the R _Soil/TER ratio correctly. When soil heterotrophic respiration was dependent on linear functions of temperature and moisture independent of soil carbon pool size, worse model-data fits were produced. Adding additional complexity to the soil pool marginally improved the model-data fit from the base model, but issues remained. The soil models were not successful in modelling R _Root/R _Soil. This is partially attributable to estimated turnover parameters of soil carbon pools not agreeing with expected values from literature and being poorly constrained by the parameter estimation routine. We conclude that net ecosystem exchange of CO₂ alone cannot constrain specific rhizospheric and microbial components of soil respiration. Reasons for this include inability of the data assimilation routine to constrain soil parameters using ecosystem CO₂ flux measurements and not considering the effect of other resource limitations (for example, nitrogen) on the microbe biomass. Future data assimilation studies with these models should include ecosystem-scale measurements of R _Soil in the parameter estimation routine and experimentally determine soil model parameters not constrained by the parameter estimation routine.     

Real‐time monitoring of high‐gravity corn mash fermentation using in situ raman spectroscopy

In situ Raman spectroscopy was employed for real‐time monitoring of simultaneous saccharification and fermentation (SSF) of corn mash by an industrial strain of Saccharomyces cerevisiae. An accurate univariate calibration model for ethanol was developed based on the very strong 883 cm^?1 C–C stretching band. Multivariate partial least squares (PLS) calibration models for total starch, dextrins, maltotriose, maltose, glucose, and ethanol were developed using data from eight batch fermentations and validated using predictions for a separate batch. The starch, ethanol, and dextrins models showed significant prediction improvement when the calibration data were divided into separate high‐ and low‐concentration sets. Collinearity between the ethanol and starch models was avoided by excluding regions containing strong ethanol peaks from the starch model and, conversely, excluding regions containing strong saccharide peaks from the ethanol model. The two‐set calibration models for starch (R² = 0.998, percent error = 2.5%) and ethanol (R² = 0.999, percent error = 2.1%) provide more accurate predictions than any previously published spectroscopic models. Glucose, maltose, and maltotriose are modeled to accuracy comparable to previous work on less complex fermentation processes. Our results demonstrate that Raman spectroscopy is capable of real time in situ monitoring of a complex industrial biomass fermentation. To our knowledge, this is the first PLS‐based chemometric modeling of corn mash fermentation under typical industrial conditions, and the first Raman‐based monitoring of a fermentation process with glucose, oligosaccharides and polysaccharides present. Biotechnol. Bioeng. 2013; 110: 1654–1662. © 2013 Wiley Periodicals, Inc.     

A multivariate calibration procedure for UV/VIS spectrometric monitoring of BHK‐21 cell metabolism and growth

Monitoring mammalian cell culture with UV–vis spectroscopy has not been widely explored. The aim of this work was to calibrate Partial Least Squares (PLS) models from off‐line UV–vis spectral data in order to predict some nutrients and metabolites, as well as viable cell concentrations for mammalian cell bioprocess using phenol red in culture medium. The BHK‐21 cell line was used as a mammalian cell model. Spectra of samples taken from batches performed at different dissolved oxygen concentrations (10, 30, 50, and 70% air saturation), in two bioreactor configurations and with two strategies to control pH were used to calibrate and validate PLS models. Glutamine, glutamate, glucose, and lactate concentrations were suitably predicted by means of this strategy. Especially for glutamine and glucose concentrations, the prediction error averages were lower than 0.50 ± 0.10 mM and 2.21 ± 0.16 mM, respectively. These values are comparable with those previously reported using near infrared and Raman spectroscopy in conjunction with PLS. However, viable cell concentration models need to be improved. The present work allows for UV–vis at‐line sensor development, decrease cost related to nutrients and metabolite quantifications and establishment of fed‐batch feeding schemes. © 2013 American Institute of Chemical Engineers Biotechnol. Prog., 30:241–248, 2014     

An on-line technique for monitoring propionic acid fermentation

An on-line technique, based on measuring the increase in pressure due to CO₂ release in a closed air-tight reactor, was used to evaluate the fermentation of lactate by propionibacteria. The method was applied to batch cultures of Propionibacterium shermanii grown in yeast extract/sodium lactate medium containing lactate as a carbon source under micro-aerophilic conditions. Gas pressure evolution was compared both with substrate consumption and metabolites production and with acidification and growth. Linear relationships were found between gas pressure variation, lactate consumption and propionate and acetate production. The technique also enabled the evaluation of total CO₂ produced, by taking account of pressure, oxygen and pH measurements. These results tend to show that this simple and rapid method could be useful to monitor propionic acid bacteria growth.     

Propionic acid fermentation by Propionibacterium acidipropionici: effect of growth substrate

Summary Batch propionic acid fermentations by Propionibacterium acidipropionici with lactose, glucose, and lactate as the carbon source were studied. In addition to propionic acid, acetic acid, succinic acid and CO₂ were also formed from lactose or glucose. However, succinic acid was not produced in a significant amount when lactate was the growth substrate. Compared to fermentations with lactose or glucose at the same pH, lactate gave a higher propionic acid yield, lower cell yield, and lower specific growth rate. The specific fermentation or propionic acid production rate from lactate was, however, higher than that from lactose. Since about equimolar acid products would be formed from lactate, the reactor pH remained relatively unchanged throughout the fermentation and would be easier to control when lactate was the growth substrate. Therefore, lactate would be a preferred substrate over lactose and glucose for propionic acid production using continuous, immobilized cell bioreactors. Correspondence to: S. T. Yang     

Isotopic and chemical investigations of Fe (III) reduction by Shewanella putrefaciens strain 200R

Experiments were conducted using the Fe⁺³‐reducing bacterium Shewanella putrefaciens strain 200R to determine the stable carbon isotope fractionation during dissimilatory Fe (III) reduction and associated lactate oxidation at circum‐neutral pH. Previous studies used equilibrium fractionation factors (~14.3‰) between bacterial biomass and synthesized fatty acids to identify the predominant carbon fixation pathways for some of the most frequently isolated microbes including Shewanella under anaerobic conditions. We investigated the carbon isotope disproportionation among organic carbon substrate (lactate), biomass and respired carbon dioxide at the lag to stationary phase of the growth curve. Ferric citrate and sodium lactate were used as electron acceptor and donor, respectively. Sodium bicarbonate or potassium phosphate was used as buffering agent. Iron (II), iron (III), dissolved inorganic carbon (DIC) and carbon isotope ratios were measured for both bicarbonate‐ and phosphate‐buffered systems. Carbon isotope ratio measurements were made on the respired CO₂ (as DIC) and microbial biomass for both buffering conditions. The fraction of lactate consumed was estimated using DIC as a proxy and was verified by direct measurement using HPLC. Our result showed that bicarbonate‐buffered system has an enhancing effect in the reduction process compared to the phosphate system. Both systems resulted in carbon isotope fractionations between the lactate substrate and DIC that could be modelled as a Rayleigh process. The biomass produced under both buffer conditions was depleted on average by ~2‰ relative to the substrate and enriched by ~5‰ relative to the DIC. This translates to an overall isotopic fractionation of 10–12‰ between the biomass and respired CO₂ in both buffering systems.     

Lactate Metabolism and Its Effects on Glucose Metabolism in an Excised Neural Tissue

Abstract: Chains of lumbar sympathetic ganglia, excised from 15-day-old chicken embryos, were incubated for 4 h at 36°C in a bicarbonate-buffered physiological salt solution containing 5.5 mM glucose and equilibrated with 5% CO₂–95% O₂. [U-¹⁴C]Glucose and [U-¹⁴C]lactate were used as tracers to measure the products of glucose and lactate metabolism, respectively, including CO₂, lactate, and constituents of the tissue. When 5 mM lactate was added to bathing solution containing 5.5 mM glucose, lactate carbon displaced 50–70% of the glucose carbon otherwise used for CO₂ production and provided about three times as much carbon for CO₂ as did glucose. The lactate addition increased the total carbon incorporated into CO₂ and into constituents of the tissue above those observed with glucose alone and also increased the lactate released to the bathing solution from [U-¹⁴C]-glucose. The latter increase was evidently due to an interference with reuptake of the lactate released from the ganglion cells, not to an increase in the cellular release itself. When the volume of bathing solution was increased 10-fold relative to that of the tissue, the average output of CO₂ from [U-¹⁴C]glucose during a 4-h incubation was decreased by 50% when 5 mM lactate was present but was not affected significantly in the absence of added lactate. It is concluded that the effect of changing volume in the presence of lactate was due to the effects of lactate on glucose metabolism described above and resulted from a lower average lactate concentration in the smaller volume than in the larger one, due to metabolic depletion of the added lactate. Consumable substrates other than lactate, such as glutamine and certain amino acids, also affected glucose metabolism.     

Physiological consequences of prolonged aerial exposure in the American eel,Anguilla rostrata: blood respiratory and acid-base status

Summary The physiological consequences of prolonged air-exposure on blood respiratory and acid-base properties were examined in the American eel (Anguilla rostrata). Eels displayed a low capacity for aerial gas transfer as indicated by pronounced increases and decreases in arterial CO₂ and O₂ tensions, respectively. The increase in arterial CO₂ tension contributed to severe extracellular acidosis. The decrease in arterial O₂ tension, combined with a marked reduction in red blood cell pH and concomitant Bohr and Root effects, caused arterial O₂ content to decline to levels that were insufficient to support metabolic requirements, aerobically. Consequently, the rate of anaerobic glycolysis increased during air-exposure as suggested by a gradual elevation of blood lactate levels after 12 h. Increased anaerobic glycolysis and associated ATP hydrolysis and/or degradation of internal ATP stores further depressed blood pH as metabolic acid, produced by these processes, entered the circulation. Unlike other fishes previously examined, red blood cell pH was not regulated preferentially during the extracellular acidosis but simply conformed to the in vitro relationship between red blood cell and whole blood pH. Although capable of surviving prolonged air-exposure, the results demonstrate nevertheless and perhaps not surprisingly that eels, unlike true amphibious fishes that utilize gills or buccal epithelia for gas transfer, are not particularly well-adapted for gas exchange in air but do display an unusual tolerance to hypoxemia.Symbols and abbreviations B buffer value - Hct hematocrit - RBC red blood cell     

Effect of molecular hydrogen and carbon dioxide on chemo-organotrophic growth of Acetobacterium woodii and Clostridium aceticum

During growth of Acetobacterium woodii on fructose, glucose or lactate in a medium containing less than 0.04% bicarbonate, molecular hydrogen was evolved up to 0.1 mol per mol of substrate. Under an H₂-atmosphere growth of A. woodii with organic substrates was completely inhibited whereas under an H₂/CO₂-atmosphere rapid growth occurred. Under these conditions H₂+CO₂ and the organic substrate were utilized simultaneously indicating that A. woodii was able to grow mixotrophically. Clostridium aceticum differed from A. woodii in that H₂ was only evolved in the stationary phase, that the inhibition by H₂ was observed at pH 8.5 but not at pH 7.5, anf that in the presence of fructose and H₂+CO₂ only fructose was utilized.The hydrogenase activity of fructose-grown cells of C. aceticum amounted to only 12% of that of H₂+CO₂-grown cells. With A. woodii a corresponding decrease of the activity of this enzyme was not observed.     

Methanogenesis from lactate by a co-culture of Clostridium formicoaceticum and Methanosarcina mazei

Summary A co-culture of Clostridium formicoaceticum and Methanosarcina mazei converted lactate to methane and carbon dioxide at mesophilic temperatures and pH values near 7.0. Lactate was first converted to acetate by the homoacetogen, and then to CH₄ and CO₂ by the methanogen, with the second reaction as the rate-limiting step. The methane yield was about 1.45 mol/mol lactate. These two organisms formed a mutualistic association and may be useful together with the homolactic bacterium Stretococcus lactis to convert lactose to methane. Offprint requests to: S. T. Yang     

Influence of pH on Terminal Carbon Metabolism in Anoxic Sediments from a Mildly Acidic Lake

The carbon and electron flow pathways and the bacterial populations responsible for transformation of H₂-CO₂, formate, methanol, methylamine, acetate, glycine, ethanol, and lactate were examined in sediments collected from Knaack Lake, Wis. The sediments were 60% organic matter (pH 6.2) and did not display detectable sulfate-reducing activity, but they contained the following average concentration (in micromoles per liter of sediment) of metabolites and end products: sulfide, 10; methane, 1,540; CO₂, 3,950; formate, 25; acetate, 157; ethanol, 174; and lactate, 138. Methane was produced predominately from acetate, and only 4% of the total CH₄ was derived from CO₂. Methanogenesis was limited by low environmental temperature and sulfide levels and more importantly by low pH. Increasing in vitro pH to neutral values enhanced total methane production rates and the percentage of CO₂ transformed to methane but did not alter the amount of ¹⁴CO₂ produced from [2-¹⁴C]acetate (~24%). Analysis of both carbon transformation parameters with ¹⁴C-labeled tracers and bacterial trophic group enumerations indicated that methanogenesis from acetate and both heterolactic- and acetic acid-producing fermentations were important to the anaerobic digestion process.     

Novel synthetic co-culture of Acetobacterium woodii and Clostridium drakei using CO2 and in situ generated H2 for the production of caproic acid via lactic acid

Acetobacterium woodii is known to produce mainly acetate from CO₂ and H₂, but the production of higher value chemicals is desired for the bioeconomy. Using chain-elongating bacteria, synthetic co-cultures have the potential to produce longer-chained products such as caproic acid. In this study, we present first results for a successful autotrophic co-cultivation of A. woodii mutants and a Clostridium drakei wild-type strain in a stirred-tank bioreactor for the production of caproic acid from CO₂ and H₂ via the intermediate lactic acid. For autotrophic lactate production, a recombinant A. woodii strain with a deleted Lct-dehydrogenase complex, which is encoded by the lctBCD genes, and an inserted D-lactate dehydrogenase (LdhD) originating from Leuconostoc mesenteroides, was used. Hydrogen for the process was supplied using an All-in-One electrode for in situ water electrolysis. Lactate concentrations as high as 0.5 g L^–1 were achieved with the AiO-electrode, whereas 8.1 g L^–1 lactate were produced with direct H₂ sparging in a stirred-tank bioreactor. Hydrogen limitation was identified in the AiO process. However, with cathode surface area enlargement or numbering-up of the electrode and on-demand hydrogen generation, this process has great potential for a true carbon-negative production of value chemicals from CO₂.     

Coralline algal structure is more sensitive to rate,rather than the magnitude,of ocean acidification

Marine pCO₂ enrichment via ocean acidification (OA), upwelling and release from carbon capture and storage (CCS) facilities is projected to have devastating impacts on marine biomineralisers and the services they provide. However, empirical studies using stable endpoint pCO₂ concentrations find species exhibit variable biological and geochemical responses rather than the expected negative patterns. In addition, the carbonate chemistry of many marine systems is now being observed to be more variable than previously thought. To underpin more robust projections of future OA impacts on marine biomineralisers and their role in ecosystem service provision, we investigate coralline algal responses to realistically variable scenarios of marine pCO₂ enrichment. Coralline algae are important in ecosystem function; providing habitats and nursery areas, hosting high biodiversity, stabilizing reef structures and contributing to the carbon cycle. Red coralline marine algae were exposed for 80 days to one of three pH treatments: (i) current pH (control); (ii) low pH (7.7) representing OA change; and (iii) an abrupt drop to low pH (7.7) representing the higher rates of pH change observed at natural vent systems, in areas of upwelling and during CCS releases. We demonstrate that red coralline algae respond differently to the rate and the magnitude of pH change induced by pCO₂ enrichment. At low pH, coralline algae survived by increasing their calcification rates. However, when the change to low pH occurred at a fast rate we detected, using Raman spectroscopy, weaknesses in the calcite skeleton, with evidence of dissolution and molecular positional disorder. This suggests that, while coralline algae will continue to calcify, they may be structurally weakened, putting at risk the ecosystem services they provide. Notwithstanding evolutionary adaptation, the ability of coralline algae to cope with OA may thus be determined primarily by the rate, rather than magnitude, at which pCO₂ enrichment occurs.     

Inorganic carbon acquisition in the acid‐tolerant alga Chlorella kessleri

The ability of the freshwater alga, Chlorella kessleri, to maintain a carbon concentrating mechanism when grown at acid pH was investigated. The alga grows over the pH range 4.0–9.0 and was found to take up bicarbonate and CO₂ actively when grown at pH 6.0. However, when grown at acid pH (below 5.5), it does not have active CO₂ uptake. The acidotolerant species maintained an internal pH of 6.1–7.5 over the external pH range 4.5–7.5, thus the pH difference between the cell interior and the external medium was large enough to allow for the diffusive uptake of CO₂ at acid external pH. Mass spectrometric monitoring of O₂ and CO₂ fluxes by suspensions of C. kessleri, grown at acid pH, and maintained at pH 7.5 showed that the rates of O₂ evolution did not exceed those of CO₂ uptake. The final CO₂ compensation concentrations of 14.0–17.7 µM reached by photosynthetic cells were above the CO₂ equilibrium concentration in the external medium, indicating a lack of active CO₂ uptake at acid pH. Chlorella kessleri accumulated CO₂ with internal concentrations that were 9.9, 18.7 and 22.7‐fold that of the external medium for cells grown, respectively, at pH 4.5, 5.0 and 5.5. The ability of C. kessleri cells to accumulate high intracellular concentrations of inorganic carbon at acid pH would provide a sufficiently high concentration of CO₂ at the active site of Rubisco thus allowing the alga to maintain growth rates similar to those at alkaline pH.     

pH and base counterion affect succinate production in dual-phase <Emphasis Type="Italic">Escherichia coli</Emphasis> fermentations

Succinate production was studied in Escherichia coli AFP111, which contains mutations in pyruvate formate lyase (pfl), lactate dehydrogenase (ldhA) and the phosphotransferase system glucosephosphotransferase enzyme II (ptsG). Two-phase fermentations using a defined medium at several controlled levels of pH were conducted in which an aerobic cell growth phase was followed by an anaerobic succinate production phase using 100% (v/v) CO₂. A pH of 6.4 yielded the highest specific succinate productivity. A metabolic flux analysis at a pH of 6.4 using ¹³C-labeled glucose showed that 61% of the PEP partitioned to oxaloacetate and 39% partitioned to pyruvate, while 93% of the succinate was formed via the reductive arm of the TCA cycle. The flux distribution at a pH of 6.8 was also analyzed and was not significantly different compared to that at a pH of 6.4. Ca(OH)₂ was superior to NaOH or KOH as the base for controlling the pH. By maintaining the pH at 6.4 using 25% (w/v) Ca(OH)₂, the process achieved an average succinate productivity of 1.42 g/l h with a yield of 0.61 g/g.     

The influence of carbon dioxide-depletion on growth and sinking rate of two planktonic diatoms in culture

Growth in relation to CO₂-depletion and CO₂-enrichment was investigated for the freshwater diatoms Asterionella formosa and Fragilaria crotonensis in batch cultures. Algal concentration and pH were measured during growth cycles, and inorganic carbon quantities determined by potentiometric Gran titrations and from pH-alkalinity relationships. After the primary growth with CO₂-depletion and pH increase, successive CO₂-enrichments induced further such cycles and produced a final three- to fivefold increase in algal biomass over that of unenriched controls. The extent of CO₂-depletion, and pH rise, was greater in later cycles, indicative of some cellular adaptation. Values of pH reached 9·7 for Asterionella and 9·9 for Fragilaria. The lowest residual quantities of free CO₂ were 0·1 and 0·03 μmol 1^-1 for Asterionella and Fragilaria respectively, which were less than 0·05% of the corresponding residual quantities of total CO₂. The primary limitation of CO₂-uptake and growth was probably related to the concentration of free CO₂, given the relative excess of other major nutrients (N, P, Si) in he media used. Limited of CO₂-uptake could be restored without CO₂ additions if the CO₂ present was redistributed between its several forms (increasing free CO₂) by the addition of strong acid, although growth was still restricted.

Limitation of CO₂-uptake, either by CO₂-depletion or the addition of an inhibitor of photo-synthesis (DCMU), increased the sinking rate of Asterionella cells from 0·3 to 1 m day^-1. The possible ecological implications of CO₂-pH-growth and CO₂-pH-buoyancy relationships are discussed, which may contribute to the frequent paucity of diatoms during summer in manv productive lakes.     

At-line monitoring of key parameters of nisin fermentation by near infrared spectroscopy,chemometric modeling and model improvement

An analytical procedure has been developed for at-line (fast off-line) monitoring of 4 key parameters including nisin titer (NT), the concentration of reducing sugars, cell concentration and pH during a nisin fermentation process. This procedure is based on near infrared (NIR) spectroscopy and Partial Least Squares (PLS). Samples without any preprocessing were collected at intervals of 1 h during fifteen batch of fermentations. These fermentation processes were implemented in 3 different 5 l fermentors at various conditions. NIR spectra of the samples were collected in 10 min. And then, PLS was used for modeling the relationship between NIR spectra and the key parameters which were determined by reference methods. Monte Carlo Partial Least Squares (MCPLS) was applied to identify the outliers and select the most efficacious methods for preprocessing spectra, wavelengths and the suitable number of latent variables (n _LV). Then, the optimum models for determining NT, concentration of reducing sugars, cell concentration and pH were established. The correlation coefficients of calibration set (R _c) were 0.8255, 0.9000, 0.9883 and 0.9581, respectively. These results demonstrated that this method can be successfully applied to at-line monitor of NT, concentration of reducing sugars, cell concentration and pH during nisin fermentation processes.