Yeast-catalyzed alcoholic fermentation (sugar consumption) is a crucial step in industrial processes (food and biofuel) that results in the transformation of sugars (glucose and fructose) into alcohol and carbon dioxide . The yeast
A variety of techniques have been proposed for monitoring the sugar consumption by yeast, such as the measurement of the amount of CO2 released during the process using electrochemical CO2 probes  , optical fiber refractometry, density determination using flexural oscillators, and ultrasound techniques .
In addition, Piermarini et al. described a method for monitoring the consumption of glucose and fructose in real time . However, the disadvantage of this method is that it requires the construction of a biosensor that uses glucose oxidase as a bio-receptor. The production of both carbon dioxide and alcohol is a consequence of the biological activity of cells during the fermentation step. The online measurement of carbon dioxide in liquid (grape juice to wine) can be conducted by measuring changes in the medium resistance as a result of CO2 bubbles. Thus, the fermentation kinetics could be controlled (or monitored) by the measurement of the total carbon dioxide . It is known that carbon dioxide bubbles increase the value of the medium resistance .
In 2009, Perez et al. described a method for the real-time monitoring of the fermentation of sugar by yeast cells. These researchers measured changes in the medium resistance (Rm), which is defined as the resistance of a certain amount of electrolyte solution some distance from the working electrode and the reference electrode (Ruiz, G., Ph.D thesis, Universidad Nacional de Tucumán, Tucumán, Argentina, 2010); the changes in the resistance would be the result of the CO2 produced during the fermentation.
The main disadvantages of the technique developed by Perez and colleagues are the following. First, it is an off-line method, which implies that it is necessary to extract medium from the fermentation tank using a peristaltic pump. In addition, their method requires the use of temperature compensation because the temperature is not controlled during the assay. The measurement time using our technique is 80 minutes, whereas the duration of the experiments conducted by Perez and co-workers was 200 h, which allowed the growth and fermentation of the yeast. The duplication time of
Also, Perez and co-workers used a relatively high frequency to measure Rm, which increases the signal-to-noise ratio.
In this work, we propose a method to follow the sugars consumption by yeast.
As detailed above, the yeast does not exhibit significant growth during our assay time, which implies that our assay monitors the speed at which the yeast cells consume sugars. The fermentation is followed by the measurement of the changes in the medium resistance (Rm). We used Baker’s yeast (Levex) resuspended in a salt medium as the biological material in our assays.
The Rm was measured at 700 Hz and 50 mV with a cell in a tripolar configuration using a gold electrode as the working electrode. This novel method improves the existent technologies to monitor the fermentation progress using the amount of CO2 produced because it only uses a gold electrode, which economizes the process . In addition, the assay is performed at a low frequency, which decreases the measurement noise. Thus, the objective of this work is to follow the sugar consumption in yeast measuring the effect of the bubbles produced by yeast in the electrical medium resistance.
There were two different mediums used in the assay:
Saline medium (SM), which consists of 20 mM KH2PO4, 30 mM KCl, and 1 mM MgCl2 (Cicarelli, Reagents S.A. San Lorenzo, Santa Fe, Argentina) and has a pH of 6.5.The saline medium is used to keep the yeast on a basal metabolic state, so that by adding glucose or sucrose, the only activity will correspond to the metabolism of these sugars.
Yeast cell suspension (CS). The CS was obtained by first dissolving 3.5 g of dry yeast Levex® (Levex, Buenos Aires, Argentina) in 15 ml of SM and stirring the mixture for 5 min with a magnetic stirrer. The solution was then centrifuged for 10 min at 5000 rpm, and the supernatant was discarded. The cells were then resuspended in 15 ml of SM, and the solution was stirred magnetically. This process was repeated 4 times to ensure the complete removal of biological material ions.
The electrochemical measurements were performed using a Solartron 12508W impedance analyzer (Solartron Group Ltd. Victoria Road, Farnborough Hampshire England) composed of a Solartron 1287 Electrochemical Analyzer and a Solartron 1250 Frequency Response Analyzer with the software provided by the manufacturer (ZPlot®, Scribner Associates Inc.).
The electrodes were gradually polished in a series of steps and with different roughness degrees: First the electrode was polished with Diamond pasta Praxis® with a roughness of 6 um, then with Diamond pasta Praxis® with a roughness of 3 um and last with alumina powder with a roughness of 1 um. The size of the grain matches the European standards (FEPA standard 43-GB-1984; R1993) .
Two different cells were used in this work. A tripolar cell, which is composed of a working electrode (WE), an Ag/AgCl reference electrode (Re1) and an AISI 304 stainless-steel concave counter-electrode (CE) with a diameter of 85 mm, was used (Fig. 1). The WE is a solid cylinder (1.5 cm long), and only 1 cm2 of its transversal section is exposed and coated with gold (the rest was insulated with Grilon). The area of the CE was larger than that of the working electrode to minimize its impedance.
An AC signal of 50 mV with a frequency of 700 Hz was applied to the electrochemical cell. We used 700 Hz, because at this frequency the medium total resistance is equal to Rm (Rm is the medium resistance between Re1 and CE) and the interface is negligible (Fig. 2).
To determinate the effect of CO2 bubbles on the medium resistance, we designed a tetrapolar cell with a length of 28 cm, a width of 2 cm and four stainless steel electrodes (Fig. 3).
First, 95 ml of the biological suspension was placed in the previously described tripolar cell. An AC signal of 50 mV with a frequency of 700 Hz was then applied to the cell. Fifteen minutes after the start of the experiment, 85 mM glucose was added. The same experiment was repeated using sucrose as the carbon source. In the control experiment, we added glucose to the BS medium in the absence of yeast. All experiments were performed at room temperature (25±0.1 °C throughout the assay time).
This experiment was conducted to calibrate the developed method. The glucose concentration was determined using a Prodigy Autocode blood glucose meter (Diagnostic Device, Inc. Miami, Florida, USA). This measurement of the glucose concentration and the bioimpedance monitoring of the sugar metabolism were performed simultaneously.
We performed this experiment to demonstrate that the changes in the Rm were due to the presence of CO2 bubbles. First, 25 ml of the biological sample was placed in the previously described tetrapolar cell; no carbon source was used in this experiment. The CO2 gas was injected at the bottom of the cell from a high-pressure cylindrical vessel. The experiment was conducted at 20 kHz in a bipolar configuration. At this frequency, all of the variations in the cell are reflected in the medium resistance (Rm); thus, we expected changes in the Rm% as a result of the bubbling.
All of the experiments were conducted at 25 °C.
Glucose (85 mM) was added to the base solution in the absence of yeast to demonstrate that the yeast fermentation is responsible for the changes in the Rm%. The composition of the base solution and the concentrations of yeast and glucose were the same as those used in the previous experiments.
The plate count technique was performed in YPS agar (Yeast extract, Peptone, Sucrose) medium with the following concentration: Yeast extract 10 mg/L; Peptone 20 mg/L, Sucrose 20 mg/L and Agar 1,5%.
Fig. 4 shows a plot of the Rm% as a function of time.
The glucose was added 15 min after the beginning of the experiment. Approximately 7 min after the addition of glucose, the Rm% began to gradually increase to 700%. The Rm% then started to decrease until it reached approximately the value observed prior to the glucose addition. The time interval during which this increase and decrease occurred is approximately 14 min. We then performed the same experiment using sucrose as the carbon source. Fig. 5 shows the plot of the resulting Rm% as a function of time.
The sucrose was added 16 min after the beginning of the experiment. Approximately 10 min after the addition of sucrose, the Rm% increased to approximately 1200%. As shown, the maximum Rm% is reached after 15 minutes, whereas the maximum Rm% is attained within 8 minutes when glucose was used as the carbon source (Fig. 3). In addition, the time interval from the time at which the Rm% begins to increase to the end of its decrease is approximately 24 minutes when sucrose is used as the carbon source, whereas the corresponding time interval using glucose as the carbon source is 17 minutes.
Fig. 6 shows the derivative of Rm% for glucose and sucrose.
The sugar was added at t=0. We only consider the first 10 min, since the addition of sugar, in the comparison of the derivatives because there is a large amount of noise in the measurements obtained after this time point as a result of the bubbles. The derivative plots indicate the velocity at which Rm% changes as a result of the effects of CO2 bubbles over the electrode . An 11-point derivate algorithm was applied.
Fig. 6 shows that the glucose began to be consumed within 6 min after the start of the experiment, whereas the sucrose consumption started after 8 min. However, the sucrose assay exhibited a higher consumption rate than the glucose experiment. The addition of glucose to the SM demonstrated that the observed changes in Rm% were due to the yeast-catalyzed fermentation of the sugar. Fig. 7 shows the corresponding plot of Rm% as a function of time.
As shown, the changes in Rm% after glucose was added were approximately 1%. Thus, these changes are negligible compared with the changes in the Rm% observed in the CS after the addition of glucose (700%) or sucrose (1200%).
We used the previously described tetrapolar cell in this assay. Fig. 8 shows the plot of the corresponding Rm% as a function of time.
As shown in Fig. 8, the CO2 was insufflated at 7 min. Unlike the gradual increase in the Rm% shown in fig 3 and 4, the Rm% increased very rapidly from 0 to 19%. This difference is attributed to the fact that few CO2 bubbles are released at the beginning of the fermentation and that this number increases as the fermentation progresses. In this CO2 bubbling experiment, the bubble flow rate is more intensive that the obtained in the fermentation processes, which explains the sudden increase in Rm%. This finding confirms the hypothesis that the CO2 bubbles produced by the fermentation are responsible for the observed changes in the Rm%.
The bioimpedance measurement was performed simultaneously with the monitoring of the glucose concentration, which was achieved using a blood glucose meter. Fig. 9 shows plots of the Rm% and the glucose concentration [mM] as a function of time.
The experiment starts with the addition of glucose at t0=0 min. As shown in the figure, the glucose concentration starts to decrease 1 minute after the addition of glucose, whereas Rm% starts to change approximately 7 minutes after the addition of glucose (at this point, 20% of the glucose has been consumed).
The method presented here is based on monitoring the consumption of sugar by measuring the effect of the CO2 bubbles produced by fermentation on the medium resistance. The advantages of this method over existing technologies are the following. i) As shown in Fig. 3, Rm starts to increase 7 min after the addition of glucose. This feature is very important because the assays used by other researchers  who have measured the alcoholic fermentation of yeast using impedance require approximately 200 h, whereas the total assay times of our method are approximately 80 min and 100 min when using glucose and sucrose, respectively, as the carbon sources. In addition, Perez et al. detected the start of the fermentation after 50 h whereas we detected the start of the fermentation 7 minutes after the addition of the carbon source.
In Fig. 8 we can see that the changes in Rm depend on the amount of CO2 bubbles liberated by the fermentation. In the first few minutes of the fermentation, the amount of bubbles is low; however, as the fermentation progresses, this number increases until it induces changes in the Rm. These results are in agreement with the resistance results obtained by Perez et al. These researchers reported that the resistance Rm started to increase approximately 50 h after the start of the experiment.
We worked at room temperature, which varied less than 1°C in all tests. In previous literature, it was stated that variations of 1°C in temperature cause a variation of 2% in Rm (Felice, C. J., Ph.D Thesis, Universidad Nacional de Tucumán, Tucumán, Argentina, 1996). In our experiment we obtain variations of 700% and 1200% (glucose and sucrose respectively) so that a variation of 2% is negligible.
During sugar’s metabolism the pH changes from 5.28 units, prior to sugar addition, to 4.29 when the fermentation ends. According to Firstenberg-Eden  a change of 2 units in pH represent a change of 14% in conductance approximated.
Again, this change is negligible compared to the changes observed in our experiments.
We did not measure the growth of yeast: our fermentation occurs without significant growth of yeast because Rm% returned to its original value (its value prior to the addition of glucose or sucrose) after the sugar was consumed. If the yeast cells were experiencing growth, the Rm would stabilize at a different value . To confirm that the yeast does not grow during the assay we perform a plate count and obtained 1,6x109 UFC/mL before the addition of sugar and 1,68x109 UFC/mL after the fermentation ends. ii) The proposed method is inexpensive because it only requires a gold electrode and does not require the use of enzymes . In addition, the proposed method is quite simple because we applied an electric field with a known frequency and voltage instead of using other techniques, such as optical fiber refractometry and ultrasound.
It is important to note the difference in the plots obtained with different carbon sources (glucose and sucrose). The plot obtained using sucrose as the carbon source (Fig 4) showed that the increase in Rm was greater than that obtained using glucose as the carbon source (Fig 3). In addition, the time interval from the addition of the carbon source to the time at which the Rm returned to its original value (before the addition of the carbon source) was longer when sucrose was used as the carbon source than when glucose was used. Because sucrose is a disaccharide (glucose + fructose), the yeast cells use a cell wall invertase to catalyze the hydrolysis of sucrose into an equimolar mixture of glucose and fructose .
Here we are measuring the effect of the CO2 bubbles, produced by the fermentation of sugar by yeast, over the medium resistance. We are not measuring the CO2 concentration in this work because we do not use and electrochemical sensor to perform this. The saline medium used here allows us to monitor only the consumption of sugar by yeast.
The use of a transport media, in order to avoid the multiplication of cells, could serve to perform a comparative analysis of the metabolism of sugars between different yeast’s genera or subspecies
This paper presents a new method to monitor the sugar consumption in yeast. This method allows the real time monitoring of sugar’s fermentation in yeast, by measuring the effect of CO2 bubbles over the medium resistance. The CO2 bubbles, obtained as a product of sugar fermentation, produce an increase in the medium resistance (Rm) at 700 Hz. It proves to be possible to monitor the fermentation of different sugars (e.g., glucose or sucrose) and compare the speed at which they are consumed. The developed method has several advantages over existing technologies, such as its low cost, and reduced measurement time. A very important point is the fact that the observed changes are not due to the growth of yeast but only to the fermentation of sugars.