**Meet the editor**

Angela Faustino Jozala is Professor at Universidade de Sorocaba. PhD in Fermentation Technology and Master of sciences in Food Technology, both at the Faculty of Pharmaceutical Sciences at USP. Has experience in industrial biotechnology and pharmaceutical microbiology, highlighting the production process and purification (up and dowstream) of biomolecules and biopolymers

of different applications in the areas of food, medicine and pharmaceutics.

## Contents

#### **Foreword XI**


#### Chapter 2 **Fermentation and Redox Potential 23** Chen-Guang Liu, Jin-Cheng Qin and Yen-Han Lin


## Foreword

Chapter 7 **Microbial Population Optimization for Control and**

Chapter 8 **Biosensors in Fermentation Applications 145** Jianguo Shi, Derong Feng and Yiwei Li

**Section 3 Products from Fermentation Process 159**

Chapter 9 **Biogas - Turning Waste into Clean Energy 161**

Chapter 10 **Production Processes for Monoclonal Antibodies 181**

Chapter 11 **Production of Lipopeptides by Fermentation Processes:**

**Methods for Bacterial Selection 199**

Chapter 12 **Lactic Acid Bacteria and Fermentation of Cereals and**

Chapter 13 **Solid-State Culture for Lignocellulases Production 255**

Chapter 14 **Anaerobic Digestion: I. A Common Process Ensuring Energy**

**the Production of Gaseous Biofuels 271**

Sompong O‐Thong

**VI** Contents

Skorupa Parachin

Talita Souza Carmo

Miguel J. Beltran-Garcia

**Pseudocereals 223**

Trujillo

K. Błaszczyk

**Improvement of Dark Hydrogen Fermentation 119**

Otávio Bravim da Silva, Lucas Silva Carvalho, Gabriela Carneiro de Almeida, Juliana Davies de Oliveira, Talita Souza Carmo and Nádia

Lucas Silva Carvalho, Otávio Bravim da Silva, Gabriela Carneiro de Almeida, Juliana Davies de Oliveira, Nadia Skorupa Parachin and

**Endophytic Bacteria, Fermentation Strategies and Easy**

Denisa Liptáková, Zuzana Matejčeková and Ľubomír Valík

Ulises Durán Hinojosa, Leticia Soto Vázquez, Isabel de la Luz Membrillo Venegas, Mayola García Rivero, Gabriela Zafra Jiménez, Sergio Esteban Vigueras Carmona and María Aurora Martínez

**Flow and the Circulation of Matter in Ecosystems. II. A Tool for**

Anna Sikora, Anna Detman, Aleksandra Chojnacka and Mieczysław

Esteban Beltran-Gracia, Gloria Macedo-Raygoza, Juan Villafaña-Rojas, America Martinez-Rodriguez, Yur Yenova Chavez-Castrillon, Froylan M. Espinosa-Escalante, Paolo Di Mascio, Tetsuya Ogura and

Fermentation is a theme widely useful for food, feed and biofuel production. Indeed each of these areas, food industry, animal nutrition and energy production, has considerable pres‐ ence in the global market. Fermentation process also has relevant applications on medical and pharmaceutical areas, such as antibiotics production. The present book, *Fermentation Processes*, reflects that wide value of fermentation in related areas. It holds a total of 14 chap‐ ters over diverse areas of fermentation research.

This book includes a chapter about the importance and application of biosensor in fermenta‐ tion process helping to control the process. Two chapters deal with the application of fer‐ mentation for feed, focused on high-quality silage production and factors that affect rumen fermentation. Notably, these two chapters reveal the importance and advantage of fermen‐ tation process has to keep the animals healthy. Three chapters mention biofuel production, such as bioethanol and biogas. One of these reports kinetic model design to describe 1-G ethanol fermentation process and this model also has application for second-generation ethanol process. Two of these describe about biogas production by anaerobic digestion proc‐ ess using microorganism for energy. The first one uses waste as substrate. One chapter presents the microbial population optimization for control and improvement of dark hydro‐ gen fermentation. Another one shows the importance of redox potential during fermenta‐ tion. It has two chapters on the medical and pharmaceutical areas. The first one shows the production of monoclonal antibodies by fermentation. The second one is about the produc‐ tion of lipopeptides by fermentation processes. Three chapters mention enzyme production, such as cellulase, xylanases and laccase, by solid-state and/or liquid fermentation by monoand/or coculture. One chapter refers to the application of lactic acid bacteria in food.

I sincerely hope that all areas covered by those chapters in this book will be interesting for researchers involved with fermentation process. I also hope new microorganisms could be applied in more unique fermentation processes to produce advanced bioproducts.

> **Thalita Peixoto Basso** PhD in Microbiology – Esalq/USP Brazil

## **Fermentation Process**

#### **Importance of the Fermentation to Produce High-Quality Silage Importance of the Fermentation to Produce High-Quality Silage**

Thiago Carvalho da Silva, Leandro Diego da Silva, Edson Mauro Santos, Juliana Silva Oliveira and Alexandre Fernandes Perazzo Thiago Carvalho da Silva, Leandro Diego da Silva, Edson Mauro Santos, Juliana Silva Oliveira and Alexandre Fernandes Perazzo

Additional information is available at the end of the chapter Additional information is available at the end of the chapter

http://dx.doi.org/10.5772/64887

#### **Abstract**

The objective of this chapter was to discuss the importance of the fermentation processes for silage making and how it affects the final quality of the silage. The preservation of the forage crops as silage is based on a fermentation process that lows the pH and preserves the nutritive value of the fresh crop. The main principle is the production of lactic acid by the lactic acid bacteria from the metabolism of the water-soluble carbohydrates in the fresh crop. However, different fermentations may occur into the silo environment and it depends on the availability of substrate, the microbial populations, the moisture content, and the buffering capacity of the crop at the ensiling. The fermentation is quite important in the ensiling process because it affects the nutritional quality of the silage and the animal performance. If the fermentation does not occur as recommended and the undesirable fermentations will take place, which will result in a total spoiled feed that is potentially risky for animals and human's health. Well-fermented silage can be used in diets for ruminant animals without any risk for their health and without compromise the productive performance.

**Keywords:** additives, ammonia nitrogen, mycotoxins, lactic acid, organic acids, pH

#### **1. Introduction—silage production and utilization**

Grazing is the most common and economical way to feed cattle; however, it is cannot be done over the entire year, due to the climatic conditions that limit the grasses growth. The availability of pastures in livestock systems depends on the seasons because the factors that affect plant

© 2017 The Author(s). Licensee InTech. This chapter is distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/3.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. © 2017 The Author(s). Licensee InTech. This chapter is distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/3.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

growth (e.g., temperature, luminosity, and rainfall) are different for each season, which leads to periods with high forage production and periods of its shortage. In the winter, for example, there is no forage production enough to feed the animals [1].

The choice of suitable forage conservation process to provide constantly feed, essentially depends of the climatic conditions at harvest. In hot areas with dry seasons, probably the haymaking is the best choice for forage preservation, because it is a simple technology, where the fresh crop is dehydrated after cutting and the material is stable and preserved after reach an adequate moisture content.

In tropical regions with hot and humid climates, it is difficult to produce high-quality hay, due to high humidity and frequent rainfall at the optimum stage of maturity for crop with better nutritional value. In this context, ensiling is an important method of forage preservation because it is not too dependent on weather as the haymaking. In addition, in many parts of world, the silage is the major source of energy in the total mixed rations of ruminants [2, 3]. Thus, the objective of this chapter was to describe the fermentation processes for silage making and its manipulation and how it affects the final quality of the silage, which includes the effects on animal performance and health.

#### **2. Importance of the fermentation for silage making**

According to [2], in short, the silage is made by keeping chopped crop air-tight in a silo, as follows: (1) the crops are harvesting and chopping in a specific length at the better nutritional value and proper moisture content; (2) application of continuously heavy weights to pack at adequate densities; (3) and complete sealing. The preservation of the forage crops as silage depends of anaerobic environment, because it is based on a lactic acid fermentation that decreases the pH and associated with high osmotic pressure that inactivates the microorganisms preserving the nutritive value of the fresh crop (**Figure 1**). Even the presence of some mycotoxins in the fresh crop may be denatured due to the acid pH of silage.

The main principle of silage is anaerobic environment and fermentation of the water-soluble carbohydrates in the fresh crop by the epiphytic lactic acid bacteria (LAB) and production of lactic acid. However, different fermentation pathways may occur into the silo environment, depending on the availability of substrate, the predominant microbial populations, the dry matter (DM) content, and the buffering capacity of the crop at the ensiling (**Figure 2**). In addition, the fermentation must be limited to a certain extent, because it alters the chemical composition of the feed. This process may last for days or months, which may result in silage containing high levels of alcohols, butyric acid, ammonia, amines, and acetic acid that represent the major silage losses. Generally, the epiphytic microbial populations found in growing crops include pseudomonas, actinomycetes, listeria, and mainly the LAB that we expect to dominate the fermentation process to produce high-quality silage (**Table 1**) [4].

**Figure 1.** Diagram of silage fermentation process [2].

growth (e.g., temperature, luminosity, and rainfall) are different for each season, which leads to periods with high forage production and periods of its shortage. In the winter, for example,

The choice of suitable forage conservation process to provide constantly feed, essentially depends of the climatic conditions at harvest. In hot areas with dry seasons, probably the haymaking is the best choice for forage preservation, because it is a simple technology, where the fresh crop is dehydrated after cutting and the material is stable and preserved after reach

In tropical regions with hot and humid climates, it is difficult to produce high-quality hay, due to high humidity and frequent rainfall at the optimum stage of maturity for crop with better nutritional value. In this context, ensiling is an important method of forage preservation because it is not too dependent on weather as the haymaking. In addition, in many parts of world, the silage is the major source of energy in the total mixed rations of ruminants [2, 3]. Thus, the objective of this chapter was to describe the fermentation processes for silage making and its manipulation and how it affects the final quality of the silage, which includes the effects

According to [2], in short, the silage is made by keeping chopped crop air-tight in a silo, as follows: (1) the crops are harvesting and chopping in a specific length at the better nutritional value and proper moisture content; (2) application of continuously heavy weights to pack at adequate densities; (3) and complete sealing. The preservation of the forage crops as silage depends of anaerobic environment, because it is based on a lactic acid fermentation that decreases the pH and associated with high osmotic pressure that inactivates the microorganisms preserving the nutritive value of the fresh crop (**Figure 1**). Even the presence of some

The main principle of silage is anaerobic environment and fermentation of the water-soluble carbohydrates in the fresh crop by the epiphytic lactic acid bacteria (LAB) and production of lactic acid. However, different fermentation pathways may occur into the silo environment, depending on the availability of substrate, the predominant microbial populations, the dry matter (DM) content, and the buffering capacity of the crop at the ensiling (**Figure 2**). In addition, the fermentation must be limited to a certain extent, because it alters the chemical composition of the feed. This process may last for days or months, which may result in silage containing high levels of alcohols, butyric acid, ammonia, amines, and acetic acid that represent the major silage losses. Generally, the epiphytic microbial populations found in growing crops include pseudomonas, actinomycetes, listeria, and mainly the LAB that we expect to dominate the fermentation process to produce high-quality silage (**Table 1**) [4].

there is no forage production enough to feed the animals [1].

**2. Importance of the fermentation for silage making**

mycotoxins in the fresh crop may be denatured due to the acid pH of silage.

an adequate moisture content.

4 Fermentation Processes

on animal performance and health.

**Figure 2.** Effects of dry matter content and water-soluble carbohydrates: buffering capacity on silage quality [5].

High-quality silages are resulted of a fast and efficient fermentation preserving the crop nutrients, which depends if the fresh crop has high nutritional value and good characteristics for the ensiling process, as described before. In addition, the fermentation process cannot improve the crop nutritive value, but in some cases occur an increase on digestibility, always with energy losses. Efficient fermentation ensures a more palatable and digestible feed, which improves the animal performance. As noted above, the most important factors related to the crop characteristics to ensiling are adequate dry matter content, sufficient water-soluble carbohydrates for fermentation, and low buffering capacity.

The dry matter content affects directly the microbial activity, specific density, and effluent losses. Crops with dry matter content below 25% at ensiling show high effluent losses and high activity of undesirable microorganisms such as the genus *Clostridium* [6]. In addition, the LAB are more tolerant to low moisture conditions (low water activity) than other undesirable anaerobic microorganisms. However, dry matter content above 45% difficult the process of forage packing, resulting in high porosity, which may cause losses by the development of aerobic microorganisms [7].


**Table 1.** Typical microbial populations on crops before ensiling [4].

About the amount of water-soluble carbohydrates, they present a narrow range of optimum values (60–80 g/kg of dry matter), because they are readily available substrates for the LAB and other microorganisms [6]. Furthermore, the excess sugar can stimulate the growth of anaerobic yeasts that are not fully inhibited by the low pH, as occurs in sugarcane silage, which results in high DM losses because the fermentation goes to the ethanol pathway [8].

The silage resistance to the pH lowering is named buffering capacity. This is exerted by compounds present in the crop, as the crude protein, inorganic ions, organic acids, and others. The greater buffering capacity needs more water-soluble carbohydrates content for an effective fermentation by reducing pH and inhibiting undesirable fermentations [9].

The fermentation coefficient (FC) was developed to predict if the crop is suitable to ensiling or not, as follows [10]:

FC = DM (%) + 8 WSC/BC

where FC = fermentation coefficient, DM = dry matter content, WSC = water-soluble carbohydrates, and BC = buffer capacity.

The forage crops with FC < 35 can result in undesirable fermentations and high dry matter losses, requiring additive application to control silage fermentation. If the FC ≥ 35, sufficient fermentable substrates are available. However, in high DM, crops are used microbial inoculants to ensure the presence of osmotolerant LAB to dominate the fermentation process.

#### **3. Silage fermentation processes**

The ensiling process, didactically, is divided into four principal phases [4]:

**1.** Initial aerobic phase since the harvest to the oxygen exhaustion in the silo. This phase is characterized by crop respiration and activity of all obligate and facultative aerobic organisms such as molds, yeasts, and some bacteria until finish up all the oxygen (**Figure 1**). In addition, the plant enzymes such as proteases and carbohydrases remain active. This phase must be short, because the sugars are converted to CO2 and water with heat release, representing dry matter losses, increased Maillard products, and drops in the silage quality. This phase is also important because of CO2, hydrogen peroxide, and other compounds that are produced with antimicrobial effect.

forage packing, resulting in high porosity, which may cause losses by the development of

About the amount of water-soluble carbohydrates, they present a narrow range of optimum values (60–80 g/kg of dry matter), because they are readily available substrates for the LAB and other microorganisms [6]. Furthermore, the excess sugar can stimulate the growth of anaerobic yeasts that are not fully inhibited by the low pH, as occurs in sugarcane silage, which

The silage resistance to the pH lowering is named buffering capacity. This is exerted by compounds present in the crop, as the crude protein, inorganic ions, organic acids, and others. The greater buffering capacity needs more water-soluble carbohydrates content for an effective

The fermentation coefficient (FC) was developed to predict if the crop is suitable to ensiling or

where FC = fermentation coefficient, DM = dry matter content, WSC = water-soluble carbohy-

The forage crops with FC < 35 can result in undesirable fermentations and high dry matter losses, requiring additive application to control silage fermentation. If the FC ≥ 35, sufficient fermentable substrates are available. However, in high DM, crops are used microbial inoculants to ensure the presence of osmotolerant LAB to dominate the fermentation process.

results in high DM losses because the fermentation goes to the ethanol pathway [8].

fermentation by reducing pH and inhibiting undesirable fermentations [9].

The ensiling process, didactically, is divided into four principal phases [4]:

**Group Population, colony-forming units/g of fresh forage**

aerobic microorganisms [7].

6 Fermentation Processes

not, as follows [10]:

FC = DM (%) + 8 WSC/BC

drates, and BC = buffer capacity.

**3. Silage fermentation processes**

Total aerobic bacteria >10,000,000 Lactic acid bacteria 10–1,000,000 Enterobacteria 1000–1,000,000 Yeasts 1000–100,000 Molds 1000–10,000 Clostridia 100–1000 Bacilli 100–1000 Acetic acid bacteria 100–1000 Propionic acid bacteria 10–1000

**Table 1.** Typical microbial populations on crops before ensiling [4].


#### **3.1. Substrates**

The most important substrates for the fermentation are the water-soluble carbohydrates and various amino acids and vitamins of the crop. In addition, after chopping the enzymes, plants can hydrolyze starch and hemicelluloses providing more hexoses and pentoses to microbial growth. Hexose monosaccharides, oligosaccharides, and polysaccharides, such as glucose, fructose, sucrose, and fructans, are the main water-soluble carbohydrates readily available for fermentation. Other important carbohydrate is the starch, which is the main storage polysaccharide in some crops, but it is practically not used in the fermentation because it is insoluble in water [9].

#### **3.2. Types of fermentations**

In silage fermentation, several pathways occur simultaneously; the fermentation type depends on the environmental conditions, microorganism species, and substrate availability. The LAB show two basic types of hexose fermentation to lactic acid. The most efficient pathway in energy conservation is the obligate homofermentative, which produces almost exclusively lactic acid (>85%). The facultative heterofermentative lactic acid bacteria show besides the homolactic pathway; they present ability to ferment pentoses, because they have both enzymes aldolase and phosphoketolase. The obligate heterofermentative lactic acid bacteria present DM loss from hexose fermentation due the CO2 production as well as lactic acid, and acetic acid or ethanol [4]. The acetate or ethanol production depends on the fermentation substrate: if the fermentation substrate is a hexose, the end-product is acetic acid, and if it is a pentose, the endproduct is ethanol [6]. Although heterolactic pathway causes DM loss, a partial increase in acetic acid concentration improves the aerobic stability of silage, because the acetic acid inhibits the activity of yeasts during the feed-out phase [11]. The end-products of well-fermented silages are presented in **Table 2**.


**Table 2.** Amounts of common fermentation end products in various silages [12].

When the acidification is not fast and adequate and/or with high moisture content, undesirable secondary fermentations can occur by other microorganisms, which are able to compete for nutrients with the LAB. Enterobacterial fermentation pathway is similar to the heterofermentative LAB and ferments glucose to acetic acid, formic acid, and alcohol. In addition, enterobacteria can decarboxylate and deaminate amino acids and reduce NO3. Other undesirable is the clostridial fermentations, which derive their energy from organic compounds such as carbohydrates and proteins producing butyric acid, acetic acid, propionic acid, ethanol, biogenic amines, and CO2. Those processes represent major losses that decrease silage quality and increase the production cost because of the low DM recovery. In addition, other smaller fermentations like the *Propionibacterium* can ferment glucose, fructose, glycerol, lactate, lactose, sucrose, xylose, and starch producing propionic acid, acetic acid, CO2, and formic acid or isovaleric acid. The facultative anaerobic yeasts can ferment glucose, maltose, and sucrose with the main products such as ethanol, CO2, and others compounds (alcohols, volatile fatty acids and lactate). The facultative anaerobic bacilli can ferment carbohydrates to organic acids or ethanol, 2,3-butanediol, and glycerol [4].

The secondary fermentations are undesirable because they preserve less energy in its endproducts compared to the lactic acid fermentation, which is explained by the production of CO2. These fermentations can also produce toxic compounds that impair the animal health and performance.

#### **3.3. Efficiency of the fermentation process**

**3.1. Substrates**

8 Fermentation Processes

in water [9].

**3.2. Types of fermentations**

silages are presented in **Table 2**.

1

% of total nitrogen.

**Item, % of dry matter Silage and dry matter contents**

pH 4.3–4.5 4.7–5.0 4.3–4.7 3.7–4.2 4.0–4.5 Lactic acid 7.0–8.0 2.0–4.0 6.0–10.0 4.0–7.0 0.5–2.0 Acetic acid 2.0–3.0 0.5–2.0 1.0–3.0 1.0–3.0 <0.5 Propionic acid <0.5 <0.1 <0.1 <0.1 <0.1 Butyric acid <0.5 0 <0.5 0 0 Ethanol 0.5–1.0 0.5 0.5–1.0 1.0–3.0 0.2–2.0 Ammonia N1 10.0–15.0 <12 8.0–12.0 5.0–7.0 <10.0

**Table 2.** Amounts of common fermentation end products in various silages [12].

The most important substrates for the fermentation are the water-soluble carbohydrates and various amino acids and vitamins of the crop. In addition, after chopping the enzymes, plants can hydrolyze starch and hemicelluloses providing more hexoses and pentoses to microbial growth. Hexose monosaccharides, oligosaccharides, and polysaccharides, such as glucose, fructose, sucrose, and fructans, are the main water-soluble carbohydrates readily available for fermentation. Other important carbohydrate is the starch, which is the main storage polysaccharide in some crops, but it is practically not used in the fermentation because it is insoluble

In silage fermentation, several pathways occur simultaneously; the fermentation type depends on the environmental conditions, microorganism species, and substrate availability. The LAB show two basic types of hexose fermentation to lactic acid. The most efficient pathway in energy conservation is the obligate homofermentative, which produces almost exclusively lactic acid (>85%). The facultative heterofermentative lactic acid bacteria show besides the homolactic pathway; they present ability to ferment pentoses, because they have both enzymes aldolase and phosphoketolase. The obligate heterofermentative lactic acid bacteria present DM loss from hexose fermentation due the CO2 production as well as lactic acid, and acetic acid or ethanol [4]. The acetate or ethanol production depends on the fermentation substrate: if the fermentation substrate is a hexose, the end-product is acetic acid, and if it is a pentose, the endproduct is ethanol [6]. Although heterolactic pathway causes DM loss, a partial increase in acetic acid concentration improves the aerobic stability of silage, because the acetic acid inhibits the activity of yeasts during the feed-out phase [11]. The end-products of well-fermented

**Alfafa, 32.5% Alfafa, 50.0% Grass, 30.0% Corn, 37.5% High moisture corn, 75.0%**

When the acidification is not fast and adequate and/or with high moisture content, undesirable secondary fermentations can occur by other microorganisms, which are able to compete for nutrients with the LAB. Enterobacterial fermentation pathway is similar to the heterofermenThe prevalent fermentation pathways in the ensiling process depend on several factors. They are related to the fresh crop and are basically the contents of DM and water-soluble carbohydrates. In addition, there are some characteristics related to the process techniques such as particle size, specific density, and especially the length time until the installation of anaerobic conditions in the silo. According to [13], the homofermentative LAB pathway results in only 0.7% of energy loss and it can be described as follows:

Glucose or fructose + 2 ADP + 2 Pi = 2 lactate + 2 ATP + 2 H2O.

The heterofermentative LAB pathway from glucose results in 24% of DM loss and 1.7% of energy loss. When they ferment, fructose results in 4.8% of DM loss and 1.0% of energy loss, and it can be described as follows:

Glucose + ADP + Pi = lactate + ethanol + CO2 + ATP + H2O, or

Fructose + 2 ADP + 2 Pi = lactate + acetate + 2 mannitol + 2 CO2 + 2 ATP + H2O.

In the clostridial fermentations, DM loss is 51.1% and the energy loss is 18.4%, and it can be described as follows:

2 lactate + ADP + Pi = butyrate + 2 CO2 + 2 H2 + ATP + H2O.

In the yeasts' fermentation, the DM loss is 48% and the energy loss is 0.2%, and it can be described as follows:

Glucose + 2 ADP + 2 Pi = 2 ethanol + 2 CO2 + 2 ATP + 2 H2O.

#### **4. Manipulating silage fermentation**

The knowledge about silage fermentation provides technology improvement to produce highquality silages. In addition, crops that were once considered inappropriate to ensiling, mainly legumes, are routinely ensiled in many farms nowadays. Theoretically, all forage crops can be conserved as silage, if the ensiling techniques such as the finely chopped, well packed in the silo, and complete sealed through of plastic sheet are done carefully to promote adequate anaerobic condition. However, the crop intrinsic characteristics will direct the fermentation pathway and affect the final silage quality.

#### **4.1. Changing the harvest time**

Each crop, depending on environment, has the ideal stage of maturity for silage production considering the yield due to the profitability, dry matter, and fermentable sugar contents for bacteria and maximum nutritional value for livestock (**Table 3**). Practically, all factors involving the fermentation will change with crop maturity stage. In addition, the water-soluble carbohydrates have a diurnal fluctuation cycle, and their concentrations are highest at 18:00 h and lowest at 06:00 h. Generally, advancing crop maturity results in increases in dry matter, carbohydrates, and LAB population as well as total microorganism number. In addition, decreases in buffering capacity and crude protein concentration are observed, and some crops have showed a decrease in digestibility with advancing maturity [9].


**Table 3.** Harvest and dry matter recommendation for main crops conserved as silage [14].

#### **4.2. Wilting**

Some crops, like tropical grasses and some legume such as alfalfa and forage soybean (**Table 4**), have a quite low DM content at the same time when the nutritive value is high. Obtaining a good fermentation and eliminating the effluent losses must increase the dry matter content prior to chopping and ensiling. Generally, those crops need be wilted at harvest with a mowerconditioner to increase DM content and to enhance the lactic fermentation. Mowing and conditioning can increase the leaves losses and affect the microbial populations on the crop. The plant juice released can increase the nutrients losses and bacterial population and cause a shift in the microbial species present [15].


**Table 4.** Effect of crop vegetative stage and preharvest wilting time on ensiling parameters and *in vitro* rumen degradability of forage soybean silage [16].

The wilting before ensiling is more common in regions with dry weather or with well-defined seasons, because the rainfall during the wilting period may cause significative losses than the ensiling wet crop. During the wilting, the crop remains metabolically active, and the cell respiration and proteolysis cause losses, the most important factor is the time until reaching the desired DM. The fast dehydration decreases plant carbon losses and protein degradation. The respiration loss is unavoidable, and its intensity depends on the oxygen, DM, and watersoluble carbohydrate contents. Depending on environmental conditions, the crop containing high level of crude protein may have high proteolysis during wilting, which decreases the silage quality [15].

#### **4.3. Silage additives**

**4. Manipulating silage fermentation**

pathway and affect the final silage quality.

**4.1. Changing the harvest time**

10 Fermentation Processes

**4.2. Wilting**

shift in the microbial species present [15].

The knowledge about silage fermentation provides technology improvement to produce highquality silages. In addition, crops that were once considered inappropriate to ensiling, mainly legumes, are routinely ensiled in many farms nowadays. Theoretically, all forage crops can be conserved as silage, if the ensiling techniques such as the finely chopped, well packed in the silo, and complete sealed through of plastic sheet are done carefully to promote adequate anaerobic condition. However, the crop intrinsic characteristics will direct the fermentation

Each crop, depending on environment, has the ideal stage of maturity for silage production considering the yield due to the profitability, dry matter, and fermentable sugar contents for bacteria and maximum nutritional value for livestock (**Table 3**). Practically, all factors involving the fermentation will change with crop maturity stage. In addition, the water-soluble carbohydrates have a diurnal fluctuation cycle, and their concentrations are highest at 18:00 h and lowest at 06:00 h. Generally, advancing crop maturity results in increases in dry matter, carbohydrates, and LAB population as well as total microorganism number. In addition, decreases in buffering capacity and crude protein concentration are observed, and some crops

have showed a decrease in digestibility with advancing maturity [9].

**Table 3.** Harvest and dry matter recommendation for main crops conserved as silage [14].

**Crop Maturity Dry matter (%) Cut length (mm)**

Some crops, like tropical grasses and some legume such as alfalfa and forage soybean (**Table 4**), have a quite low DM content at the same time when the nutritive value is high. Obtaining a good fermentation and eliminating the effluent losses must increase the dry matter content prior to chopping and ensiling. Generally, those crops need be wilted at harvest with a mowerconditioner to increase DM content and to enhance the lactic fermentation. Mowing and conditioning can increase the leaves losses and affect the microbial populations on the crop. The plant juice released can increase the nutrients losses and bacterial population and cause a

Corn Milk line 1/2–2/3 down the kernel 28–37 9.5–12.7 Alfafa Mid-bud 1/10 bloom, wilt to 30–40 6.4–9.5 Cereal Milk or soft dough, wilt to 28–37 6.4–9.5 Grasses When the first stems head out 28–37 6.4–9.5 Clover 1/4–1/2 bloom, wilt to 28–37 6.4–9.5 Sorghum Grain medium to hard dough 30–35 9.5–12.7

> In specific cases, when all ensiling techniques and fermentation process are understood and managed properly, the use of additives is necessary to regulate the fermentation process and to obtain high-quality silages. Silage additives can be used to help fixing some historic problems of the crops (low LAB epiphytic, and low DM and soluble sugars contents), oversized silos, silage storage for prolonged time, or silage moved from silo to another structure [17]. In addition, the additives are used to reduce heating and DM losses improving the silage fermentation quality and profitability. Most commercial additives contain more than one active ingredient in order to enhance efficacy and broad range of applicability [10]. According to [18], the additives, basically, have five functions (**Table 5**). Once again, it is important to emphasize that the use of additives will never correct or fix failures from poor management of the silagemaking process.


**Table 5.** Silage additives [18].

#### *4.3.1. Fermentation stimulators*

The additives that promote the desirable lactic acid fermentation are called fermentation stimulators, by either providing additional fermentable sugars or increasing the LAB population in the ensiled crop.

Additives containing water-soluble carbohydrates will improve the fermentation in crops containing low sugars such as some legumes and tropical grasses. The use of molasses in the ensiling process was a practice widely used in the past to accelerate and increase the lactic acid fermentation. However, it was recommended to be used in relatively high concentrations (40– 50 g/kg of fresh matter) and crops containing low DM content showed increase in effluent losses. Due to the high cost and viscosity, which are difficult to apply the molasses, today it is not too used in the farms. Other products or by-products can also be used for the same purpose, but attention should be paid to the availability and cost [19].

Enzyme additives usually are active enzyme combination (cellulases, hemicellulases, and amylases) used to break down the crop fiber and starch to release water-soluble carbohydrates, which could be fermented by LAB. The best results are improvements in silage fermentation and decreases in fiber content. However, the enzymes require certain conditions for maximum activity such as the pH, temperature, surface area, dry matter content, and crop proteases may inhibit enzyme activity. In addition, their positive effects also depend on the LAB initial population, crop characteristics, and application rate. The most suitable role for enzymes may be in combination with microbial inoculants [17, 19].

Inoculants containing homofermentative LAB are used with the purpose of increasing the initial population of this bacteria ensuring efficient fermentation to produce lactic acid. In addition, the use of homofermentative inoculants may accelerate pH reduction because the lactic acid is a stronger acid (p*K*a 3.86) than acetic acid (p*K*a 4.76) [4]; improving the lactic acid:acetic acid ratio consequently reduces dry matter losses. Homofermentative inoculation would also limit degradation and deamination of crop proteins and reduce ammonia production, which increases silage quality [20]. It was observed by [21], when evaluating the effects of homofermentative inoculants in alfalfa silage; they observed that some of the evaluated inoculants, with faster growing and ability to dominate the epiphytic microflora, decreased the pH since the first day of fermentation (**Figure 3**).

**Figure 3.** The pH (a), ammonia nitrogen (b) and lactic acid of alfalfa silages as a function of microbial inoculant within each fermentation period. a–cMeans followed by different letters in bars are different according to the predicted difference (*P* < 0.05). CTRL = control (without inoculant); CI = commercial inoculant, Sil-All® 4 × 4 W.S. (Alltech, Sao Paulo, Brazil); S1 = *Pediococcus acidilactici*, Strain 10.6; S2 = *P. pentosaceus*, Strain 6.16 [21].

Microbial inoculants include one or more of these bacteria: *Lactobacillus plantarum, L. acidophi‐ lus, L. salivarius, Pediococcus acidilactici, P. pentacaceus, Enterococcus faecium*, and *Streptococcus bovis*. Some combinations are used in accordance with the LAB capacity and potential of synergistic actions. For example, the use of *Streptococcus*, which exhibit faster growth and simultaneous drop in pH, combined with *Pediococcus*, which are more tolerant to conditions of temperature, pH, and high dry matter content. However, *Lactobacillus plantarum* is the most common species used [17]. According to [22], the inoculant should be added at a rate that is at least 10% of the epiphytic population to fermentation improvement. For commercial inoculants, recommendation ranges from 1 × 105 to 1 × 106 colony-forming units (cfu)/g of fresh forage.

#### *4.3.2. Fermentation inhibitors*

**Functions Examples**

Aerobic deterioration

12 Fermentation Processes

**Table 5.** Silage additives [18].

*4.3.1. Fermentation stimulators*

lation in the ensiled crop.

inhibitors

Fermentation stimulators Homofermentative lactic acid bacteria

acid

Nutrients Urea, ammonia, biuret, and limestone

but attention should be paid to the availability and cost [19].

be in combination with microbial inoculants [17, 19].

sodium hydroxide

Glucose, sucrose, molasses, cereals, wheat, citrus pulp, and enzymes

Formaldehyde, sodium nitrite, sodium metabisulfite, sodium chloride, antibiotics, and

Fermentation inhibitors Formic acid, acetic acid, lactic acid, benzoic acid, acrylic acid, citric acid, and sorbic

Propionic acid, caproic acid, sorbic acid, and ammonia

The additives that promote the desirable lactic acid fermentation are called fermentation stimulators, by either providing additional fermentable sugars or increasing the LAB popu-

Additives containing water-soluble carbohydrates will improve the fermentation in crops containing low sugars such as some legumes and tropical grasses. The use of molasses in the ensiling process was a practice widely used in the past to accelerate and increase the lactic acid fermentation. However, it was recommended to be used in relatively high concentrations (40– 50 g/kg of fresh matter) and crops containing low DM content showed increase in effluent losses. Due to the high cost and viscosity, which are difficult to apply the molasses, today it is not too used in the farms. Other products or by-products can also be used for the same purpose,

Enzyme additives usually are active enzyme combination (cellulases, hemicellulases, and amylases) used to break down the crop fiber and starch to release water-soluble carbohydrates, which could be fermented by LAB. The best results are improvements in silage fermentation and decreases in fiber content. However, the enzymes require certain conditions for maximum activity such as the pH, temperature, surface area, dry matter content, and crop proteases may inhibit enzyme activity. In addition, their positive effects also depend on the LAB initial population, crop characteristics, and application rate. The most suitable role for enzymes may

Inoculants containing homofermentative LAB are used with the purpose of increasing the initial population of this bacteria ensuring efficient fermentation to produce lactic acid. In addition, the use of homofermentative inoculants may accelerate pH reduction because the lactic acid is a stronger acid (p*K*a 3.86) than acetic acid (p*K*a 4.76) [4]; improving the lactic acid:acetic acid ratio consequently reduces dry matter losses. Homofermentative inoculation would also limit degradation and deamination of crop proteins and reduce ammonia produc-

Heterofermentative lactic acid bacteria

Moisture absorbents Citrus pulp, ground corn, cassava meal, straw, and coffee hulls

These are all chemical additives that affect the undesirable fermentation and microorganism growth. Based on the same principle of food conservation, several substances are used for this purpose. However, the choice of a suitable additive depends on cost-efficiency and historical occurrence of silage with poor-quality fermentation. Generally, they are used in wet crops with low WSC content and/or high buffer capacity. In addition, in crops containing high WSC, the acid-tolerant yeast can proliferate and decrease the silage quality. Salts of acids have become the most popular fermentation inhibitors, because they are easier and safer to handle [10], and they are effective on controlling yeast growth [23].

#### *4.3.3. Inhibitors of aerobic deterioration*

During the feed-out phase, when opening the silo, the presence of oxygen allows the development of molds, yeasts, and aerobic bacteria that consume the silage nutrients. The length of time that silage remains cool and does not spoil after it is exposed to air is called of aerobic stability. There are chemical and biological additives that are used to improve the aerobic stability by inhibit aerobic spoilage, mainly yeasts and acetic acid bacteria, because these microorganisms are responsible to initiate the aerobic deterioration. Generally, the chemical additives are more expensive and difficult to handle than are biological, and successful treatment depends on application rate. However, the variation in the effects when chemical additives are used is lower than the biological additives. Chemical additives with strong antimycotic activity are sorbic and benzoic acid [19, 23]. Besides the use of chemical additives, there is the possibility of using of biological additives based on heterofermentative LAB, such as *Lactobacillus buchneri*, which anaerobically degrade lactic acid to acetic acid and 1,2 propanediol causing a yeast inhibition [10, 23]. Yeast inhibition by organic acids is due to the undissociated form in acid pH. The inhibition effectiveness depends on the dissociation constant (pk) of organic acid; the acids with the highest pk are more effective in inhibiting. The ascending order of pk is formic acid, lactic acid, acetic acid, and propionic acid (3.75, 3.86, 4.76, and 4.87, respectively) [4].

#### *4.3.4. Nutrients*

The quality of crop can be improved by supplementation of dietary components that are essential for ruminants through of specific additives at the time of ensiling. In addition, despite of the buffering effect, the urea and ammonia can improve the aerobic stability of silage and increase crude protein content [6]. Grains can be added to increase levels of metabolizable energy in the silage. In other cases, some minerals can be added in order to meet a possible deficiency of the crop to better animal performance [19].

#### *4.3.5. Moisture absorbents*

Good results have been obtained in crops with a low DM content (<25%) at the ensiling to prevent excessive effluent losses and clostridial fermentations. Some additives can also improve the nutritive value and final silage quality [6]. Grains can be added to increase moisture absorbent to reduce silage effluent losses [19].

#### **4.4. Using mixed crops**

It can be used with several goals always taking advantage of a potential synergistic effect from improvement of soil tillage and fertilization and increased nutritive value, and/or supply the dry matter content and water-soluble carbohydrates to ensure a high-quality silage. Mixing legumes with cereal crops has been to increase grain yields and crude protein of crops while improving soil fertility but can increase the buffering capacity, which can decrease the fermentation efficiency in drops in the pH [24].

#### **5. How does the fermentation process affect silage quality?**

the most popular fermentation inhibitors, because they are easier and safer to handle [10], and

During the feed-out phase, when opening the silo, the presence of oxygen allows the development of molds, yeasts, and aerobic bacteria that consume the silage nutrients. The length of time that silage remains cool and does not spoil after it is exposed to air is called of aerobic stability. There are chemical and biological additives that are used to improve the aerobic stability by inhibit aerobic spoilage, mainly yeasts and acetic acid bacteria, because these microorganisms are responsible to initiate the aerobic deterioration. Generally, the chemical additives are more expensive and difficult to handle than are biological, and successful treatment depends on application rate. However, the variation in the effects when chemical additives are used is lower than the biological additives. Chemical additives with strong antimycotic activity are sorbic and benzoic acid [19, 23]. Besides the use of chemical additives, there is the possibility of using of biological additives based on heterofermentative LAB, such as *Lactobacillus buchneri*, which anaerobically degrade lactic acid to acetic acid and 1,2 propanediol causing a yeast inhibition [10, 23]. Yeast inhibition by organic acids is due to the undissociated form in acid pH. The inhibition effectiveness depends on the dissociation constant (pk) of organic acid; the acids with the highest pk are more effective in inhibiting. The ascending order of pk is formic acid, lactic acid, acetic acid, and propionic acid (3.75, 3.86, 4.76,

The quality of crop can be improved by supplementation of dietary components that are essential for ruminants through of specific additives at the time of ensiling. In addition, despite of the buffering effect, the urea and ammonia can improve the aerobic stability of silage and increase crude protein content [6]. Grains can be added to increase levels of metabolizable energy in the silage. In other cases, some minerals can be added in order to meet a possible

Good results have been obtained in crops with a low DM content (<25%) at the ensiling to prevent excessive effluent losses and clostridial fermentations. Some additives can also improve the nutritive value and final silage quality [6]. Grains can be added to increase

It can be used with several goals always taking advantage of a potential synergistic effect from improvement of soil tillage and fertilization and increased nutritive value, and/or supply the dry matter content and water-soluble carbohydrates to ensure a high-quality silage. Mixing legumes with cereal crops has been to increase grain yields and crude protein of crops while

they are effective on controlling yeast growth [23].

deficiency of the crop to better animal performance [19].

moisture absorbent to reduce silage effluent losses [19].

*4.3.3. Inhibitors of aerobic deterioration*

14 Fermentation Processes

and 4.87, respectively) [4].

*4.3.5. Moisture absorbents*

**4.4. Using mixed crops**

*4.3.4. Nutrients*

All microorganisms present in the silo, crop epiphytic population, and possible contamination primarily consume energy of water-soluble carbohydrates and other compounds for their growth and proliferation. Theoretically, the homolactic fermentation recovers 99% of the energy from glucose. However, in the silage fermentation process, many pathways occur simultaneously with different extensions, beyond the initial cellular respiration and enzymes activity, which are decisive in the final silage quality. Reducing losses by effluent is also important because it contains cellular content with high nutritional value that can contaminate the environment [6, 25]. High-quality silage is the result of adoption of appropriate techniques, starting with soil preparation and fertilization. In addition, the crop must have high DM yield, adequate nutritional value, and good characteristics for fermentation at the ensiling. Actually, even if high-quality crops are harvested efficiently, significant losses in the quality can occur if the ensiling process is inadequate (**Table 6**).


**Table 6.** Dry matter losses in silage under good or poor management [26].

Forages should be harvested for silage making when they have high nutritional value and the DM content is between 30 and 35%. Therefore, the monitoring of dry matter content at harvest period is essential, because some crops are required to be wilted or ensiled with additives to reach the recommended DM content. The crop must be chopped to about 0.5–1.5 cm length so that the work of packing and taking out is carried out easily. The chopped forage must be well packed in the silo, so less air will be trapped inside the stack, and the peripheral area should have packed more intensely. Filling the silo as quick as possible (within 3 days) limits the forage exposure to air, but each night until it is filled, the stack should be covered. The last step is complete seal with plastic as soon as filling and compaction is completed. In addition, the plastic should be covered, usually with tires or soil to eliminate gases and to prevent damage of the plastic. The packing density at of a good silage should be about 650 kg of fresh silage per cubic meter.

#### **5.1. Chemical composition and nutritive value**

Changes are inevitable in chemical composition during the ensiling process; it is due the conversion of soluble carbohydrates into organic acids, as well as degradation of fiber and protein of fresh crop. First, changes in the composition start immediately after cutting, still in an aerobic environment. Early in this phase, enzymes break down fructans, starch, and hemicellulose, releasing simple sugars, and also degrade protein to peptides, amino acids, amides, and ammonia. In addition, during the respiration, soluble carbohydrates are converted to CO2 and water by releasing heat. If the respiration period is extended, it can increasing losses due the development of molds and yeasts. Also, the heat released by respiration may decrease the digestibility due to the Millard reaction. The heat binds amino acids to the hemicellulose increasing the indigestible fiber and undegradable protein [4].

During LAB fermentation, the soluble carbohydrates are converted to lactic acid, acetic acid, ethanol, CO2, and water, which represents slight losses of DM and energy. However, if there is a clostridial fermentation, which causes major problems in the silage quality, it converts the soluble carbohydrates and amino acids to organic acids, glycine, biogenic amines, ammonianitrogen, H2, and CO2. The fermentation length is important in the crop preservation. When the fermentation length is extensive, the losses and changes in nutritional value are greater [4].

Another major problem about the silage chemical composition is at the silo opening. With air exposure, the microorganisms, which were inhibited, can proliferate and consume the silage energy. Heating and spoilage during feed-out is one of the greatest contributors to DM losses. In addition, it can produce some substances, like mycotoxins, that may pose risks to animals fed with this silage [26].

#### **5.2. Animal performance**

The feed intake is the key constraint limiting performance of ruminant animals fed diets containing forages. Regulation of feed intake in ruminants involves multiple mechanisms and complicated interactions between animal and feed characteristics. Evaluating factors that affect the silage intake of dairy cows, [27] concluded which silage intake can be predicted based on the silage digestibility, total acids, and DM content. Silage intake increased with increasing silage digestibility which was influenced by stage of maturity at harvest. The same authors showed that the total organic acids produced by silage fermentation process depress the silage intake, but it will depend on the proportion of the silage included in the diet. In addition, a positive association between DM content and silage intake, and DM content independently affects the silage fermentation and animal performance.

Feeding spoiled silage can be a big problem, because the deterioration decreases silage digestibility and intake in cattle. In addition, molds in spoiled silage can produce mycotoxins that cause serious health problems in the animals and farmers [26]. Silage additive is one of the ways to try to ensure efficient fermentation and thus obtain high-quality silage. When studies from North America evaluating the effects of silage additives on animal responses were summarized, [28] showed that although not replace good techniques of the ensiling process, the microbial inoculation can improve the silage quality and animal performance. This activity in animal performance is not well understood and might inhibit detrimental microorganisms in both silage and rumen to enhance the animal health and performance [29].

#### **5.3. Animal health**

**5.1. Chemical composition and nutritive value**

16 Fermentation Processes

fed with this silage [26].

**5.2. Animal performance**

increasing the indigestible fiber and undegradable protein [4].

affects the silage fermentation and animal performance.

Changes are inevitable in chemical composition during the ensiling process; it is due the conversion of soluble carbohydrates into organic acids, as well as degradation of fiber and protein of fresh crop. First, changes in the composition start immediately after cutting, still in an aerobic environment. Early in this phase, enzymes break down fructans, starch, and hemicellulose, releasing simple sugars, and also degrade protein to peptides, amino acids, amides, and ammonia. In addition, during the respiration, soluble carbohydrates are converted to CO2 and water by releasing heat. If the respiration period is extended, it can increasing losses due the development of molds and yeasts. Also, the heat released by respiration may decrease the digestibility due to the Millard reaction. The heat binds amino acids to the hemicellulose

During LAB fermentation, the soluble carbohydrates are converted to lactic acid, acetic acid, ethanol, CO2, and water, which represents slight losses of DM and energy. However, if there is a clostridial fermentation, which causes major problems in the silage quality, it converts the soluble carbohydrates and amino acids to organic acids, glycine, biogenic amines, ammonianitrogen, H2, and CO2. The fermentation length is important in the crop preservation. When the fermentation length is extensive, the losses and changes in nutritional value are greater [4].

Another major problem about the silage chemical composition is at the silo opening. With air exposure, the microorganisms, which were inhibited, can proliferate and consume the silage energy. Heating and spoilage during feed-out is one of the greatest contributors to DM losses. In addition, it can produce some substances, like mycotoxins, that may pose risks to animals

The feed intake is the key constraint limiting performance of ruminant animals fed diets containing forages. Regulation of feed intake in ruminants involves multiple mechanisms and complicated interactions between animal and feed characteristics. Evaluating factors that affect the silage intake of dairy cows, [27] concluded which silage intake can be predicted based on the silage digestibility, total acids, and DM content. Silage intake increased with increasing silage digestibility which was influenced by stage of maturity at harvest. The same authors showed that the total organic acids produced by silage fermentation process depress the silage intake, but it will depend on the proportion of the silage included in the diet. In addition, a positive association between DM content and silage intake, and DM content independently

Feeding spoiled silage can be a big problem, because the deterioration decreases silage digestibility and intake in cattle. In addition, molds in spoiled silage can produce mycotoxins that cause serious health problems in the animals and farmers [26]. Silage additive is one of the ways to try to ensure efficient fermentation and thus obtain high-quality silage. When studies from North America evaluating the effects of silage additives on animal responses were summarized, [28] showed that although not replace good techniques of the ensiling process, the microbial inoculation can improve the silage quality and animal performance. This activity The microorganisms in the microbial epiphytic population are usually nonpathogenic. However, the contamination, especially with the soil, may increase the presence of enterobacteria and spores of clostridium and bacillus in the silage. Therefore, in some cases, the silage can be a contamination source of animal products, such as meat, milk, and cheese, besides affecting the animal health [4]. During the silage fermentation process, a succession of microorganisms and denaturation and production of several compounds occur. However, the main problem is the occurrence of undesirable fermentations, which reduces the nutritive value of silage. Furthermore, the presence of some microorganisms or compounds produced may be a risk to the animal health [26].

Enterobacteria present in the crop may have a small positive effect on the hygienic quality of silage because during the first stage of ensiling, they reduce the nitrate (NO3 − ) to intermediates as nitrite and nitric oxide which inhibit clostridial fermentations. However, enterobacteria are undesirable because they have an endotoxin, which can reduce the silage intake and increase cases of mastitis, besides the less effective fermentation than LAB [4].

The anaerobic environment into the silo is essential for high-quality silage and inhibition of molds that produce mycotoxins. Generally, the mycotoxins in silage are related to molds with high tolerance to CO2 concentrations. Feeding spoiled silage results in reduced intake, increased abortions, hormonal imbalances, and suppressed immune function. In addition, good ensiling conditions reduce the most of the population of potential pathogens such as *Listeria monocytogenes, Escherichia coli*, and several *Salmonella* species, because they are strongly inhibited by acid pH (<4.5). Actually, the biggest problems are caused by clostridia and bacilli due the ability to form endospore and their presence later in food production systems, requiring special treatment for their elimination [4].

#### **6. Future trends**

With the knowledge of the silage process, some techniques are being developed to improve the efficiency of the preservation and production of high-quality silage. The development of monitors for the DM content of the crop at harvest will help farmers to know the crop quality and ensiling characteristics to choice of additives when needed at accurate rates. In addition, the development of specific additives for each culture that are used in the world with ample effect, since the silage fermentation until the animal performance. Furthermore, today many researches are aimed to developing plastic films more resistant and impermeable to oxygen. Through improvements in the plant, breeding is possible to obtain suitable crops for the most different environmental conditions with high quality and productivity, besides the suitable characteristics for silage production and animal performance.

#### **7. Conclusions**

Despite being a well-known technique, it is not easy to produce high-quality silage. Starting with the crop containing high nutritional value that usually is expensive and requires much care. In addition, the ensiling process needs specific machinery, physical structure (silos), and plastic sheets for the coverage. Moreover, the farmers cannot afford the risk of losing the entire crop with a poorly made silage.

Some crops at better nutritional value also have good ensiling characteristics such as corn and sorghum. However, to ensure a high-quality silage is often necessary to use techniques such as crop wilting and application of additives, which can become the process more expensive. A quick and efficient fermentation in reducing the pH is the most desired at ensiling. It depends on the anaerobic environment, water activity, and substrate for LAB fermentation. The homofermentative LAB are the most efficient in preserving the crop characteristics. However, some heterofermentative LAB are also desirable because of its effect on the aerobic stability of silage.

#### **Author details**

Thiago Carvalho da Silva1\*, Leandro Diego da Silva2 , Edson Mauro Santos3 , Juliana Silva Oliveira3 and Alexandre Fernandes Perazzo3

\*Address all correspondence to: timao22@hotmail.com


3 Department of Animal Science, Federal University of Paraiba, Paraiba, Brazil

#### **References**


[3] Adesogan, A.T. Challenges of tropical silage production. In Proc. 15th International Silage Conference. Madison, WI: University of Wisconsin, 2009, pp. 139–154.

**7. Conclusions**

18 Fermentation Processes

silage.

**Author details**

Juliana Silva Oliveira3

**References**

Orange, 2004, pp. 1–24.

Thiago Carvalho da Silva1\*, Leandro Diego da Silva2

\*Address all correspondence to: timao22@hotmail.com

and Alexandre Fernandes Perazzo3

1 Department of Animal Science, Federal University of Goias, Goias, Brazil

2 Department of Animal Science, Federal University of Vicosa, Vicosa, Brazil

3 Department of Animal Science, Federal University of Paraiba, Paraiba, Brazil

[1] Doonan, B.M., Kaiser, A.G., Stanley, D.F., Blackwood, I.F., Piltz, J.W., and White, A.K. Silage in the farming system. In: A.G. Kaiser, J.W. Piltz, H.M. Burns, N.W. Griffiths (Eds.), Successful Silage, Chapter 1. New South Wales Dept. of Primary Industry:

[2] Chiba, S., Chiba, H., and Yagi, M. A guide for silage making and utilization in the tropical regions. A publication of the Japanese Livestock Technology Association.

Tokyo: Ministry of Agriculture, Forestry and Fisheries, 2005, pp. 29.

crop with a poorly made silage.

Despite being a well-known technique, it is not easy to produce high-quality silage. Starting with the crop containing high nutritional value that usually is expensive and requires much care. In addition, the ensiling process needs specific machinery, physical structure (silos), and plastic sheets for the coverage. Moreover, the farmers cannot afford the risk of losing the entire

Some crops at better nutritional value also have good ensiling characteristics such as corn and sorghum. However, to ensure a high-quality silage is often necessary to use techniques such as crop wilting and application of additives, which can become the process more expensive. A quick and efficient fermentation in reducing the pH is the most desired at ensiling. It depends on the anaerobic environment, water activity, and substrate for LAB fermentation. The homofermentative LAB are the most efficient in preserving the crop characteristics. However, some heterofermentative LAB are also desirable because of its effect on the aerobic stability of

, Edson Mauro Santos3

,


degradation of forage soybean silage. Animal Feed Science and Technology. v. 200, pp. 102–106, 2015.


[28] Kung Jr., L. and Muck, R.E. Animal Response to Silage Additives. Proceedings from the Silage: Field to Feedbunk North American Conference, Hershey, 1997, NRAES-99, pp. 200–210.

degradation of forage soybean silage. Animal Feed Science and Technology. v. 200, pp.

[17] Kung Jr, L., Stokes, M.R., Lin, C.J. Silage additives. In: D.R. Buxton, R.E. Muck, J.H. Harrison (Eds.), Silage Science and Technology. 1st ed. Madison, WI: American Society

[18] McDonald, P. The biochemistry of silage. Chichester: John Wiley & Sons, 1981. 218 p. [19] Kaiser, A.G. Silage additives. In: A.G. Kaiser, J.W. Piltz, H.M. Burns, N.W. Griffiths (Eds.), Successful Silage, Chapter 7. New South Wales Dept. of Primary Industry:

[20] Muck, R.E., and Kung, Jr, L. Effect of silage additives on ensiling. Proceedings of the conference on Silage: Field to feedbunk. North American Conference Hershey, PA, 1997.

[21] Silva, V.P., Pereira, O.G., Leandro, E.S., Da Silva, T.C., Ribeiro, K.G., Mantovani, H.C., and Santos, S.A. Effects of lactic acid bacteria with bacteriocinogenic potential on the fermentation profile and chemical composition of alfalfa silage in tropical conditions.

[22] Muck, R.E. Silage microbiology and its control through additives. Revista Brasileira de

[23] Da Silva, T.C., Pereira O.G., Kung Jr, L., Silva, L.D., Paula, R.A., Martins, R.M., Silva, V.P., and Ribeiro, K.G. Effect of a chemical additive containing sodium benzoate, potassium sorbate, and sodium nitrite on the microbial populations and aerobic stability of sugar cane. Proceedings of the XVII International Silage Conference,

[24] Titterton, M., and Bareeba, F.B. Grass and legume silages in the tropics. In: L. t'Mannetje (Ed.), Silage Making in the Tropics with Particular Emphasis on Smallholders. Proc. FAO Electronic Conference on Tropical Silage. Plant Production and Protection Paper 161, Rome, Italy: Food and Agricultural Organization of the United Nations, 2000. [25] Rotz, C.A. and Muck, R.E. Changes in Forage Quality during Harvest and Storage. In: G.C. Fahey, Jr. et al (Eds.), Forage Quality, Evaluation, and Utilization, Madison, WI:

[26] Adesogan, A.T. Improving forage preservation with additives. Covington, KY: Annual

[27] Huhtanen, P., Rinne, M., and Nousiainen, J. Evaluation of the factors affecting silage intake of dairy cows: a revision of the relative silage dry-matter intake index. Animal,

Journal Dairy Science. v. 99, n. 3, pp. 1895–1902, 2016.

American Society of Agronomy, 1994. pp. 828–868.

Meeting of the American Forage and Grassland Council, 2013.

Zootecnia, Viçosa, v. 39, pp. 183–191, 2010.

Piracicaba, Brazil, 2015, pp. 148–149.

v. 1, pp. 758–770, 2007.

102–106, 2015.

20 Fermentation Processes

NRAES-99.

of Agronomy, 2003. pp. 305–360.

Orange, 2004, pp. 172–196.

[29] Gollop, N., Zakin, V. and Weinberg, Z.G. Antibacterial activity of lactic acid bacteria included in inoculants for silage and in silages treated with these inoculants. Journal of Applied Microbiology, v. 98, pp. 662–666, 2005.

#### **Fermentation and Redox Potential Fermentation and Redox Potential**

Chen-Guang Liu, Jin-Cheng Qin and Yen-Han Lin Chen-Guang Liu, Jin-Cheng Qin and Yen-Han Lin

Additional information is available at the end of the chapter Additional information is available at the end of the chapter

http://dx.doi.org/10.5772/64640

#### **Abstract**

Redox potential, known as oxidation–reduction or oxidoreduction potential (ORP), not only indicates the reduction and oxidation capacity of the environment but also reflects the metabolic activity of microorganisms. Redox potential can be monitored online and controlled in time for more efficient fermentation operation. This chapter reviews the enzymes that modulate intracellular redox potential, the genetically engineered strains that harbor specific redox potential–regulated genes, the approaches that were used to manipulate and control redox potential toward the production of desired metabolites, the role of redox potential in metabolic pathway, and the impact of redox potential on microbial physiology and metabolism. The application of redox potential–controlled ethanol fermentation and the development of three redox potential–controlled fermentation processes are illustrated. In the end, the future perspective of redox potential control is provided.

**Keywords:** redox potential, ORP, fermentation, bioprocess, ethanol

#### **1. Introduction**

The fermentation industry has a long history since human ancestor occasionally produced alcohol, yogurt, and pickled food. Most of these fermentation products are related to the pathways of glycolysis and TCA cycle, which required microaerobic or anaerobic conditions to avoid the desired products being oxidized by oxygen.

Precisely controlling microaerobic or anaerobic states is a challenge when using a general dissolved oxygen electrode because of the detection limit of the probe. Therefore, the meas‐ urement of redox potential (aka oxidoreduction potential, ORP) is considered as an ideal alternative approach because of its rapid response and high sensitivity to oxidation reaction.

© 2017 The Author(s). Licensee InTech. This chapter is distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/3.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. © 2017 The Author(s). Licensee InTech. This chapter is distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/3.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

What's more, redox potential also correlates to metabolic network, involving the genes, proteins, and metabolites. Since maintaining intracellular redox potential balance is a basic demand of cells, either intracellular or intercellular redox potential control could be the effective methods to redistribute metabolic flux toward targeted products. This idea has been applied to make a broad range of fermented products.

In this chapter, the basic principle of redox potential and its intracellular influence on genes, proteins, and metabolites are reviewed. Furthermore, redox potential control by metabolic modification and process engineering on the various metabolite fermentations are illustrated, specifically for ethanol production as an example.

#### **2. Basic theory of redox potential**

Chemically, the oxidation–reduction potential (aka ORP or redox potential) is defined as the tendency for a molecule to acquire electrons. It involves two components known as redox pair during the electron transfer process, of which the oxidizing one (Ox) attracts electrons and then becomes the reducing one (Red). This relationship is illustrated below:

$$\text{Ox} + \text{ne}^- = \text{Red} \text{ l} \tag{1}$$

Electrons are exchanged during a redox reaction, in which a pair of oxidation reaction and reduction reaction must be involved. As an illustration, when oxidizing iodide by ferric iron to form iodine, the iodine ion loses two electrons to from iodine (known as oxidation), concurrently ferric ion receives the same amount electrons to form ferrous ion (known as reduction). As a result, a complete redox reaction is established.

$$\text{Oxidation:}\ 2\text{I}^- = \text{I}\_2 + 2\text{e}^-$$

Reduction: 2Fe3+ + 2e− = 2Fe2+

Redox reaction: 2Fe3+ + 2I− = 2Fe2+ + I2

In an aqueous system, the redox potential is related to the capacity of releasing or accepting electrons from all redox reactions. Similar to pH where it indicates the availability of hydrogen ions, the overall redox potential portrays a relative state of gaining or losing electrons. However, the net changes of redox potential are caused by all oxidizing and reducing agents in the aqueous system, not just alkalis and acids that determine pH values.

In 1889, Walter Hermann Nernst (1864–1941; Nobel Prize: 1920) developed an equation to interpret the theory of galvanic cells by taking the changes of Gibbs free energy (ΔG) and the mass ratio into account. The Gibbs free energy is a thermodynamic potential, a reduction of G is a necessary condition for the spontaneity of processes at constant pressure and temperature. The chemical reaction can occur only if the ΔG is negative.

$$\mathbf{E}\_{\mathrm{h}} = \mathbf{E}^{0} + \frac{\mathbf{R}\mathbf{T}}{\mathbf{n}\mathbf{F}} \ln \frac{\left[\mathbf{Ox}\right]}{\left[\text{Red}\right]} \mathbf{2} \tag{2}$$

$$
\Delta \mathbf{G} = -\mathbf{n} \mathbf{F} \Delta \mathbf{E}\_{\mathbf{h}} \mathbf{\mathcal{J}} \tag{3}
$$

E0 is the standard redox potential of a system obtained at standard state. Every chemical pair has its own intrinsic redox potential. The greater affinity for electrons, the higher standard redox potential could be. Generally, NAD+ /NADH, NADP+ /NADPH, GSSG/2GSH, ubiquinone (ox/red), and oxygen/water are some of the most common chemical pairs in cells, whose E0 were −320, −315, −240, +100, and +820 mV, respectively.

R is the universal gas constant; T is the absolute temperature; F, Faraday constant (96,485 C/ mol), is the number of coulombs per mole of electrons, and n is the number of transferred electrons. The equation implies the concentration of species and temperature plays the key roles for redox potential change.

For instance, the reaction of NADH oxidized by oxygen is the final step of electron transport chain during aerobic respiration in mitochondrion. Usually, the reaction involves two redox pairs, just like oxygen/water (+820 mV) and NAD+ /NADH (320 mV), thus ΔEh = 820 mV – (−320 mV) = 1140 mV, ΔG = −125 kJ/mol, which indicates that this process occurs spontaneously due to the negative value of ΔG.

Although the Nernst equation has been broadly used in biological systems because of the involvement of electron transfer chain, one fact should be noticed that the redox potential measured by a platinum electrode is not a thermodynamically calculated value. It measures the redox state in an aqueous system as voltages. Although a living biosystem centers on cell growth and metabolism, it is an open system where the intracellular equilibrium state is not always established. Nevertheless, the significance of redox potential on functioning biological systems was predicted nearly one century ago by two prestigious British scientists at the University of Cambridge [1]. Many scientists, since then, have successfully explored various correlations between extracellular redox potential measured by an electrode and intracellular biological properties.

#### **3. Extracellular redox potential**

What's more, redox potential also correlates to metabolic network, involving the genes, proteins, and metabolites. Since maintaining intracellular redox potential balance is a basic demand of cells, either intracellular or intercellular redox potential control could be the effective methods to redistribute metabolic flux toward targeted products. This idea has been

In this chapter, the basic principle of redox potential and its intracellular influence on genes, proteins, and metabolites are reviewed. Furthermore, redox potential control by metabolic modification and process engineering on the various metabolite fermentations are illustrated,

Chemically, the oxidation–reduction potential (aka ORP or redox potential) is defined as the tendency for a molecule to acquire electrons. It involves two components known as redox pair during the electron transfer process, of which the oxidizing one (Ox) attracts electrons and

Electrons are exchanged during a redox reaction, in which a pair of oxidation reaction and reduction reaction must be involved. As an illustration, when oxidizing iodide by ferric iron to form iodine, the iodine ion loses two electrons to from iodine (known as oxidation), concurrently ferric ion receives the same amount electrons to form ferrous ion (known as

In an aqueous system, the redox potential is related to the capacity of releasing or accepting electrons from all redox reactions. Similar to pH where it indicates the availability of hydrogen ions, the overall redox potential portrays a relative state of gaining or losing electrons. However, the net changes of redox potential are caused by all oxidizing and reducing agents

In 1889, Walter Hermann Nernst (1864–1941; Nobel Prize: 1920) developed an equation to interpret the theory of galvanic cells by taking the changes of Gibbs free energy (ΔG) and the mass ratio into account. The Gibbs free energy is a thermodynamic potential, a reduction of G is a necessary condition for the spontaneity of processes at constant pressure and temperature.

Ox ne Red 1 - + = (1)

then becomes the reducing one (Red). This relationship is illustrated below:

reduction). As a result, a complete redox reaction is established.

= 2Fe2+ + I2

The chemical reaction can occur only if the ΔG is negative.

in the aqueous system, not just alkalis and acids that determine pH values.

applied to make a broad range of fermented products.

specifically for ethanol production as an example.

**2. Basic theory of redox potential**

Oxidation: 2I−

24 Fermentation Processes

Reduction: 2Fe3+ + 2e−

Redox reaction: 2Fe3+ + 2I−

= I2 + 2e−

= 2Fe2+

The extracellular redox potential is different from intracellular redox state due to cytomem‐ brane separation and cell redox homeostasis. Environmental factors are critical to indirectly shift the cellular redox potential. Based on Nernst Equation, the redox potential is simply determined by the ratio of oxidative state to reductive state at a fixed temperature, which is always a constant parameter in most biological processes. **Figure 1** illustrates three general approaches to control extracellular redox potential in biological devices.

**Figure 1.** Approaches to control extracellular redox potential. (A) energy input, (B) redox reagents, and (C) gas sparging.

#### **3.1. Control extracellular redox potential by energy input**

Bioelectrical reactors (BERs), equipped with anodic and cathodic electrodes, were developed to regulate extracellular redox state in the medium through an external power source. It was used to replace chemical electron donor and acceptor in biosystem. BERs control redox potential at a certain level as easy as tuning a radio. It has been applied to microorganism cultivation and metabolites production [2]. Nevertheless, BERs have been implemented in a laboratory setting or for the production of high‐value products in order to compensate for its complicated equipment requirement and extra electrical energy consumption.

#### **3.2. Control extracellular redox potential by redox reagents**

Numerous chemicals with higher or lower standard redox potential than common metabolic components are supplemented into fermentation broth in order to alter environmental redox potential. Some commonly used reductants and oxidants to control extracellular redox potential include FeCl3, Na2S, potassium ferricyanide, dithiothreitol, cysteine, methyl violo‐ gen, neutral red, H2O2, and even directly NADH and NAD+ as additives. Unlike BERs requiring the design of a specific reactor, supplementing redox reagents can be employed in any type of bioreactor. However, the disadvantages are obvious: (a) extra chemicals added in media potentially interfere with intended bioprocessing and (b) some chemicals are too costly for industrial fermentation.

Those problems could be solved using substrates with different reducing degree. Girbal and Soucaille [3] used mixed substrates (glucose, glycerol, and pyruvate) to interfere with the intracellular NADH/NAD+ ratio in *Clostridium acetobutylicum*. Snoep et al. [4] chose some energy source substrates, such as mannitol, glucose, and pyruvate, to govern cellular redox potential in *Enterococcus faecalis*.

#### **3.3. Control extracellular redox potential by gas sparging**

Oxygen and nitrogen are commonly used in aerobic and anaerobic fermentation, respectively. Thus, sparging pure or mixed gases into fermentation broth is one of the desired approaches to avoid unwanted reactions caused by redox salts. Generally speaking, oxygen elevates redox potential and hydrogen depresses it, whereas nitrogen and helium as inert gases remove dissolved oxygen or hydrogen from the medium. Furthermore, by adjusting the ratio of mixed gases, a different redox potential level can be maintained. Carbon monoxide and SO2 were also utilized to reduce the redox potential sometimes [5]. However, aerating a fermenter during fermentation is considered cost-effective only when air is used. As a mix of nitrogen, hydrogen and helium were applied to regulate redox potential in the above settings, these methods become too luxurious for industrial applications.

#### **3.4. Extracellular redox potential and dissolved oxygen**

**Figure 1.** Approaches to control extracellular redox potential. (A) energy input, (B) redox reagents, and (C) gas

complicated equipment requirement and extra electrical energy consumption.

Bioelectrical reactors (BERs), equipped with anodic and cathodic electrodes, were developed to regulate extracellular redox state in the medium through an external power source. It was used to replace chemical electron donor and acceptor in biosystem. BERs control redox potential at a certain level as easy as tuning a radio. It has been applied to microorganism cultivation and metabolites production [2]. Nevertheless, BERs have been implemented in a laboratory setting or for the production of high‐value products in order to compensate for its

Numerous chemicals with higher or lower standard redox potential than common metabolic components are supplemented into fermentation broth in order to alter environmental redox potential. Some commonly used reductants and oxidants to control extracellular redox potential include FeCl3, Na2S, potassium ferricyanide, dithiothreitol, cysteine, methyl violo‐

the design of a specific reactor, supplementing redox reagents can be employed in any type of bioreactor. However, the disadvantages are obvious: (a) extra chemicals added in media potentially interfere with intended bioprocessing and (b) some chemicals are too costly for

Those problems could be solved using substrates with different reducing degree. Girbal and Soucaille [3] used mixed substrates (glucose, glycerol, and pyruvate) to interfere with the

energy source substrates, such as mannitol, glucose, and pyruvate, to govern cellular redox

Oxygen and nitrogen are commonly used in aerobic and anaerobic fermentation, respectively. Thus, sparging pure or mixed gases into fermentation broth is one of the desired approaches to avoid unwanted reactions caused by redox salts. Generally speaking, oxygen elevates redox potential and hydrogen depresses it, whereas nitrogen and helium as inert gases remove dissolved oxygen or hydrogen from the medium. Furthermore, by adjusting the ratio of mixed

ratio in *Clostridium acetobutylicum*. Snoep et al. [4] chose some

as additives. Unlike BERs requiring

**3.1. Control extracellular redox potential by energy input**

**3.2. Control extracellular redox potential by redox reagents**

gen, neutral red, H2O2, and even directly NADH and NAD+

**3.3. Control extracellular redox potential by gas sparging**

industrial fermentation.

intracellular NADH/NAD+

potential in *Enterococcus faecalis*.

sparging.

26 Fermentation Processes

Controlling the level of dissolved oxygen in a fermenter is essential for microorganisms to propagate under optimum physiological condition, not only because oxygen is involved in maintaining cell membrane integrity and function by synthesizing unsaturated fatty acid and sterol, but also for keeping metabolic flux channeling toward the production of desired products.

A number of bioreactions toward the syntheses of intended metabolites requires maintaining dissolved oxygen at a proper level. For most microaerobic and anaerobic fermentations, conventional oxygen probe has trouble in distinguishing trace level dissolved oxygen from background noise, and its response time is not sufficient for the purpose of regulating dissolved oxygen level. Even for aerobic fermentation, redox potential still offers much more details about gaseous conditions than that collected from dissolved oxygen measurement [6]. The standard redox potential for the O2/H2O pair has the highest value among typical metabolites related to microbial metabolism during fermentation. If electrons were transferred to acceptors, oxygen must be the preferable choice even though its concentration is lower than other metabolites. Therefore, redox potential is much more sensitive in monitoring the presence of a trace amount of dissolved oxygen under microaerobic and anaerobic conditions.

#### **4. Intracellular redox potential**

Currently, advanced technologies, such as a nanosensor that can embed into individual cells, have been developed to measure intracellular redox potential directly for in-depth understanding on intracellular redox balance and its impact on cell physiology and metabolism. However, the indirect approaches, such as the measurement of NAD(P)H pools, NAD(P)+ / NAD(P)H, GSH/GSSG, and the total oxidization power, are still commonly adopted to monitor the distribution of intracellular redox potential.

#### **4.1. Universal redox pairs in a cell**

A conjugate pair that constitutes a complete redox reaction is the fundamental of metabolic network in a cell. Many metabolic functions are realized through keeping intracellular redox balance with the main redox pairs, such as glutathione (GSH)/glutathione disulfide (GSSG), thioredoxin (TrxSS/Trx(SH)2), nicotinamide adenine dinucleotide (NAD), and nicotinamide adenine dinucleotide phosphatase (NADP). These redox systems, such as NADP+ /NADPH, GSSG/2GSH, and TrxSS/Trx(SH)2, are not isolated systems. Both the Trx and GSH systems use NADPH as a source of reducing equivalents; thus, they are thermodynamically connected to each other. The role of NAD(P)+ /NAD(P)H in redox reaction is illustrated in **Figure 2**.

**Figure 2.** The structure (A) and function (B) of NAD(P)H.

Both glutathione (GSH) and thioredoxin are important reducing agents in all organisms, involved in cell oxidative stress response where they play an antioxidant role. Glutathione is a tripeptide (glutamine, cysteine, and glycine) that prevents damage to cellular components caused by reactive oxygen species such as free radicals and peroxides, lipid peroxides, and heavy metals. Thioredoxin is another class of small redox proteins with thiol system in the cell, which appears in many crucial biological processes, including redox signaling.

The coenzymes are essential electron carriers in cellular redox reactions with the oxidized form NAD(P)+ and the reduced form NAD(P)H. The reduction reaction requires an input of energy and the oxidation reaction is exergonic. During carbohydrate metabolism, NADH plays as a notable reducing substance in catabolism, whereas NADPH, the other reducing component connected to anabolism, favors formation of amino acids, fatty acids, and nucleic acids. There are 129 enzymes that need NAD+ as cofactor in order to serve 931 redox reaction and 108 enzymes that require the involvement of NADP+ as cofactor in order to catalyze 1099 redox reaction (KEGG, 2016‐3).

#### **4.2. Redox effect across the membrane**

Cytosol is isolated from the extracellular environment by a selectively permeable cytomem‐ brane, which not only prevents the main redox pair escaping from the plasma freely but also conditionally allows the external redox chemicals to enter into the cytoplasm. As shown in **Figure 3**, chemicals with different reduction degrees, such as dithiothreitol (DDT), diamine, hydrogen peroxide, and oxygen, can unrestrictedly cross the membrane bilayer, causing the changes to the intracellular redox potential. However, most of these chemicals are prohibited to across the membrane. In another scenario, membrane proteins, such as ox‐ idoreductase, involved in electron transport will respond and change the extracellular redox potential. For example, ferric reductase assists ferrous iron transport across the cell mem‐ brane [7]. Hydrogenase facilitates electron flow through the membrane with the conversion of NADH and NAD+ [8]. A low redox potential level results in the changes of thiol and di‐ sulfide balance on membrane proteins, making the membrane more permeable to protons [9]. A thiol‐rich membrane protein transduces external GSH reducing power across the er‐ ythrocyte membrane, which can be explained as a thiol/disulfide exchange mechanism [10].

**Figure 3.** Intracellular redox response to extracellular redox potential and effects of redox potential on cellular metabo‐ lism and stress response.

#### **4.3. Effects of redox potential on a cell**

NADPH as a source of reducing equivalents; thus, they are thermodynamically connected to

Both glutathione (GSH) and thioredoxin are important reducing agents in all organisms, involved in cell oxidative stress response where they play an antioxidant role. Glutathione is a tripeptide (glutamine, cysteine, and glycine) that prevents damage to cellular components caused by reactive oxygen species such as free radicals and peroxides, lipid peroxides, and heavy metals. Thioredoxin is another class of small redox proteins with thiol system in the cell,

The coenzymes are essential electron carriers in cellular redox reactions with the oxidized form NAD(P)+ and the reduced form NAD(P)H. The reduction reaction requires an input of energy and the oxidation reaction is exergonic. During carbohydrate metabolism, NADH plays as a notable reducing substance in catabolism, whereas NADPH, the other reducing component connected to anabolism, favors formation of amino acids, fatty acids, and nucleic acids. There

Cytosol is isolated from the extracellular environment by a selectively permeable cytomem‐ brane, which not only prevents the main redox pair escaping from the plasma freely but also conditionally allows the external redox chemicals to enter into the cytoplasm. As shown in **Figure 3**, chemicals with different reduction degrees, such as dithiothreitol (DDT), diamine, hydrogen peroxide, and oxygen, can unrestrictedly cross the membrane bilayer, causing the changes to the intracellular redox potential. However, most of these chemicals are prohibited to across the membrane. In another scenario, membrane proteins, such as ox‐ idoreductase, involved in electron transport will respond and change the extracellular redox potential. For example, ferric reductase assists ferrous iron transport across the cell mem‐ brane [7]. Hydrogenase facilitates electron flow through the membrane with the conversion of NADH and NAD+ [8]. A low redox potential level results in the changes of thiol and di‐

as cofactor in order to serve 931 redox reaction and 108

as cofactor in order to catalyze 1099 redox

which appears in many crucial biological processes, including redox signaling.

/NAD(P)H in redox reaction is illustrated in **Figure 2**.

each other. The role of NAD(P)+

28 Fermentation Processes

**Figure 2.** The structure (A) and function (B) of NAD(P)H.

are 129 enzymes that need NAD+

**4.2. Redox effect across the membrane**

reaction (KEGG, 2016‐3).

enzymes that require the involvement of NADP+

The influences of redox potential on enzymes activity have also been reported. Almost all enzymes related to oxidation–reduction reaction are redox potential sensitive, such as alcohol dehydrogenase, D‐glyceraldehyde‐3‐phosphate dehydrogenase, quinone reductase (involved in quinone detoxification), NADH diphosphatase (involved in peroxisomal function), ubiqui‐ none oxidoreductase (catalyzing the oxidation of NADH in the respiratory chain or in cytoplasm), mitochondrial NADH kinase (response to oxidative stress), and so on. The above‐ mentioned proteins have been investigated in *Saccharomyces cerevisiae* in the past decades. Numerous proteins contain sulfhydryl groups (PSH) due to their cysteine content. In fact, the concentration of PSH groups in cells and tissues is much greater than that of GSH. These groups can be present as thiols (‐SH), disulfides (PS‐SP), or mixed disulfides; Hsp33 as a possible chaperone and cysteine protease in heat shock protein families is regulated by redox potential, whose conformation changes from reduced state to oxidized state with the exposure of hydrophobic surface [11]. Being a key regulator of glutathione and, in turn, of redox potential, the identification of GSTp as, a JNK regulator, provides an important link between cellular redox potential and the regulation of stress kinase activities [12].

Gene expression is controlled by redox states as well. It has been reported that overexpressing genes related to redox process in *Escherichia coli* resulted in the decrease of NADH/NAD+ ratio, which improve the cell growth profiles, because sufficient NAD+ is required to oxidize carbohydrate substrate during cell growth [13]. *GPD2* encodes NAD‐dependent glycerol 3‐ phosphate dehydrogenase, the key enzyme of glycerol synthesis, and is essential for cell survival under osmotic and low redox potential conditions. Unlike its homologous gene *GPD1* controlled by high osmolality glycerol response pathway, *GPD2* is regulated under anoxic conditions or, more accurately, oxygen‐independent reducing environment [14]. *YAP1*, a transcription factor for sensing the high redox state (e.g. H2O2), usually exists in the cytoplasm but is transferred into nucleus to activate the transcription of antioxidant genes *SOD1*, *TWF*, *TRX2*, *GLR1*, and *GSH1*, when Yap1p C‐terminal region with three conserved cysteine residues is oxidized in response to oxidative stress [15]. A redox sensing protein (RSP) binds transcrip‐ tional regulation regions located upstream from *adh*A, *adh*B, and *adh*E as a transcriptional repressor. The structure of RSP was changed from α‐helix to β‐sheet rich conformation when redox potential declined by adding NADH. Meanwhile, the repression of an alcohol dehy‐ drogenase transcription caused by RSP was reversed [16]. Thioredoxin reduces cysteine moieties in the DNA‐binding sites of several transcription factors and is therefore important in gene expression [17].

External redox potential correlates the net balance of intracellular reducing equivalents and the changes in the cellular redox environment can alter signal transduction, DNA and RNA synthesis, protein synthesis, enzyme activation, and even regulation of the cell cycle. Thus, monitoring and controlling environmental redox potential helps to elucidate cellular physi‐ ology and intracellular metabolic interaction.

#### **5. Redox potential and metabolic flux**

Strategies to control intracellular redox potential can be developed by altering intracellular redox potential pools, consequently resulting in redistribution of metabolic profiles. However, cells have a series of built‐in mechanisms to adjust their own intercellular redox balance by cofactor regeneration through the oxidoreductase‐harboring genes, including mitochondrial alternative oxidase (AOX), formate dehydrogenase (FDH), cytoplasmic H2O‐forming NADH oxidase (NOX), and mitochondrial NADH kinase (POS5). Therefore, modification of these genes is a promising strategy to "design" a robust strain subjected to redox regulation through extracellular manipulation, although such an alternation may result in unexpected outcomes.

#### **5.1. Alternative oxidase**

The alternative oxidase (AOX, EC: 1.10.3.11), also named ubiquinol oxidase, forms a part of the electron transport chain in mitochondria. The function of this oxidase is believed to dissipate excess reducing power. The reaction catalyzed by AOX oxidase (ubiquinol oxidase) is shown in Reaction (4).

$$\text{ubiquione} + \text{H}^+ + \text{NADH} = \text{ubiquino} \text{l} + \text{NAD}^+ \tag{4}$$

When a cell subjected to increasing glycolytic fluxes under aerobic conditions, a decrease in respiratory capacity is caused by the presence of excess glucose that repressed respiratory pathways. Introducing a heterologous alternative oxidase into *S*. *cerevisiae*, increased metabolic flux toward respiration and reduced aerobic ethanol formation [18]. In other investigation, the introduction of AOX pathway improved reactive oxygen species and pyruvate levels simul‐ taneously under stressful conditions, such as suboptimal temperature and hyperosmotic pressure [19].

#### **5.2. Formate dehydrogenase**

controlled by high osmolality glycerol response pathway, *GPD2* is regulated under anoxic conditions or, more accurately, oxygen‐independent reducing environment [14]. *YAP1*, a transcription factor for sensing the high redox state (e.g. H2O2), usually exists in the cytoplasm but is transferred into nucleus to activate the transcription of antioxidant genes *SOD1*, *TWF*, *TRX2*, *GLR1*, and *GSH1*, when Yap1p C‐terminal region with three conserved cysteine residues is oxidized in response to oxidative stress [15]. A redox sensing protein (RSP) binds transcrip‐ tional regulation regions located upstream from *adh*A, *adh*B, and *adh*E as a transcriptional repressor. The structure of RSP was changed from α‐helix to β‐sheet rich conformation when redox potential declined by adding NADH. Meanwhile, the repression of an alcohol dehy‐ drogenase transcription caused by RSP was reversed [16]. Thioredoxin reduces cysteine moieties in the DNA‐binding sites of several transcription factors and is therefore important

External redox potential correlates the net balance of intracellular reducing equivalents and the changes in the cellular redox environment can alter signal transduction, DNA and RNA synthesis, protein synthesis, enzyme activation, and even regulation of the cell cycle. Thus, monitoring and controlling environmental redox potential helps to elucidate cellular physi‐

Strategies to control intracellular redox potential can be developed by altering intracellular redox potential pools, consequently resulting in redistribution of metabolic profiles. However, cells have a series of built‐in mechanisms to adjust their own intercellular redox balance by cofactor regeneration through the oxidoreductase‐harboring genes, including mitochondrial alternative oxidase (AOX), formate dehydrogenase (FDH), cytoplasmic H2O‐forming NADH oxidase (NOX), and mitochondrial NADH kinase (POS5). Therefore, modification of these genes is a promising strategy to "design" a robust strain subjected to redox regulation through extracellular manipulation, although such an alternation may result in unexpected outcomes.

The alternative oxidase (AOX, EC: 1.10.3.11), also named ubiquinol oxidase, forms a part of the electron transport chain in mitochondria. The function of this oxidase is believed to dissipate excess reducing power. The reaction catalyzed by AOX oxidase (ubiquinol oxidase)

When a cell subjected to increasing glycolytic fluxes under aerobic conditions, a decrease in respiratory capacity is caused by the presence of excess glucose that repressed respiratory pathways. Introducing a heterologous alternative oxidase into *S*. *cerevisiae*, increased metabolic

+ + ubiquinone + H + NADH = ubiquinol + NAD (4)

in gene expression [17].

30 Fermentation Processes

**5.1. Alternative oxidase**

is shown in Reaction (4).

ology and intracellular metabolic interaction.

**5. Redox potential and metabolic flux**

Formate dehydrogenases (FDH, EC: 1.2.1.2) are a set of enzymes that catalyze the oxidation of formate to carbon dioxide (see Reaction 6), donating electrons to a second substrate, such as NAD+ or cytochrome. NAD+ ‐dependent formate dehydrogenases are important in methylo‐ trophic yeast and bacteria and are vital in the catabolism of C1 compounds, such as methanol.

$$\text{(formatate} + \text{NAD}^\* = \text{CO}\_2 + \text{NADH} + \text{H}^\* \tag{5}$$

As the FDH gene from *Candida boidinii* was introduced into *Paenibacillus polymyxa*, highly expressed exogenous FDH increased NADH/NAD+ and the titers of NADH‐dependent products such as lactic acid and ethanol, while resulting in significantly decreased acetoin and formic acid [20]. In addition, the increased capacity of a FDH gene in *Bacillus subtilis* efficiently enhanced the production of 2,3‐butanediol and decreased the formation of acetoin through increasing the availability of NADH [21]. In another case, an engineered strain for the conver‐ sion of D‐fructose to allitol was developed by constructing a multienzyme coupling pathway and cofactor recycling system in *E*. *coli*. FDH gene was used to support the cofactor recycling system for the availability of NADH [22].

#### **5.3. NADH oxidase**

NADH oxidase (NOX, EC: 1.6.3.4) is a membrane‐associated enzyme that catalyzes the production of superoxide, a reactive free radical, by transferring one electron from NADH to oxygen as the electron acceptor (see Reaction 7). It is considered one of the major sources of producing superoxide anions in humans as well as bacteria, subsequently used in oxygen‐ dependent killing mechanisms for invading pathogens.

$$2\text{H}^+ + 2\text{NADH} + \text{O}\_2 = 2\text{H}\_2\text{O} + 2\text{NAD}^+\tag{6}$$

Glycerol is a main by‐product in the 2,3‐butanediol metabolic pathways. To minimize glycerol accumulation by an engineered *S*. *cerevisiae*, the *Lactococcus lactis* NOX gene was inserted and expressed, resulting in substantial decreases in intracellular NADH/NAD+ ratio. As a result, the carbon flux was redistributed from glycerol to 2,3‐butanediol [23]. NADH oxidase was also expressed with l‐arabinitol dehydrogenase in *E*. *coli* to efficiently produce l‐xylulose. Thus, the efficiency above 96% for the conversion of l‐arabinitol into l‐xylulose was achieved under optimized conditions [24].

#### **5.4. NADH kinase**

NADH kinase (like POS5, EC: 2.7.1.86) catalyzes the replacement reaction with two substrates ATP and NADH and two products ADP and NADPH (see Reaction 8). It provides a key source of the important cellular antioxidant NADPH.

$$\text{ATP} + \text{NADH} \rightleftharpoons \text{ADP} + \text{NADHH} \tag{7}$$

NADPH is a key cofactor for carotenoid biosynthesis. *Corynebacterium glutamicum* was always used for the production of amino acids, such as L‐isoleucine. By implementing NADPH‐ supplying strategies based on NAD kinase (PpnK), NADH kinase, glucose‐6‐phosphate dehydrogenase (Zwf), and PpnK coupling with Zwf, the expression of all genes increased both the intracellular NADPH concentration and the L‐isoleucine production [25]. Researchers constructed the NADPH regenerators of heterologous NADH kinase to increase the availa‐ bility of NADPH and resulted in a superior S‐adenosylmethionine production in *E*. *coli* without requiring L‐methionine addition [26]. When a *S*. *cerevisiae* strain‐producing carotenoid was constructed by overexpressing glucose‐6‐phosphate dehydrogenase and NADH kinase individually, the final product β‐carotene yield increased by 18.8% and 65.6%, respectively. Thus, NADPH supply improved by overexpression of NADH kinase is more important than glucose‐6‐phosphate dehydrogenase [27].

#### **6. Application of redox potential to fermentation processes**

Controlling redox potential at a desired level alters the intracellular metabolic flow in order to favor the formation of desired product(s). Many researches have been conducted in this regard with a large number of examples for enhanced production of metabolites under redox potential–controlled conditions. Most studied metabolites using redox potential–controlled approaches are hydrogen, pyruvate, 1,3‐propanediol, butanol, and 2,3‐butanediol, and the following metabolites are reviewed but provided with references: acetoin [28], succinic acid [29], xylitol [30], and so on.

#### **6.1. Hydrogen**

Hydrogen, as a clean and high‐combustion energy in widespread areas, can be generated by fermentative anaerobes. Hydrogen production from anaerobic fermentation by bacteria demands reducing level because the standard redox potential of H2/H+ is low. Zhang et al. [8] showed that the addition of NAD+ during hydrogen fermentation by *Enterobacter aerogenes* resulted in the increase of overall hydrogen. Nakashimada et al. [31] investigated *E*. *aerogenes* for its hydrogen production under different intracellular redox state through the utilization of different substrates bearing various reduction degrees. Low redox potential accelerated the NAD(P)H‐dependent hydrogenase activity in membrane and favors high H2 evolution capability. Ren et al. [32] assessed H2 production during butyric acid fermentation, propionic acid fermentation, and ethanol fermentation by controlling redox potential and pH simulta‐ neously. Besides, the NAD+ synthetase encoded by nadE gene was homologously overex‐ pressed in *E*. *aerogenes* to decrease the NADH/NAD+ ratio and thus enhanced hydrogen yield [33].

#### **6.2. Pyruvate**

**5.4. NADH kinase**

32 Fermentation Processes

of the important cellular antioxidant NADPH.

glucose‐6‐phosphate dehydrogenase [27].

[29], xylitol [30], and so on.

showed that the addition of NAD+

**6.1. Hydrogen**

NADH kinase (like POS5, EC: 2.7.1.86) catalyzes the replacement reaction with two substrates ATP and NADH and two products ADP and NADPH (see Reaction 8). It provides a key source

NADPH is a key cofactor for carotenoid biosynthesis. *Corynebacterium glutamicum* was always used for the production of amino acids, such as L‐isoleucine. By implementing NADPH‐ supplying strategies based on NAD kinase (PpnK), NADH kinase, glucose‐6‐phosphate dehydrogenase (Zwf), and PpnK coupling with Zwf, the expression of all genes increased both the intracellular NADPH concentration and the L‐isoleucine production [25]. Researchers constructed the NADPH regenerators of heterologous NADH kinase to increase the availa‐ bility of NADPH and resulted in a superior S‐adenosylmethionine production in *E*. *coli* without requiring L‐methionine addition [26]. When a *S*. *cerevisiae* strain‐producing carotenoid was constructed by overexpressing glucose‐6‐phosphate dehydrogenase and NADH kinase individually, the final product β‐carotene yield increased by 18.8% and 65.6%, respectively. Thus, NADPH supply improved by overexpression of NADH kinase is more important than

Controlling redox potential at a desired level alters the intracellular metabolic flow in order to favor the formation of desired product(s). Many researches have been conducted in this regard with a large number of examples for enhanced production of metabolites under redox potential–controlled conditions. Most studied metabolites using redox potential–controlled approaches are hydrogen, pyruvate, 1,3‐propanediol, butanol, and 2,3‐butanediol, and the following metabolites are reviewed but provided with references: acetoin [28], succinic acid

Hydrogen, as a clean and high‐combustion energy in widespread areas, can be generated by fermentative anaerobes. Hydrogen production from anaerobic fermentation by bacteria demands reducing level because the standard redox potential of H2/H+ is low. Zhang et al. [8]

resulted in the increase of overall hydrogen. Nakashimada et al. [31] investigated *E*. *aerogenes* for its hydrogen production under different intracellular redox state through the utilization of different substrates bearing various reduction degrees. Low redox potential accelerated the NAD(P)H‐dependent hydrogenase activity in membrane and favors high H2 evolution capability. Ren et al. [32] assessed H2 production during butyric acid fermentation, propionic

during hydrogen fermentation by *Enterobacter aerogenes*

**6. Application of redox potential to fermentation processes**

ATP + NADH = ADP + NADPH (7)

Pyruvate, a product of glycolysis, serves as an effective starting material for the synthesis of many drugs and agrochemicals and is presently used in the food industry. By combining adaptive evolution and cofactor engineering, a series of engineered yeasts that can produce pyruvate using glucose as the sole carbon source was obtained. Consequently, the constructed strains were able to produce 75.1 g/L pyruvate, increased by 21% compared with the wild strain. The production yield of this strain reached 0.63 g pyruvate/g glucose [34].

#### **6.3. Propanediol**

1,3‐propanediol, made from glycerol under anaerobic condition, is a monomer for producing various industrial polymers. Du et al. [35] demonstrated that controlling redox potential at −190 mV was preferable for *Klebsiella pneumoniae* to ferment glycerol into 1,3‐propanediol. They further developed a redox potential–based strategy for screening high productivity strain using the correlation between redox potential level and growth rate [36]. Zheng et al. [37] regulated redox potential under low levels (−200 and −400 mV) during 1,3‐propanediol fermentation in order to avoid the accumulation of by‐product. Wu et al. [38] engineered the pathways of 2,3‐butanediol and formic acid in a recombinant *K*. *pneumonia* to improve 1,3‐ propanediol production. The intracellular metabolic flux was redistributed pronouncedly by shrinking all nonvolatile by‐products and supplying the availability of NADH. Jain et al. [39] established novel metabolic pathways for 1,2‐propanediol in *E*. *coli* by disrupting the major competing pathways for acetate production as well as the ubiquinone biosynthesis pathway that conserved more NADH.

#### **6.4. Butanol**

Butanol attracts public attentions due to its favorable physicochemical properties for blending with or for directly substituting for gasoline. Fermentation of butanol by *C*. *acetobutylicum* is generally a biphasic process consisting of acidogenesis and solventogenesis. It has been reported that an earlier initiation of solvent genesis under redox potential control at −290 mV could increase solvent production by 35% [40]. Li et al. [41] supplemented nicotinic acid, the precursor of NADH and NADPH, into the growth medium, and led to a significant increase of NADH and NADPH levels for a wild‐type *Clostridium sporogenes* strain. As a result, the metabolic pattern was shifted toward the production of more reduced metabolites, in which butanol production was then enhanced. Bui et al. [42] constructed the recombinant *K*. *pneumoniae* by overexpressing the genes *kivD*, *leuABCD*, and *adhE1*, with several NADH regeneration strategies to overcome redox imbalance, including the introduction of NAD+ ‐ dependent enzymes or elimination of the NADH competition pathway (1,3‐propanediol synthesis). The NADH/NAD+ ratio was increased resulting in butanol titer increase [42].

#### **6.5. Butanediol**

2,3‐butanediol (2,3‐BD) is a promising bulk chemical with extensive industry applications. In order to enhance the production of 2,3‐BD, various strategies for increasing the NADH availability were developed through regulation of low dissolved oxygen, supplement of reducing substrates and gene modification. An *udhA* encoding transhydrogenase was intro‐ duced and more NADH from NADPH was provided to allow the enhancement of production [43]. For the same reason, two NADH regeneration enzymes, glucose dehydrogenase and formate dehydrogenase, were introduced into *E*. *coli* with 2,3‐butanediol dehydrogenase, respectively [44]. In other case, an engineered *S*. *cerevisiae* harboring NADH oxidase gene (noxE) from *L. lactis* minimized glycerol accumulation, because intracellular NADH/NAD+ ratio was decreased substantially and carbon flux was redirected to 2,3‐BD from glycerol [23].

#### **7. Redox potential process design: a case study of ethanol fermentation**

Fuel ethanol, the most successful renewable energy so far, is produced worldwide and applied in transportation as alternative to fossil fuel. However, the high cost associated with bioethanol production urges researchers to innovate new fermentation technologies like redox potential– controlled ethanol fermentation. In this section, the role of redox potential in *S*. *cerevisiae* pathways, the correlation between yeast growth and redox potential, and the application of redox potential to very high gravity fermentation will be reviewed.

#### **7.1. The role of redox potential in yeast pathway**

*S*. *cerevisiae* has been considered as a model microorganism, whose genome, proteome, and relevant pathway information are almost unveiled. As illustrated in **Figure 4**, glucose is converted into small molecules through the coupling of redox reactions, in which NADH plays an essential role in key metabolites production such as ethanol, glycerol, and lactate. In this process, glucose is oxidized by NAD+ to make pyruvate and NADH. The surplus of reducing power is then balanced by the formation of glycerol and ethanol, where NAD+ is restored. When the growth environment favors the production of acetic acid, the implementation of redox potential control can alter the trend, leading to a more reduced state toward ethanol production.

Compared with other control parameters, such as temperature, pH, and the ingredients of medium, redox potential has less influence on improving fermentation results. Hence, the implementation of redox potential control in ethanol fermentation was not popular until the new concept of "very high gravity (VHG)" was proposed. VHG is generally regarded as the final ethanol concentration is greater than 15% (v/v) or initial glucose concentration is greater than 250 g/L. VHG is a promising technology to reduce energy consumption and labor cost, as well as elevate the efficiency of the fermenter. However, high sugar concentration depresses cell growth and bioconversion. Redox potential control helps cells survive from osmotic pressure and ethanol toxicity by constructing healthier membranes or other potential mecha‐ nisms. Yeast grown under VHG condition without redox potential control requires much longer fermentation times in order to completely utilize substrate [45]; therefore, the improve‐ ment of ethanol production by redox potential control would be expected.

**Figure 4.** Metabolic pathway of glucose degradation in *Saccharomyces cerevisiae*.

**6.5. Butanediol**

34 Fermentation Processes

production.

2,3‐butanediol (2,3‐BD) is a promising bulk chemical with extensive industry applications. In order to enhance the production of 2,3‐BD, various strategies for increasing the NADH availability were developed through regulation of low dissolved oxygen, supplement of reducing substrates and gene modification. An *udhA* encoding transhydrogenase was intro‐ duced and more NADH from NADPH was provided to allow the enhancement of production [43]. For the same reason, two NADH regeneration enzymes, glucose dehydrogenase and formate dehydrogenase, were introduced into *E*. *coli* with 2,3‐butanediol dehydrogenase, respectively [44]. In other case, an engineered *S*. *cerevisiae* harboring NADH oxidase gene (noxE) from *L. lactis* minimized glycerol accumulation, because intracellular NADH/NAD+ ratio was decreased substantially and carbon flux was redirected to 2,3‐BD from glycerol [23].

**7. Redox potential process design: a case study of ethanol fermentation**

redox potential to very high gravity fermentation will be reviewed.

**7.1. The role of redox potential in yeast pathway**

Fuel ethanol, the most successful renewable energy so far, is produced worldwide and applied in transportation as alternative to fossil fuel. However, the high cost associated with bioethanol production urges researchers to innovate new fermentation technologies like redox potential– controlled ethanol fermentation. In this section, the role of redox potential in *S*. *cerevisiae* pathways, the correlation between yeast growth and redox potential, and the application of

*S*. *cerevisiae* has been considered as a model microorganism, whose genome, proteome, and relevant pathway information are almost unveiled. As illustrated in **Figure 4**, glucose is converted into small molecules through the coupling of redox reactions, in which NADH plays an essential role in key metabolites production such as ethanol, glycerol, and lactate. In this process, glucose is oxidized by NAD+ to make pyruvate and NADH. The surplus of reducing

When the growth environment favors the production of acetic acid, the implementation of redox potential control can alter the trend, leading to a more reduced state toward ethanol

Compared with other control parameters, such as temperature, pH, and the ingredients of medium, redox potential has less influence on improving fermentation results. Hence, the implementation of redox potential control in ethanol fermentation was not popular until the new concept of "very high gravity (VHG)" was proposed. VHG is generally regarded as the final ethanol concentration is greater than 15% (v/v) or initial glucose concentration is greater than 250 g/L. VHG is a promising technology to reduce energy consumption and labor cost, as well as elevate the efficiency of the fermenter. However, high sugar concentration depresses cell growth and bioconversion. Redox potential control helps cells survive from osmotic pressure and ethanol toxicity by constructing healthier membranes or other potential mecha‐ nisms. Yeast grown under VHG condition without redox potential control requires much

is restored.

power is then balanced by the formation of glycerol and ethanol, where NAD+

Lin et al. [45] controlled redox potential under −150 mV, −100 mV, and no control conditions and demonstrated that VHG ethanol fermentation under −150 mV resulted in the highest final ethanol concentration and the highest ethanol‐to‐glucose yield. Compared with the case of 200g glucose/L, the effect of redox potential control becomes significant under VHG condi‐ tions [45]. Jeon and Park [46] cultivated *Zymomonas mobilis* and *S*. *cerevisiae* to produce ethanol in two separate compartments of an electrochemical bioreactor. The results showed that *Z*. *mobilis* favors the reducing environment, but *S*. *cerevisiae* produced more ethanol under higher redox potential conditions [46]. Na et al. [47] observed that ethanol production was enhanced in the anode compartment than in the cathode one, although the reduced environment would be better for fermentation process.

#### **7.2. Correlation between cell growth and redox potential**

During ethanol fermentation, changes of redox potential are caused by two major substances, electron donor NAD(P)H resulting from dissimilatory processes (e.g. glycolysis) and assimi‐ latory processes (e.g. biomass formation), and electron acceptor oxygen dissolved from sparging and/or agitation. The redox potential profiles are thus correlated to cellular activities and oxygen tension.

A typical redox potential profile resembles a bathtub curve. In the beginning, yeast was inoculated into the autoclaved medium where redox potential is as high as normal oxygen tension. Yeast consumes oxygen as the final electron acceptor during respiration process for rapid propagation, causing a steep fall of redox potential (Stage I, **Figure 5**). When dissolved oxygen is nearly depleted, yeast modulates the respiratory requirement from aerobic to anaerobic stages where a short transition is seen in order to alter relevant gene expression and pathways (between Stage I and II, **Figure 5**). After adjustment, yeast cells accelerate their growth rate in the exponential phase with rapid glucose utilization. Although ethanol production is a redox neutral process in theory, the use of reducing substrate like sugar tends to lower fermentation redox potential. The trend of decline in redox potential continues as fermentation proceeds and could drop as low as to −300 mV if there is no other oxidizing reagent present in the fermentation broth (Stage II–III, **Figure 5**). Due to the substrate depletion and the decline of cell viability attributed to ethanol toxicity, the lowest trough in redox potential level is observed (Stage III, **Figure 5**). Near the end of fermentation, an abrupt increase in redox potential is attributed to constant aeration or well agitation. Technically, an uprising curve appearing reveals that the fermentation is about to finish (Stage IV, **Figure 5**).

**Figure 5.** Profiles of redox potential, biomass, and dissolved oxygen.

#### **7.3. Process design using redox potential**

The performance of VHG ethanol fermentation can be further improved by (1) searching for the optimal redox potential setting and (2) extending redox potential control period to prolong the exponential growth phase. Three redox potential control schemes are collected [48]. The simple aeration‐controlled scheme (ACS) has a short redox potential–controlled period. For glucose‐controlled feeding scheme (GCFS), glucose was supplemented along with dissolved oxygen presented in the feed stream. For combined chemostat and aeration‐controlled scheme (CCACS), a constant glucose was fed along with air supply determined by redox potential– controlled device. The GCFS extends the redox potential–controlled period by offering enough glucose for yeast propagation and maintaining the low residual glucose. As a result, the ethanol yield is increased noticeably. The operation of GCFS as a fed batch, as such the buildup of ethanol causes yeast cessation, resulting in incomplete fermentation. The CCACS is a set of continuous equipment that feeds the fresh medium into a fermenter and discharge spent broth into aging vessels at a constant dilution rate. Sterilized air was used to adjust the fermentation redox potential at a predetermined level. In the chemostat fermenter, both intracellular and extracellular factors should reach their respective steady states. Thus, constant growth rate and yeast viability are sustained under a preset redox potential level, which is helpful to prolong the redox potential–controlled duration and to maximize the benefits from redox potential control. The CCACS achieved the longest controlled period and the highest ethanol yield among all three schemes. However, a chemostat device alone could not result in zero glucose discharge. The incorporation of aging vessel design into fermentation operation thus was developed [49].

#### **8. Future work of redox potential and fermentation**

A typical redox potential profile resembles a bathtub curve. In the beginning, yeast was inoculated into the autoclaved medium where redox potential is as high as normal oxygen tension. Yeast consumes oxygen as the final electron acceptor during respiration process for rapid propagation, causing a steep fall of redox potential (Stage I, **Figure 5**). When dissolved oxygen is nearly depleted, yeast modulates the respiratory requirement from aerobic to anaerobic stages where a short transition is seen in order to alter relevant gene expression and pathways (between Stage I and II, **Figure 5**). After adjustment, yeast cells accelerate their growth rate in the exponential phase with rapid glucose utilization. Although ethanol production is a redox neutral process in theory, the use of reducing substrate like sugar tends to lower fermentation redox potential. The trend of decline in redox potential continues as fermentation proceeds and could drop as low as to −300 mV if there is no other oxidizing reagent present in the fermentation broth (Stage II–III, **Figure 5**). Due to the substrate depletion and the decline of cell viability attributed to ethanol toxicity, the lowest trough in redox potential level is observed (Stage III, **Figure 5**). Near the end of fermentation, an abrupt increase in redox potential is attributed to constant aeration or well agitation. Technically, an uprising

curve appearing reveals that the fermentation is about to finish (Stage IV, **Figure 5**).

The performance of VHG ethanol fermentation can be further improved by (1) searching for the optimal redox potential setting and (2) extending redox potential control period to prolong the exponential growth phase. Three redox potential control schemes are collected [48]. The simple aeration‐controlled scheme (ACS) has a short redox potential–controlled period. For glucose‐controlled feeding scheme (GCFS), glucose was supplemented along with dissolved oxygen presented in the feed stream. For combined chemostat and aeration‐controlled scheme (CCACS), a constant glucose was fed along with air supply determined by redox potential– controlled device. The GCFS extends the redox potential–controlled period by offering enough glucose for yeast propagation and maintaining the low residual glucose. As a result, the ethanol yield is increased noticeably. The operation of GCFS as a fed batch, as such the buildup of ethanol causes yeast cessation, resulting in incomplete fermentation. The CCACS is a set of

**Figure 5.** Profiles of redox potential, biomass, and dissolved oxygen.

**7.3. Process design using redox potential**

36 Fermentation Processes

Although many fermentation processes have been well developed with long‐term operability, cost saving is an endless effort, particularly for the production of biofuels and bio‐based chemicals at bulk quantity. Every penny in cost savings is destined to bring huge economic returns. Since redox reactions and homeostasis are the basis for intracellular metabolism, monitoring and controlling redox potential status inside a cell could potentially re‐route metabolic material and energy flow. Numerous works have been done and confirmed that proper redox potential control could alter cellular metabolism, thereby enhancing the conver‐ sion of targeted metabolites.

**Figure 6.** Research and prospect in redox potential–controlled fermentation.

With the availability of technologies that can detect intracellular redox potential levels, an integrated approach, including gene expression, protein biosynthesis, and biomolecular interacting network, should be employed to identify effects of redox potential control on the multiple hierarchy (**Figure 6**). The underlying mechanism of this phenomenon can then be elucidated at molecular and bioprocess engineering levels. The more details obtained, the better applications of redox potential control can be exploited. Consequently, robust strains and optimized processes can be developed toward high‐yield production.

Future perspective of redox potential control is attractive. Fermentation will be carried out using gene‐modified strains featuring tailor‐made redox potential balance. The strain will be subjected to tight regulation through precise redox potential level. Metabolic flux profiles obtained at different redox potential levels will be quantified to achieve the maximum production of various desired metabolites or used to locate potential bottleneck for strain improvement. Benefits from the development of new redox potential–controlled fermentation technology are thus anticipated.

#### **Author details**

Chen‐Guang Liu1 , Jin‐Cheng Qin2 and Yen‐Han Lin3\*

\*Address all correspondence to: yenhan.lin@usask.ca

1 State Key Laboratory of Microbial Metabolism, and School of Life Sciences & Biotechnology, Shanghai Jiao Tong University, Shanghai, China

2 School of Life Sciences and Biotechnology, Dalian University of Technology, Dalian, China

3 Department of Chemical and Biological Engineering, University of Saskatchewan, Saska‐ toon, Saskatchewan, Canada

#### **References**


[5] Kukec A, Berovic M, Celan S. The role of on‐line redox potential measurement in Sauvignon blanc fermentation. Food Technology and Biotechnology. 2002;40:49‐55.

elucidated at molecular and bioprocess engineering levels. The more details obtained, the better applications of redox potential control can be exploited. Consequently, robust strains

Future perspective of redox potential control is attractive. Fermentation will be carried out using gene‐modified strains featuring tailor‐made redox potential balance. The strain will be subjected to tight regulation through precise redox potential level. Metabolic flux profiles obtained at different redox potential levels will be quantified to achieve the maximum production of various desired metabolites or used to locate potential bottleneck for strain improvement. Benefits from the development of new redox potential–controlled fermentation

and Yen‐Han Lin3\*

1 State Key Laboratory of Microbial Metabolism, and School of Life Sciences & Biotechnology,

2 School of Life Sciences and Biotechnology, Dalian University of Technology, Dalian, China

3 Department of Chemical and Biological Engineering, University of Saskatchewan, Saska‐

[1] Needham J G. Curtis gates lloyd. Science. 1926;64:569‐570. DOI: 10.1126/science.

[2] Thrash J C, Coates J D. Review: Direct and indirect electrical stimulation of microbial metabolism. Environmental Science Technology. 2008;42:3921‐3931. DOI: 10.1021/

[3] Girbal L, Soucaille P. Regulation of *Clostridium acetobutylicum* metabolism as revealed by mixed‐substrate steady‐state continuous cultures ‐ role of NADH/NAD ratio and

[4] Snoep J L, Joost M, Demattos T. Effect of the energy‐source on the NADH/NAD ratio and on pyruvate catabolism in anaerobic chemostat cultures of *Enterococcus faecalis*

ATP pool. Journal of Bacteriology. 1994;176:6433‐6438.

nctc‐775. FEMS Microbiology Letters. 1991;81:63‐66.

and optimized processes can be developed toward high‐yield production.

technology are thus anticipated.

toon, Saskatchewan, Canada

64.1667.569‐a

es702668w

**References**

, Jin‐Cheng Qin2

\*Address all correspondence to: yenhan.lin@usask.ca

Shanghai Jiao Tong University, Shanghai, China

**Author details**

38 Fermentation Processes

Chen‐Guang Liu1


water‐forming NADH oxidase in *Bacillus subtilis*. Metabolic Engineering. 2014;23:34‐41. DOI: 10.1016/j.ymben.2014.02.002

[29] Balzer G J, Thakker C, Bennett G N. Metabolic engineering of *Escherichia coli* to minimize byproduct formate and improving succinate productivity through increasing NADH availability by heterologous expression of NAD(+)‐dependent formate dehydrogenase. Metabolic Engineering. 2013;20:1‐8. DOI: 10.1016/j.ymben.2013.07.005

[18] Vemuri G N, Eiteman M A, McEwen J E. Increasing NADH oxidation reduces overflow metabolism in *Saccharomyces cerevisiae*. Proceedings of the National Academy of Sciences of the United States of America. 2007;104:2402‐2407. DOI: 10.1073/pnas.

[19] Dinakar C, Vishwakarma A, Raghayendra A S. Alternative oxidase pathway optimizes photosynthesis during osmotic and temperature stress by regulating cellular ROS, malate valve and antioxidative systems. Frontiers in Plant Science. 2016;7:68. DOI:

[20] Zhang L, Xu Y Y, Gao J. Introduction of the exogenous NADH coenzyme

[21] Yang T W, Rao Z M, Hu G Y. Metabolic engineering of *Bacillus subtilis* for redistributing the carbon flux to 2,3‐butanediol by manipulating NADH levels. Biotechnology for

[22] Zhu Y M, Li H Y, Liu P P. Construction of allitol synthesis pathway by multi‐enzyme coexpression in *Escherichia coli* and its application in allitol production. Journal of Industrial Microbiology & Biotechnology. 2015;42:661‐669. DOI: 10.1007/

[23] Kim J W, Seo S O, Zhang G C. Expression of *Lactococcus lactis* NADH oxidase increases 2,3‐butanediol production in Pdc‐deficient *Saccharomyces cerevisiae*. Bioresource

[24] Gao H, Kim I W, Choi J H. Repeated production of L‐xylulose by an immobilized whole‐ cell biocatalyst harboring L‐arabinitol dehydrogenase coupled with an NAD(+) regeneration system. Biochemical Engineering Journal. 2015;96:23‐28. DOI: 10.1016/

[25] Shi F, Li K, Huan X J. Expression of NAD(H) kinase and glucose‐6‐phosphate dehy‐ drogenase improve NADPH supply and l‐isoleucine biosynthesis in *Corynebacterium glutamicum ssp lactofermentum*. Applied Biochemistry and Biotechnology.

[26] Chen Y W, Xu D B, Fan L H. Manipulating multi‐system of NADPH regulation in *Escherichia coli* for enhanced S‐adenosylmethionine production. RSC Advances.

[27] Zhao X, Shi F, Zhan W. Overexpression of ZWF1 and POS5 improves carotenoid biosynthesis in recombinant *Saccharomyces cerevisiae*. Letters in Applied Microbiology.

[28] Zhang X, Zhang R, Bao T. The rebalanced pathway significantly enhances acetoin production by disruption of acetoin reductase gene and moderate‐expression of a new

Technology. 2015;191:512‐519. DOI: 10.1016/j.biortech.2015.02.077

Biofuels. 2015;8:129. DOI: 10.1186/s13068‐015‐0320‐1

2013;171:504‐521. DOI: 10.1007/s12010‐013‐0389‐6

2015;5:41103‐41111. DOI: 10.1039/c5ra02937f

2015;61:354‐360. DOI: 10.1111/lam.12463

regeneration system and its influence on intracellular metabolic flux of *Paenibacillus polymyxa*. Bioresource Technology. 2016;201:319‐328. DOI: 10.1016/j.biortech.

0607469104

40 Fermentation Processes

2015.11.067

s10295‐014‐1578‐1

j.bej.2014.12.017

10.3389/fpls.2016.00068

