Thesis for the Degree of Doctor of Philosophy

Fungi-based biorefinery model for food industry waste Progress toward a circular economy

Pedro F. Souza Filho

ABSTRACT

The food industry, one of the most important industrial sectors worldwide, generates large amounts of biodegradable waste with high organic load. In recent years, the traditional management methods to treat this waste (e.g., landfilling) have been considered not suitable

because they do not exploit the potential of the waste material. Alternatively, valorization of

food industry waste via a biorefinery model using filamentous fungi is considered to represent an attractive strategy because it minimizes the negative impacts while recovering the nutrients and energy of the waste, in accordance with the concept of the circular economy.

In this thesis, four food processing wastes were utilized as case studies: potato protein liquor (PPL, the soluble fraction of potato starch production waste), the peels wasted during orange juice production, the starchy byproduct of pea protein processes, and the wastewater of a wheat-starch plant. Rhizopus oryzae, a zygomycetous filamentous fungus, was grown with these wastes as a substrate, yielding biomass containing 43% (w/w) protein together with 51% removal of the chemical oxygen demand when cultivated in tenfold-diluted PPL. Moreover,

protein-rich biomass was produced using the pea-processing byproduct (55%) and wheat- Fungi-based biorefinery model for food industry waste: Progress toward a circular economy starch wastewater (51%). In contrast, cultivation in orange peel extract yielded a biomass rich Copyright © Pedro Ferreira de Souza Filho in lipids (20%). The use of PPL was also studied in terms of the economy of fungal cultivation. The biotreatment was found to require only 46% of the capital investment Swedish Centre for Resource Recovery necessary for treating PPL by the traditional strategy (application as fertilizer). In comparison, University of Borås the ascomycetous fungus Aspergillus oryzae yielded superior results compared to those of R. oryzae when grown in the starchy residues. The high protein content of the fungal biomass SE-501 90 Borås, Sweden encouraged the investigation of its use for bioplastic production. The addition of 20% fungal

biomass in a pectin matrix increased the tensile yield of the film and reduced the elongation at Digital version: http://urn.kb.se/resolve?urn=urn:nbn:se:hb:diva-14888 break. Moreover, a positive effect on water vapor permeability of the film was also observed. ISBN 978-91-88838-00-1 (printed) These results indicate the ability of the filamentous fungi to convert resources wasted by the ISBN 978-91-88838-01-8 (PDF) food industry into new products with positive impacts on the economy and the environment. ISSN: 0280-381X, Skrifter från Högskolan i Borås, nr. 89 Keywords: Filamentous fungi; circular economy; biorefinery; food industry; fungal biomass; NMÄ NE RK A E V T S bioplastic; resource recovery. Cover photo: smoked tempeh

Trycksak Borås, 2018 3041 0234

II III

LIST OF PUBLICATIONS

This thesis is based on the results presented in the following peer-reviewed articles:

I. Souza Filho, P.F.; Zamani, A.; Taherzadeh, M. J. Production of Edible Fungi from Potato Protein Liquor (PPL) in Airlift Bioreactor. Fermentation, 2017, 3(1), 12.

II. Souza Filho, P.F.; Brancoli, P.; Bolton, K.; Zamani, A.; Taherzadeh, M. J. Techno- Economic and Life Cycle Assessment of Wastewater Management from Potato Starch Production: Present Status and Alternative Biotreatments. Fermentation, 2017, 3(4), 56.

III. Souza Filho, P.F.; Zamani, A.; Taherzadeh, M. J. Edible protein production by filamentous fungi using starch plant wastewater. Waste and Biomass Valorization, 2018, in Press.

IV. Souza Filho, P.F.; Nair, R.; Andersson, D.; Lennartsson, P.; Taherzadeh, M. J. Vegan- mycoprotein concentrate from pea-processing industry byproduct using edible filamentous fungi. Fungal Biology and Biotechnology, 2018, 5(1), 5.

V. Gurram, R.; Souza Filho, P. F.; Taherzadeh, M. J.; Zamani, A. A solvent free approach for production of films from pectin and fungal biomass – Manuscript submitted for publication.

STATEMENT OF CONTRIBUTION

My contributions to the above publications were:

I. Responsible for the experimental work. Together with the co-authors, I was responsible for the processing of the data and writing of the manuscript.

II. Conception and design of the simulation scenarios, contributed to analysis of the data and writing of the paper.

III. Responsible for the experimental work. Together with the co-authors, I was responsible for the processing of the data and writing of the manuscript.

IV V

IV. Responsible for part of the processing of the data and writing of the manuscript. PREFACE

V. Supervisor of the experimental work. Responsible for the processing of the data and part

of the writing of the manuscript. This dissertation has been produced as part of the requirements for completing the Ph.D. PUBLICATION NOT INCLUDED IN THIS THESIS program in Resource Recovery at the University of Borås. Herein are summarized the main VI. Souza Filho, P.F.; Taherzadeh, M.J. (2019). Integrated biorefineries for the production of results found during my studies under the supervision of Prof. Mohammad Taherzadeh and bioethanol, biodiesel, and other commodity chemicals. In: Vertes A.; Qureshi N.; Dr. Akram Zamani. Blaschek H.P.; Yukawa H. (eds) Green Energy to Sustainability: Strategies for Global Changes in the global climate, accumulation of waste, loss of biodiversity, and exploitation of Industries, Wiley-Blackwell, USA, ISBN: 978-1119152026, Chapter 23, in Press. natural resources have been topics of focus of the international community. Increasing global

population and consequent increasing demand for food can heighten the impact of human activity on the environment. Filamentous fungi act in nature as recyclers of nutrients and have the potential to contribute to the development of a new socio-economic system. Application of the fungi in food, feed, and bioplastic sectors was investigated in this Ph.D. thesis using real waste streams from food and drink processing industries. The scientific bases of the experiments, the main results, and their implications are presented and discussed in the subsequent chapters.

I hope you enjoy the reading.

VI VII

RESEARCH JOURNEY

My fascination for microorganisms started approximately 10 years ago when, as an undergraduate student, I joined the Environmental Engineering group at the Federal University of Rio Grande do Norte (Brazil). There, I learned the enormous potential of these microscopic agents to assist human-kind in pursuing less harmful relationships with the environment. When I came to Borås, I decided to work with filamentous fungi and scrutinize their role as biocatalysts in a biorefinery model.

My first task in the development of this doctoral thesis was to apply the filamentous fungus Rhizopus oryzae to the management of the liquid byproduct of potato starch production (Paper I). The goal of this investigation was to reduce the storage time of the Potato Protein Liquor (PPL) and add value to this material. Protein-rich biomass with potential application as fish feed supplement was obtained together with a reduction in the chemical oxygen demand of the stream. Subsequently, the technological, economic, and environmental implications of this alternative were studied in Paper II. The results suggested a reduction in the initial capital cost and environmental impact when shifting the PPL from its traditional use (i.e., land application as fertilizer) to the fungal strategy.

The treatment of wheat starch manufacturing wastewater containing high content of suspended solids was also investigated using R. oryzae and Aspergillus oryzae (Paper III). The performance of both fungi was compared in terms of their ability to produce biomass, recover nitrogen, and remove organic matter from the discarded water.

Another side stream studied was the starchy byproduct obtained during the isolation of pea protein. The solid material was tested at different concentrations in water, aiming at the production of high protein content fungal biomass (also termed mycoprotein) with intended application as human food. Five different strains were tested including ascomycetes and zygomycetes. The results of this scrutiny are summarized in Paper IV.

Lastly, the waste of industrial orange juice production was used to cultivate R. oryzae, whose biomass was tested for the production of bioplastic films made of pectin using a solvent-free approach (Paper V). Notably, addition of the fungal biomass improved the mechanical properties of the pectin films and reduced the permeability of water vapor through the film.

VIII IX

Therefore, based on the above findings, this dissertation is intended to contribute to the CONTENTS scientific debate regarding how to shape a new socio-techno-economic structure in consideration of the environment.

ABSTRACT ...... III LIST OF PUBLICATIONS ...... V PREFACE ...... VII RESEARCH JOURNEY ...... IX CONTENTS ...... XI ABBREVIATION ...... XIII LIST OF TABLES AND FIGURES ...... XV

Chapter 1 ...... 1 Introduction ...... 1 1.1 Overview ...... 1 1.2 Thesis structure ...... 4

Chapter 2 ...... 7 Food industry ...... 7 2.1 Food industry waste ...... 8 2.1.1 Potato Starch and Potato Protein Liquor (PPL) ...... 11 2.1.2 Wheat Starch Wastewater ...... 13 2.1.3 Pea protein ...... 16 2.1.4 Orange waste ...... 18

Chapter 3 ...... 19 Biorefinery approach for management of food processing waste ...... 19

Chapter 4 ...... 25 Filamentous fungi ...... 25 4.1 Fungi and the environment ...... 25 4.2 Biomass for food and feed applications ...... 27 4.3 Rhizopus oryzae ...... 29 4.4 Aspergillus oryzae ...... 33

Chapter 5 ...... 37 Bioplastics ...... 37

X XI

5.1 Pectin biofilms ...... 38 ABBREVIATION 5.2 Biomass for bioplastic production ...... 40

Chapter 6 ...... 43 Circular economy approach ...... 43 BAT Best Available Techniques 6.1 Economic evaluation of alternative biotreatment of industrial waste ...... 44 COD Chemical Oxygen Demand 6.2 Social and ethical aspects ...... 46 CWFS Citrus Waste Free Sugars Chapter 7 ...... 49 Concluding remarks and future work ...... 49 EU European Union 7.1 Concluding remarks ...... 49 HTPL Heat Treated Potato Liquor 7.2 Future work ...... 50 IPCC Intergovernmental Panel on Climate Change Acknowledgements ...... 51 References ...... 53 NPV Net Present Value

NREL National Renewable Energy Laboratory

PPI Pea Protein Isolate

PPL Potato Protein Liquor

PPP Polluter Pays Principle

SEM Scanning Electron Microscopy

TKN Total Kjeldahl Nitrogen

USD United States Dollar

WVPC Water Vapor Permeability Coefficient

XII XIII

LIST OF TABLES AND FIGURES

Table 2.1. Examples of food processing wastes and amount generated according to geographic location. 10 Table 2.2: Characterization of the PPL used in this thesis. 13 Table 2.3. Characterization of the wheat-starch industry’s wastewater streams. 16 Table 2.4. Characterization of the pea-processing byproduct. 17 Table 5.1. Mechanical properties and thickness of the films with and without incorporation of fungal biomass. 41

Figure 1.1. Key factors affecting food security. 2 Figure 1.2. Graphical representation of the waste hierarchy applied to food waste. 3 Figure 2.1. Block flow diagram representing the preliminary wastewater treatment plant and the streams in the wheat-starch facility. 15 Figure 3.1. Valuable compounds derivable from A) fruit and vegetable and B) meat and derivatives and dairy products. 21 Figure 3.2. Agro industrial wastes with potential application as substrate in a fungal- based biorefinery and possible value-added products. 23 Figure 4.1. Microscopic photo of N. intermedia mycelium grown in submerged culture using starch as substrate. 27 Figure 4.2. Changes in the morphology of R. oryzae cultivated in Erlenmeyer flasks when varying the PPL dilution rate. 30 Figure 4.3. Concentration profile of metabolites during cultivation of R. oryzae in 10-fold diluted PPL in airlift bioreactors. 31 Figure 4.4. Concentration profiles of metabolites during cultivation of R. oryzae in ED1 and ED2 media with and without enzyme supplementation. 32 Figure 4.5. Concentration profiles of metabolites during cultivation of A. oryzae in ED1 and ED2 media with and without enzyme supplementation. 34 Figure 4.6. Biomass yield after 36 h of cultivation in airlift bioreactor and 2% w/v pea-processing byproduct medium. 35 Figure 5.1. Representative chemical structure of pectin showing typical repeating groups. 38

XIV XV

Figure 5.2. Optical photo of a pectin/glycerol biofilm produced by compression Chapter 1 molding. 39 Figure 5.3. SEM images of the surface and cross-section of the pectin-based bioplastic films. 42 Figure 6.1. Process flow diagram describing the three scenarios studied for

management of the potato liquor. 45 Figure 6.2. Net present value for the different scenarios considering different project Introduction life time. 46

1.1 Overview

The continuous growth of the global population has raised concerns regarding food and nutrition security for subsequent generations [1]. Food and nutrition security is defined as the situation in which all people, at all times, retain physical, social, and economic access to sufficient, safe, and nutritious food that meets their dietary needs and food preferences, coupled with a sanitary environment along with adequate health services and care to support an active and healthy life [2]. Presently, the uneven global distribution of food creates contradictory outcomes: whereas approximately 800 million people suffer from chronic hunger, more than two billion are overweight or obese [2, 3]. Additionally, in the next 20 years, rates of diabetes and obesity as well as food insecurity are predicted to worsen drastically if changes in the food system are not manifested [4]. Although calories (i.e., energy) constitute an important factor when examining global food requirements, access to sufficient macro- and micronutrients must also be considered. Currently, around one billion people worldwide do not meet their daily protein requirement, thereby becoming susceptible to health problems such as growth failure, muscle weakness, and an impaired immune system [5].

The drivers of human nutrition include ecological, social, and economic dimensions (Figure 1.1) [6]. Over the past 50 years, the growing food requirement has primarily been met through improvements in agricultural productivity. Although a further increase in agricultural productivity can assist in meeting the nutritional demands of the future generations, this alone may not be sufficient. Moreover, such increase would demand a large percentage of the Earth’s resources (e.g., land, water, and minerals), which are already overexploited [7]. New

XVI 1 approaches should therefore incorporate socioeconomic aspects to improve food availability, In addition, the biodegradable nature of the raw-materials used in the food industry results in access, and utilization, while enhancing ecosystem services [6]. One alternative approach to the generation of large amounts of biodegradable waste. Although this is exploited to a small guarantee food security is the improvement of food processing industry. The mechanization of extent; e.g., as animal feed, the majority of food processing waste is rather disposed of via the food processing from the middle of the 19th century has improved the efficiency of the environmentally sensitive routes [11]. formulation, manufacture, distribution, and sales of food. Moreover, in industrialized Management of this waste has been regulated in the EU since 2008 via the directive countries, industrial food processing, with its economies of scale, has increased the 2008/98/EC, which is based on the waste hierarchy and the Polluter Pays Principle (PPP) availability of foods causing the concomitant decrease of food insecurity and nutrient [12]. The waste hierarchy ranks waste prevention and management options in order of deficiency [8]. In particular, food processing benefits include improvements in preservation, priority: i) prevention, ii) reuse, iii) recycling, iv) recovery, and v) disposal (Figure 1.2). The nutritional content, bioavailability of nutrients, shelf life, and safety of food [3]. ordering is related to the environmental impact and the extent of resources that can be saved by each option. Specifically, the system is organized to divert the waste from landfills, a management option with potential threat to the environment and human health and with large land requirements [13, 14]. The waste hierarchy has received wide support in most developed countries as the main framework on which to develop waste management policies [15].

Figure 1.1. Key factors affecting food security. Reprinted with permission [6].

Figure 1.2. Graphical representation of the waste hierarchy applied to food waste. Reprinted Currently, the provision of food and drink constitutes one of the largest sectors of global with permission [16]. industrial activity. In the European Union (EU), this represents the largest manufacturing sector in terms of jobs and value-added products [9]. Industrial processing of food includes several unit operations such as washing and cleaning, milling, sorting and separation, mixing, Waste management is generally carried out by agencies handling waste management smoking, heating, drying, fermentation, canning, and extrusion cooking. Each of these facilities. The main role of the food industry in waste management, as the food industry does operations causes both detrimental and beneficial changes to the components of food [3, 10].

2 3 not act in this sphere, is as main player in the first measure; i.e., prevention [17]. It is  In Chapter 3, the valorization of food industry waste using a biorefinery model is important to highlight that a significant amount of food processing waste can be avoided [18]. described. In this regard, food industrial waste can be classified as wanted and unwanted. The wanted  Chapter 4 introduces the group of microorganisms known as filamentous fungi and waste constitutes the fraction generated through the processing of the food (e.g., inedible presents the reasons justifying their use in a biorefinery. Two specific strains: Rhizopus parts, washing water). The unwanted waste, in contrast, represents losses (e.g., transportation oryzae and Aspergillus oryzae are presented together with the results obtained. losses) and, therefore, the need for additional raw material [10]. Both fractions have in common that they represent the spoilage of natural resources.  Chapter 5 provides the definition of bioplastic. The properties of the pectin biofilms with incorporation of fungal biomass are presented. The emission of pollutants by industries is regulated in the EU by another directive: the Industrial Emission Directive (2010/75/EU) [19]. This legislation aims to reduce harmful  Chapter 6 introduces the topic of circular economy including a discussion based on industrial emissions by fixing limit values for the emission of pollutants according to the Best economic, social, and ethical aspects related to the thesis. Available Techniques (BAT). The BATs are determined according to the industrial sector,  Lastly, in Chapter 7 the conclusions of the thesis are summarized and directions for future taking into account the latest technical developments and are thus periodically reviewed and works are suggested. updated. The PPP claims that the entity which profits from a process that causes pollution should bear the costs of such pollution [20] in the form of a clean-up or taxation [21]. Therefore, rather than identifying alternatives within the industry to recover wastes, many industries send their wastes to be treated by third parties.

An alternative to the traditional treatment methods that focus on the simple removal of nutrients is the use of the waste for the production of useful biomaterials by cultivating microorganisms [22]. Accordingly, the main goals of this thesis were to create alternative solutions for side streams generated by the food industry by applying filamentous fungi as biocatalysts and to produce fungal biomass with promising characteristics for application as feed, food, and bioplastics component. Several industrial residues were explored and the produced biomass and recovered nutrients were analyzed with regard to their potential utility.

1.2 Thesis structure

This thesis is organized in six chapters:

 Chapter 1 introduces the thesis and presents its associated motivations and aims.

 Chapter 2 provides a brief review of the food industry, describes food industry wastes, and presents the specific wastes investigated in this thesis.

4 5

Chapter 2

Food industry

The industrial processing of food has as a main goal to provide safe products with increased shelf life [23]. The industrialization of food processing is responsible for the present food supply being plentiful, varied, inexpensive, and devoid of strict dependence on geographical or seasonal constraints [24]. In the EU, the food and drink industry has an annual turnover of more than 1 trillion euros and employs more than 4 million people [25].

Different from other chemical industries, food industry products are not chosen according to objective parameters such as purity. Instead, the main reasons underlying product choice are related to elusive characteristics such as tradition and emotion, along with other difficult to measure features including taste, texture, and smell [23]. The results from applying these parameters can be found on modern supermarket shelves, with the number of different products having increased from approximately 1 400 in 1954 to currently over 30 000 [26]. However, the quality of raw material and products varies seasonably, making it difficult to determine the most efficient way to convert the raw materials into final products [10]. Historically, most processes used in the food industry were scaled up from artisanal practices utilized in smaller kitchens. Conversely, from the middle of the 20th century, the application of engineering principles to food processing began to appear in academia, increasing production efficiency by providing a scientific basis for the empirical methods used [10, 23].

However, there remains a need to increase efficiency in the use of resources by the food industry as well as by the whole food supply chain. Several steps in the food supply chain, such as processing, packaging, transportation, and storage are highly inefficient in their current form, considering the volume of waste they generate during their various stages. Furthermore, landfilling still constitutes a popular option for food waste disposal [27, 28].

6 7

Innovation is not as common in the food industry as in other industrial sectors, as it relies according to the specific process and available options. Common strategies for waste recovery upon widely-accepted products and well-established markets. Generally, innovation in a include its application as animal feed, fertilizer, and other applications of industrial ecology business sector depends on a network of mediators. Following the adoption of the EU [32]. directives cited above, waste management in the food industry has been subjected to more Owing to the organic nature of food processing waste, spoilage usually comprises a relevant stringent laws regarding environmental impacts. Moreover, escalating costs for the purchase concern. Fish and meat processing residue, for example, exhibit high sanitary risk and are of fresh water and effluent treatment has stimulated the industrial sector to improve waste therefore unlikely candidates for recovery. However, even these byproducts have been management practices [11]. In addition to the legal and economic factors, social aspects explored [28], with e.g., the production of protein hydrolysate and amino acids from sardine regarding consumer behavior has further pushed the need for improvement in waste solid waste having been tested [33]. To avoid the decay of food industry waste, it is usually management processes. recommended that its valorization occur in a decentralized way, such as at the site of waste production. Such approaches also reduce the costs related to the transportation of this low value material [31]. 2.1 Food industry waste

The characteristics of food industry waste differ from those of municipal food waste. In particular, it is easily accessible and collectable as it is produced in large amounts at a specific site rather than over a large area such as a city. Moreover, unlike agricultural waste, food industry waste needs fewer pretreatment steps to be converted into value-added products as it has previously been treated during food processing. New production practices and processes should be implemented in the food industry to increase efficiency with regard to the use of resources and reduce waste generation [23]. Notably, the most effective way of reducing the environmental footprint of food waste is the incorporation of this waste into productive processes [29]. Historically, however, the value added to a material by processing did not supersede the relative cost of disposing this material as waste, offering little incentive to search for alternative management techniques [11].

A drawback in the analysis of food industry waste is the lack of information regarding its amount. This data unavailability may be due in part similar wastes or by-products being often named and categorized differently and to the production data (including waste amounts) being kept confidential [30]. It is clear, however, that a wide variety of liquid, solid, and gaseous side streams are released by the food industry. This variety is a result of the multiple processes used in the food industry and the seasonal variations in food, which cause fluctuations in the chemical composition and pH of the waste [31]. Table 2.1 shows some examples of waste streams generated during food industrial processing and their amounts according to geographical location. Therefore, some degree of diversification for efficient waste management is required [11]. The most efficient management technique can be chosen

8 9

Table 2.1. Examples of food processing wastes and amount generated according to geographic 2.1.1 Potato Starch and Potato Protein Liquor (PPL) location. Adapted from [31]. PPL comprises the liquid stream generated after the extraction and refining of potato starch. Annual amount available Food industry waste Geographical location Potato constitutes the fourth most produced food crop worldwide: in 2016, over 370 million (thousand metric tons) metric tons was produced globally [34]. Over half of the potatoes harvested in North America Olive mill residue 30 000 Mediterranean basin Tomato pomace 4 000 Europe and Europe are processed to produce French fries, chips, starch, and other products [35, 36]. Wheat straw (surplus) 5 700 UK In Europe, 13.3% of the starch produced comes from potatoes [37]. Starch production starts Whey (surplus) 13.5 Europe with the reception of the potatoes in the facility and subsequent washing. The next step in Brewer's spent grain 3 400 EU-27 starch extraction is the rasping of the tuber. This is performed in rapidly-rotating drums with Pea pods 54 UK blades to open the cells and release the starch granules. Thereafter, washing of the rasped Egg shells 11 UK potato flushes the granules out from the cells. A mixture of water, cell walls, starch, and other Spent coffee grounds 15.6 UK soluble substances is obtained and sent to centrifuges, where the fruit juice is collected. Tomato pomace 53.8 Spain Subsequently, filters are used to collect the pulp and the concentrated starch is washed to Potato peels 100 (dry basis) UK remove impurities [38]. On average, each metric ton of potato yields approximately 200 kg of Sugarcane bagasse 194 692 Brazil starch, 100 kg of pulp, and 700 L of fruit juice [39]. Potato pulp is mostly composed of starch Grape pomace 15 000 USA (37% w/w dry basis), pectin (17%), cellulose (17%), hemicellulose (14%), and protein (4%). Nut shells and pits 400 California, USA Corn residue 42 000 Brazil The risk of biological contamination causes it to be sold as a low-value supplement for animal Cocoa pods 20 000 Ivory coast feed. However, it has a negative impact on the sensory properties of the feed [40-42]; Cashew shell nut moreover, despite the high pectin content, little information regarding the extraction and 20 Tanzania liquid application of this polymer in potato pulp is available [43]. Other minor applications include Palm shells (from 4 300 Malaysia the use as fertilizer and as a substrate in anaerobic digestion for the production of biogas [41]. palm oil production) Rice husks 110 000 Global The fruit juice contains approximately 5% dry matter, of which one-third is comprised of Citrus fruit 15 600 Global proteins, peptides, amino acids, and amines. The pH of the potato juice is corrected and steam processing residues injected (110 °C) to recover the proteins by coagulation [39, 44]. Proteins present in potato Apple pomace 3 000 – 4 200 Global are considered to be of high quality owing to the presence of lysine, an amino acid not Banana processing 9 000 Global frequently found in other vegetables and cereals. The soluble proteins present in the fruit juice Oat straw 11 000 Global Barley straw 58 000 Global are generally divided in three groups: patatin, protease inhibitors, and others [45]. Patatin Rice straw 731 000 Global forms up to 40% of the soluble proteins and has a molecular weight between 40 and 45 kDa. Rapeseed meal 35 000 Global Patatin comprises a dimer glycoprotein with demonstrated antioxidant abilities, lipid acyl hydrolase activity, and excellent foaming and emulsifying properties. The protease inhibitors are found in the range between 5 and 25 kDa and correspond to approximately 50% of the In this thesis, four side streams from the food and drink industry were studied. Their proteins present in the potato fruit juice. Their anti-carcinogenic and high satiety properties production process and characteristics are presented in the following sections: have been studied [46]. However, the isolation of proteins using the described methods causes them to lose their functional properties and renders a salty, bitter taste, reducing their value

10 11 and preventing their use as food additives [47, 48]. Although different methods (salt, acid and Table 2.2: Characterization of the PPL used in this thesis [44]. ethanol precipitation, carboxymethyl cellulose complexation, and chromatographic Composition (g/L) Average techniques) for the isolation of proteins from potato fruit juice have been studied [45, 49-51], Sugars: Glucose1 72.63 ± 6.93 the conflicting results have discouraged their current application in the industry [46]. Present as monosaccharide 17.49 ± 0.34 As component of: Sucrose 24.97 ± 3.20 The residual liquid after protein removal, termed Heat-Treated Potato Liquor (HTPL, 3% dry Raffinose 1.66 ± 0.04 matter), is generally concentrated by evaporation, with 700 L of fruit juice yielding 23 kg of Glucans2 0.75 ± 0.06 1 PPL and 17 kg of protein. The steam is used in the installation as a heating utility and the PPL Fructose Present as monosaccharide 16.17 ± 0.28 (40% solids) is most commonly used as a fertilizer [39], which can improve the potato As component of: Sucrose 24.97 ± 3.20 harvesting yield, especially on poor, permeable, and water-deficient soils [52]. However, the Raffinose 1.66 ± 0.04 long-term application of PPL to the soil leads to several problems such as clogging of the soil Galactose1 10.00 ± 2.43 by the fine fibers and starch granules remaining in the PPL, contamination of the ground Present as monosaccharide 0.08 ± 0.01 As component of: Raffinose 1.66 ± 0.04 water, and disaffection of the neighboring population owing to the associated bad odor. 1 Arabinose 1.30 ± 0.82 Moreover, potassium and nitrogen can accumulate in the soil, resulting in excessive Total 126.71 ± 8.05 concentrations of these compounds [52, 53]. Lastly, as the majority of potato processing Nitrogen compounds: Total Kjeldahl N 20.25 ± 0.86 occurs during the winter, the potato starch facilities to store large volumes of PPL for several Ammonia 1.86 ± 0.12 months because the soil during this period exhibits reduced biological activity that prevents Asparagine 20.98 ± 1.26 the uptake of the liquor [53]. Total Solids 438.77 ± 3.98 Insolubles 9.09 ± 0.41 PPL, kindly provided by Lyckeby Starch AB (Kristianstad, Sweden), was characterized in Ashes 114.19 ± 12.82 terms of sugars, nitrogen compounds, total solids, ashes, and Chemical Oxygen Demand COD 300 ± 26 pH 5.35 (COD). The results are summarized in Table 2.2 [44]. The stream contained a total 1 Total monosaccharides (except fructose) determined using the NREL methodology [54]. 2 Insoluble carbohydrate content above 120 g/L, almost 20% comprising hexoses present as in ethanol 95% (v/v). monosaccharides (glucose, fructose and galactose). Sucrose and raffinose (di- and trisaccharides of hexoses, respectively) were present in substantial amounts, representing approximately 40% of the carbohydrates. Total nitrogen as determined by the Kjeldahl 2.1.2 Wheat Starch Wastewater method was approximately 20 g/L. As the main amino acid, L-asparagine concentration was Wheat constitutes the crop with the largest harvested area worldwide and plays an important 20.98 ± 1.26 g/L, representing 22% of the total nitrogen content. The density of the PPL was role in the diet of the global population. It provides more protein than poultry, swine, and determined to be 1.15 g/mL and the COD was 300 g/L. bovine meat combined [34, 55]. The EU is the main producer, representing 20% of the global production [56]. Processing of wheat for production of its fractions such as gluten and starch is growing in importance [57]. In particular, wheat represents the main crop used for starch

production in the EU, accounting for 36.4% of the total 23.6 million tons of crops annually processed in the region and 40.0% of the 10.7 million tons of starch produced [37].

12 13

Several methods for the production of wheat starch were described by Van Der Borght et al. (2005) [57]. These can be classified according to the starting material used: wheat flour or the whole grain. The use of the whole grain increases the yield of starch production although the harsh conditions applied reduce the recovery of gluten. Generally, the processes start with the grains being steeped in water for a period of 12 h to one week. Control of pH and temperature as well as the addition of enzymes and sulfur dioxide improves the efficiency of this Figure 2.1. Block flow diagram representing the preliminary wastewater treatment plant and operation. Thereafter, grinding and sieving removes the bran fraction. The starch present in the streams in the wheat-starch facility [61]. the liquid phase is recovered generally by centrifugation or tabling (a method consisting of flowing a suspension of starch and gluten over an inclined table, allowing the starch to settle and the gluten, water, and water extractable fraction to run off). Finally, the starch is washed The wastewater streams were characterized in terms of total solids, ashes, carbohydrates, and and dried. nitrogen compounds with the results being presented in Table 2.3. According to the obtained

Industrially, the majority of the processes for producing wheat starch utilize wheat flour. The results, the ED1 and ED2 samples, representing the clarified liquid from the first and second flour is mixed with water to form a suspension. Gluten proteins agglomerate and are separated decanters, respectively, contained most of the carbohydrates as soluble components. In from the starch by sieving, centrifugation, tabling, hydrocyclones, or decanters [57]. These contrast, the solid-rich stream (SS) had a high content of suspended solids (71.79 g/L), 86% methods share the use of water as a solvent for fractionation of the wheat. The process, of which was formed by glucans. The content of soluble solids in the three samples was therefore, generates wastewater that requires treatment prior to discharge. This wastewater similar, with most of the carbohydrates not differing significantly (within a 95% confidence represents an environmental challenge owing to the large COD, which ranges between 6 and interval). The results were also similar for total Kjeldahl nitrogen (TKN), which varied 10 g/L [22, 58-60]. between 1.25 and 1.40 g/L. COD of the SS stream (58.7 ± 2.1 g/L) was more than double that of ED1 (27.3 ± 1.2 g/L). In the case studied in this thesis, wastewater from a wheat grain processing plant was analyzed [61]. The material was kindly provided by Interstarch GmbH (Elsteraue, Germany). The stream has a high content of fibers, which are currently removed by decantation using two decanters in series prior to sending the clarified liquid to the municipal wastewater treatment plant. Figure 2.1 illustrates the scheme of the decanters and shows the three studied streams.

14 15

Table 2.3. Characterization of wheat-starch industry wastewater streams [61]. contributions because of their functional and nutritional properties, relatively low cost, low ED1 ED2 SS allergenicity, and non-genetically modified organism status [63]. Moreover, pea proteins are

pH 5.28 5.29 3.85 rich in essential branched-chain amino acids; i.e., leucine, isoleucine, and valine, which TKN (g/L) 1.25 ± 0.06 1.34 ± 0.32 1.40 ± 0.04 account for approximately 35% of the essential amino acids in muscle proteins [64, 65]. Total solids (g/L) 28.52 ± 0.08 32.19 ± 0.43 96.68 ± 0.70 Ashes (g/L) 1.93 ± 0.02 1.90 ± 0.11 1.80 ± 0.03 Pea is the most common pulse used for protein isolate production because it can be cultivated Suspended solids (g/L) 3.70 ± 0.11 6.78 ± 0.02 71.79 ± 0.14 Glucan (mg/g) 446 ± 9 600 ± 73 857 ± 90 extensively in many diverse areas worldwide and the hull can be easily removed [66]. Pea Starch (mg/g)* 53 ± 4 361 ± 2 771 ± 6 proteins are commercialized in three main forms: pea flour, pea protein concentrate, and pea Xylan (mg/g) 103 ± 49 31 ± 0 65 ± 19 protein isolate (PPI). Pea is hulled and dry milled for the production of flour. Pea protein Galactan (mg/g) ND ND 3 ± 0 Arabinan (mg/g) ND ND 51 ± 8 concentrate with up to 65% protein content can be produced through the use of dry separation Mannan (mg/g) ND ND ND methods (e.g., air classification). The PPI is obtained by solubilizing the proteins in alkali or Lignin (mg/g) 102 ± 7 76 ± 18 49 ± 16 acid, followed by isolation through isoelectric precipitation and membrane separation Soluble solids (g/L) 24.82 ± 0.14 25.41 ± 0.43 24.88 ± 0.71 processes such as ultrafiltration and diafiltration. These methods can yield pea protein with up Glucose (g/L) 1.68 ± 0.02 1.62 ± 0.08 3.65 ± 0.29 Fructose (g/L) 1.43 ± 0.02 1.44 ± 0.01 1.41 ± 0.30 to 95% purity [66-68]. The main byproduct of the PPI is pea starch; however, this material Galactose (g/L) 0.50 ± 0.03 0.53 ± 0.01 0.56 ± 0.01 exhibits poor functional properties and does not afford many applications in food formulation Raffinose (g/L) 0.24 ± 0.08 0.04 ± 0.03 ND [69]. Glucan (g/L) 7.22 ± 0.07 8.07 ± 0.17 10.44 ± 0.35 Xylan (g/L) 2.32 ± 0.02 2.41 ± 0.03 3.91 ± 0.04 The pea-processing byproduct used was generously provided by Protein Consulting AB Galactan (g/L) 0.08 ± 0.03 0.08 ± 0.01 0.07 ± 0.01 Arabinan (g/L) 1.75 ± 0.07 1.81 ± 0.04 3.19 ± 0.04 (Sweden). It is a white, odorless, fine powder containing 56% starch and 18% proteins [5]. Mannan (g/L) 0.04 ± 0.01 0.02 ± 0.00 ND The residual protein present in the byproduct originates from protein bodies, agglomerates, Lactic acid (g/L) 1.29 ± 0.02 1.29 ± 0.01 3.23 ± 0.14 chloroplast membrane remnants (which enclose the starch granule), and a water-soluble Acetic acid (g/L) 1.06 ± 0.07 1.04 ± 0.05 1.48 ± 0.13 Glycerol (g/L) 0.23 ± 0.02 0.24 ± 0.01 0.59 ± 0.00 fraction that is presumably derived from the dehydrated starch [69]. A detailed composition of Ethanol (g/L) 0.29 ± 0.01 0.68 ± 0.03 2.04 ± 0.15 the pea-processing byproduct is presented in Table 2.4. COD (g/L) 27.3 ± 1.2 36.7 ± 4.7 58.7 ± 2.1 ND: not detected. *Part of the measured glucans

Table 2.4. Characterization of the pea-processing byproduct [5]. Component Content (% w/w in dry basis) 2.1.3 Pea protein Protein 18.19 ± 0.33 Legumes, which comprise the edible seeds of pod-bearing plants, are an important source of Ash 2.98 ± 0.03 Moisture 10.54 ± 0.19 nutrients such as proteins, minerals, starch, vitamins, and poly-phenols [62]. Pea (Pisum Arabinans 2.61 ± 0.06 sativum) constitutes the second most important legume worldwide with almost 20 million Xylans 0.00 metric tons produced annually [34, 62]. It has applications in feed, food, and food ingredients Galactans 2.30 ± 0.04 Glucans 62.38 ± 0.51 including proteins, starches, flour, and fibers. Notably, an increasing demand of non-animal of which: alternative protein sources for reasons of consumer preference, nutritional requirements, and Starch 56.34 ± 2.52 population growth has been observed. In this context, pea proteins provide important

16 17

2.1.4 Orange waste Chapter 3 Orange constitutes the most produced citric fruit with an annual global production of over 70 million metric tons [34]. Approximately half of the harvested oranges are industrially used for juice production. During juice extraction, half of the orange weight ends up as waste, formed by the seeds, peel, segment membrane, and residual pulp [70]. The composition of orange waste includes poly-, di-, and monosaccharides. The main polysaccharides are cellulose, hemicellulose, starch, and pectin, which are found in the peels. Di- and monosaccharides Biorefinery approach for management of food (sucrose, glucose, and fructose) derive from the pulp [71]. processing waste Orange waste is industrially used for the production of fiber and food ingredients such as pigments and flavonoids as well as cattle feed [72]. However, the main industrial application of orange waste is for the production of pectin, comprising most of the pectin globally produced [73]. However, a large quantity of the waste is still not industrially used and ends up Our present society has a very strong dependence on petroleum as a consequence of the being disposed in the environment [70]. The presence of soluble sugars negatively affects the intensive consumption of its derivative chemicals and fuels. The large spectrum of process of pectin extraction. Additionally, the sugars can be utilized by microorganisms, applications of petroleum products has led to the reduction of oil reserves, raising political resulting in the fast degradation of the material. Removal of sugars is, therefore, important for and economic concerns [75]. Additionally, as noted by the Intergovernmental Panel on the industrial application of orange waste [74]. Climate Change (IPCC), climate changes recently observed worldwide are extremely likely to

As the sugars are soluble in water, they can be extracted by washing the orange waste. The be caused by human activity, among other factors, owing to the emission of greenhouse gases. result is a nutrient-rich liquid stream that can be used for bioprocess applications. The Burning of fossil fuels is the largest known contributor for such emission [76]. The depletion extraction of sugars was performed according to a method described previously[74]. Orange of oil reserves is not, however, the only threat to the sustainability of the present production waste, kindly provided by Brämhults Juice AB (Borås, Sweden), was mixed with water in a system. Approximately 90% of all the material resources used for production of goods end up predetermined ratio (1 kg of wet orange waste and 1.5 L of tap water) and left at room as waste. Therefore, extensive research has been performed to identify renewable sources for temperature (20 °C) to soak overnight. Thereafter, the mixture was added into a fruit press energy and chemicals to be used as substitutes for oil as well as to develop new technologies machine (Lancman™ VSPIX120) where it passed through a textile filter (pore size 3 mm). that bond economic growth with environmental sustainability [77].

The citrus waste free sugar solution (CWFS) was collected and stored at 5 °C until use. The Reduction in oil dependence and consequent impact on climate changes can be achieved by CWFS was characterized as containing 18.3 ± 0.2 g of glucose and 18.7 ± 0.9 g of other the implementation of a sustainable economy based on renewable sources. Recent results have sugars. reinforced the potential of plant-based raw materials to replace fossil resources in many

industrial applications. The processing of plant biomass in a biorefinery platform leads to the production of food, feed, chemicals, materials, fuels, energy, and goods. Moreover, the sustainable use of biomass in an integrated manner can take advantage of the differences in biomass constituents and intermediates, maximizing the economic value of the biomass and reducing the resulting waste streams [30, 77]. This new model of production can be made

18 19 possible by increasing the efficiency and performance of process, and conversion and distribution technologies of the products [77].

Although human utilization of organic materials as feedstock for chemical, energy, and material production is not new, the interest in the effective exploitation of unavoidable organic wastes has been renewed in recent years, ignited by the aim to reduce the environmental footprint and achieve a more secure supply of renewable resources [78]. The emergence of a new concept of biorefinery occurred at the end of the 19th century, when numerous solutions to the exploitation of the residues of the paper and pulp industry started being investigated. A plant to produce ethanol from spent sulfite liquor is claimed to have been first operated in Skutskär, Sweden, as early as 1909 [79].

Biomass is considered as environmentally sustainable materials since its origin is mostly from industrial waste and agricultural and forestry management [77]. The use of the waste of one facility as an input in another facility consists in the concept of an integrated biorefinery. The integration in this system goes beyond the material streams and also includes energy exchanges, resulting in an overall increased production with minimum energy requirement [80]. In this context, residues of the food industry appear particularly useful as raw material for biorefinery applications. Such applications may help to reduce the costs involved in waste disposal. Additionally, rerouting of waste streams for new applications may lead to the recovery of nutrients via chemical purification (Figure 3.1), thermochemical processing, or through the cultivation of microorganisms [81].

Generally, biorefinery processes can be divided in four different platforms: thermochemical, biochemical, mechanical/physical, and chemical processes. Thermochemical platforms convert biomass mainly by gasification and pyrolysis. In the former process, syngas is produced by heating the biomass usually to the range from 800 to 900 °C with approximately one third of the oxygen necessary for a total combustion. The syngas can be directly burned to generate power or be used to produce methanol and hydrogen, both having application as fuel. In the pyrolysis process, the biomass is heated to around 500 °C in complete absence of oxygen, producing solid (char), gas (H2, CO, CH4, C2H2, C2H4), and liquid phase products.

The liquid phase obtained, called bio-oil can be used in engines and turbines [82, 83]. Figure 3.1. Valuable compounds derivable from A) fruit and vegetable and B) meat and its

derivatives along with dairy products. Adapted with permission from [84].

20 21

The mechanical platform involves processes that do not change the state or composition of the feedstock. They are used for size reduction and separation of the components of biomass [75]. Hydrolysis and transesterification are chemical processes commonly used in a biorefinery. Hydrolysis reactions are performed by the use of acids, alkalis or enzymes to depolymerize polysaccharides and proteins into their component monomers (e.g., glucose from cellulose) or derivate chemicals (e.g., furfural from xylose). Through the transesterification process, vegetable oils can be converted to methyl or ethyl esters of fatty acids, consisting in the most common method currently used to produce biodiesel [75].

Lastly, the biochemical platform focuses on fermentation of carbohydrates. The biomass processing starts with the preparation of the feedstock, followed by the conversion of fermentable sugars, bioconversion of the sugars using biocatalysts, and processing to deliver value-added chemicals, fuel-grade ethanol and other fuels, heat, and/or electricity [82]. Low processing temperature and pressure, and high selectivity and specificity of components targeted and products generated are among the advantages of the biochemical processes for refining of the biomass [85]. Moreover, the use of microorganisms as catalysts in the refining of food industry waste via Figure 3.2. Agro industrial wastes with potential application as substrate in a fungal-based fermentation processes can lead to the production of ingredients for the food industry itself. biorefinery using filamentous ascomycetes and possible value-added products [88]. Thickening or gelling agents (e.g., xanthan, curdlan, and gellan), flavor enhancers (e.g., yeast hydrolysate and monosodium glutamate), flavor compounds (e.g., gamma-decalactone, diacetyl and methyl-ketones), and acidulants (e.g., lactic acid and citric acid) are examples of fermentation-derived ingredients [86]. The number of products which can be obtained from biorefineries is already extensive and will become even greater with further research [87].

In particular, the use of industrial waste materials as an alternative to the use of refined sugars for the production of bioproducts by filamentous fungi is considered a promising approach as these microorganisms produce a cocktail of enzymes able to degrade at a great extent both lignocellulosic and starchy substrates [88]. Some possible agro industrial residues and value- added products in a filamentous fungal-based biorefinery are presented in Figure 3.2.

22 23

Chapter 4

Filamentous fungi

4.1 Fungi and the environment

The Kingdom Fungi (also termed Eumycota) is formed by very diverse organisms, ranging from unicellular yeasts to some of the largest organisms found on Earth. Fungi are eukaryotic, heterotrophic, and osmotrophic organisms with partial extracorporeal digestion, with a diffused, branched, tubular body, and which reproduce by means of spores [89]. Recent studies have divided the Kingdom into eight phyla: Chytridiomycota, Zygomycota, Glomeromycota, Ascomycota, Basidiomycota, Blastocladiomycota, Microsporidia, and Neocallimastigomycota [89]. Moreover, based on their life cycles, fungi can also be divided into three large groups of unicellular, macro filamentous, and multicellular filamentous fungi [90].

Yeasts comprise the classic unicellular fungi. This group includes Saccharomyces cerevisiae, which is possibly one of the most important microbial species in human history. Goddard & Greig (2015) [91] argued that S. cerevisiae lives in the environment as a nomad. Possessing a rich metabolism, S. cerevisiae is able to survive as a generalist at low abundance in a wide range of environments, albeit no especially well. The long lists of conditions in which it can survive and of metabolites it can produce have resulted in the wide use of S. cerevisiae in the food and chemical industries to produce e.g. ethanol, building block chemicals, and oil [90].

Macro-filamentous fungi are generally referred as mushrooms. The visible part, nevertheless, represents just the reproductive structure (fruiting bodies) of organisms that otherwise live as microscopic filaments of cells in substrates such as soil, wood, or living tissues of associated plants. Many mushrooms are edible, have good flavor, and, therefore, have traditionally been part of human nutrition [92, 93]. Ecologically, macro fungi can be classified as pathogens

24 25

(those that attack and kill living tissues and sometimes even the host organism), saprobes (causing the decay of dead organic matter), and mutualists (living in intimate association with plants and interacting in various ways benefiting both partners) [94].

Filamentous fungi comprise a group of eumycetes that have been extensively used in several biotechnological applications for the production of food [5], feed [44], enzymes, organic acids, antibiotics, cholesterol-lowering statins, and biofuels. This group is often considered to be superior to bacteria and yeast with regard to metabolic versatility, robustness, and secretory capacity [95]. In general, they have simple growth requirements and easily adapt to large scale processes [96]. Ecologically, fungi play a critical role, acting in the decomposition of organic material and forming symbiotic relationships with prokaryotes, plants, and animals

[97]. Thus, they are very important for the recycling of organic material in nature [98]. Figure 4.1. Microscopic image of Neurospora intermedia mycelium grown in submerged Morphologically, filamentous fungi are complex beings. Their basic growth structure is a culture using starch as the substrate. tubular filament termed a hypha. As they grow, the hyphae form branches, creating an entangled mass known as mycelium. When grown in submerged fermentation, the shape of mycelium growth varies from dispersed filaments to dense clumps (Figure 4.1). Growth in 4.2 Biomass for food and feed applications pellet-form has been reported as advantageous because it enables repeated-batch fermentation Fungi have been long recognized as a source of palatable food by many societies worldwide and improves the rheology of the broth, resulting in better mass and oxygen transfer in the [104]. Mushrooms and truffles have long been used as food because of their preferred taste reactor with lower energy consumption [99-102]. Additionally, the morphology can affect the [92]. However, out of a large number of species, less than 25 species of mushroom are production of primary and secondary metabolites [103]. The form of growth is determined not accepted as food and few have reached commercial significance [105]. Pleurotus ostreatus, only by the genetic material but also by the chemical and physical conditions used during also known as oyster mushroom, is an example of a mushroom used for food. This species cultivation [101]. constitutes a white rot fungus capable of metabolizing lignin. In nature, it is found living in Among the filamentous fungi, two groups have been the focus in biorefinery studies: parts of plants that are poor in nutrients and vitamins. Therefore, these mushrooms are well ascomycetes and zygomycetes owing to their ability to consume a large variety of carbon and adapted to adverse conditions, rendering them popular for commercial exploitation through nitrogen sources. Therefore, several residues with particular compositions can be used by industrial cultivation using cheap lignocellulose substrates [106, 107]. these fungi as substrates without further supplementation, except for some salts [88, 100]. Fungi can also be used for the production of fermented food. Overall, beneficial aspects related to the fermentation of food include prolonged shelf-life, reduced volume, shorter cooking times, and improved nutritive value compared to those of the non-fermented ingredients [108]. Examples of the use of filamentous fungi in food in Asia include Rhizopus sp. in the preparation of tempeh [109], Aspergillus oryzae in the preparation of hamanatto, miso, and shoyu, and Neurospora intermedia in the preparation of oncom [110]. In Europe, Penicillium roqueforti and Penicillium camemberti are used in the production of blue cheese (e.g., Roquefort, Gorgonzola, Stilton, Danish Blue, and Blue Cheshire) and soft ripened

26 27 cheese (e.g., Camembert and Brie), respectively. The fungi are responsible for providing Thus, substitution of fishmeal with fungal biomass can have an economic impact in fish specific characteristics to the final product such as odor, flavor, texture, and color [110]. farming as fish feed represents a major cost in this activity [121]. Moreover, aquaculture can benefit from the fat composition of R. oryzae, which is comprised of high polyunsaturated In addition to their application in the processing of food, the filamentous fungal biomass can fatty acid content and a large amount of proteins [39]. also be found in the market as an alternative to animal-derived protein sources. Mycoprotein constitutes a high-protein, high-fiber, low-fat food ingredient. The dietary fibers derive from the cell walls and are composed of 2/3 branched 1–3 and 1–6 β-glucans and 1/3 chitin. This 4.3 Rhizopus oryzae fibrous chitin-glucan matrix is typical of fungal mycelium and has low water solubility [111]. The proteins arise from the cytoplasm and contain all nine amino acids essential for the R. oryzae is among the most widely used strains of zygomycetous fungi. It produces a human diet [112]. The lipids, which derive from the cell membranes, are rich in plethora of bio-products with importance for the pharmaceutical, food, and fuel industries polyunsaturated and poor in saturated fatty acids [113, 114]. such as fumaric acid, lactic acid, pectinase, amyloglucosidase, α-amylase, and ethanol [99, 122]. Additionally, R. oryzae is recognized for its ability to produce almost optically pure Inclusion of mycoprotein in the human diet has been demonstrated to improve health in L(+)-lactic acid from starch with low nutritional requirements. This isomer is especially regard to blood cholesterol concentration, lipid profiles, and glycemic response for both important as it is the preferred form used in the production of polylactic acid, a biodegradable healthy and metabolically-compromised individuals [115]. Additionally, it can help in weight- polymer that can substitute for fossil-based plastics [123]. control because of its contribution to satiety, which consequently reduces the energy intake in subsequent meals [114, 116]. Nutritionally, mycoprotein is also a good source of zinc and Growth of R. oryzae in PPL in airlift reactors resulted in a reduction of over 50% of the initial selenium although it affords low content of vitamin B12 and low bioavailable iron [114]. COD of the waste stream [44]. Moreover, fungal biomass containing 49.9% (w/w dry basis) of protein was harvested after 54 h cultivation using an aeration rate of 0.86 vvm (volume of Filamentous fungal biomass can also be used as supplement for animal feed. The interest of air per volume of medium per minute). The morphology of the fungus was also studied in the scientific community in the use of fungi for animal feed originated in the 1960’s, when preliminary shake flask tests. PPL is a complex medium and through different dilution rates it raising awareness of population growth led to the search for alternative food and feed sources. was possible to control the fungal morphology. The cultivation of R. oryzae in 40- and 20- The low prices of more conventional feeds, such as soybean and fishmeal, have discouraged fold-diluted PPL yielded mycelial growth as filaments whereas 10-fold-diluted PPL resulted the entrance of fungal feed into the market. Recently, however, traditional feed sources have in pellet growth, and clumps were obtained by cultivating the microorganism in 5-fold-diluted faced economic and environmental problems [117]. Fishmeal, for example, has been PPL (Figure 4.2). subjected to an increase in price of almost 150% in six years, increasing from 682.00 to 1681.00 USD/ton between 2004 and 2010 [118]. Soybean cultivation, in contrast, has expanded considerably since the 1960’s to meet the speculated global protein demand and is now considered as a primary agent of destruction of ecosystems such as the Amazon rainforest [119, 120]. Consequently, there has been a new resurgence in research on the use of filamentous fungi for animal feed.

Rhizopus oryzae is considered one interesting alternative to fishmeal [100, 112]. With an annual growth of approximately 8%, roughly 50% of the global fish food supply is currently provided through fish farming, which consumes 73% of the fishmeal and 90% of the fish oil globally produced. This leads to over-fishing and, therefore, lack of fish for food [100, 117].

28 29

Figure 4.3. Cultivation of R. oryzae in 10-fold-diluted PPL in an airlift bioreactor. Concentration profiles of glucose (♦); sucrose (▲); fructose (×); ammonia (■); asparagine (●); ethanol (-); lactic acid (Ж); and glycerol (+) at 0.57 vvm aeration rate. Error bars represent ± one standard deviation. pH profile is shown in the right graph as a dashed line [44]. Figure 4.2. Changes in the morphology of R. oryzae cultivated in Erlenmeyer flasks when

varying the PPL dilution rate: (a) 40-fold; (b) 20-fold; (c) 10-fold and; (d) 5-fold [44]. R. oryzae was also examined for cultivation in wheat starch wastewater [61]. This stream

contains large amounts of cellulosic fibers in suspension. After 24 h cultivation, the COD Sugars present in PPL were consumed within 24 h. Thereafter, consumption of ethanol and removal of the clarified streams ED1 and ED2 were less than 50%. No variation in the COD lactic acid was observed. Moreover, the profile of consumption of sugars and pH of the of SS was detected. The harvested biomass grown in the clarified streams contained over 30% medium during cultivation (Figure 4.3) presented similarities: during sugar consumption, the w/w of proteins. The addition of cellulose and amyloglucosidase significantly affected the pH decreased as lactic acid was being produced. During the second part of the cultivation, biomass protein content, reaching values up to 51%. In the SS stream, the high content of following depletion of the sugars, consumption of different compounds resulted in an increase suspended solids, which were used as supports for the biomass growth and collected together of the pH. The results of the experiments in Erlenmeyer flasks (without pH control) also with the biomass, caused the protein content to be in the range of 14–16%. Moreover, in this showed a final pH superior to the initial value. This increase was associated with an increase stream, the addition of enzymes did not significantly change the protein content. in the ammonium concentration in the medium consequent to degradation of the amino acids The profile of the metabolites during cultivation of R. oryzae in ED1 and ED2 media (Figure present in the PPL. However, the measured ammonium concentration in the airlifts did not 4.4) indicated that 24 h cultivation was sufficient for consumption of the glucose, including increase through cultivation, suggesting a higher nitrogen uptake by the fungus in aerobic that released by the action of the enzymes. The higher availability of glucose in the medium conditions. caused by the hydrolysis of the oligo- and polysaccharides led to higher production of ethanol in both streams. Biomass yield (in g/L), conversely, only significantly changed (considering a 95% confidence interval) in the case of the ED2 medium.

30 31

contained 0.31 g of protein per g of biomass and a total lipid concentration of 0.2 g/g. In

A B contrast, the biomass cultivated in the semi-synthetic medium was characterized as having 2 2 0.60 g of protein but only 0.1 g of lipids per g of biomass. The difference may be related to the different nitrogen compositions of the media as microbial growth under nitrogen 1 1 starvation usually leads to accumulation of lipids [124]. In particular, the semi-synthetic medium was prepared with a total nitrogen content of 2.0 g/L, whereas the nitrogen Concentration (g/L) 0 0 concentration was 1.3 g/L in CWFS medium, as determined by the Kjeldahl method (with no 0 6 12 18 24 0 6 12 18 24 supplementation).

16 C 16 D

12 12

8 8 4.4 Aspergillus oryzae

4 4 Aspergillus oryzae constitutes an ascomycetous fungus involved in the production of many

Concentration (g/L) 0 0 value-added products including enzymes such as amylase, protease, lipase, phytase, lactase, 0 6 12 18 24 0 6 12 18 24 Time (h) Time (h) catalase, cellulase, and xylanase. A. oryzae importance can be highlighted by its production of

citric acid, representing the second most important fermentation product after ethanol. Figure 4.4. Concentration profiles of glucose (X), other sugars (■), ethanol (●), and lactic acid Production of citric acid by A. oryzae is preferable to its production by a synthetic route (▲) using R. oryzae: (A) in ED1 medium; (B) in ED2 medium; (C) in ED1 medium with because the starting materials for the latter have a higher cost than that of citric acid [88]. enzyme supplementation; and (D) in ED2 medium with enzyme supplementation. Error bars Moreover, A. oryzae is among the most studied filamentous fungi at the industrial scale and represent ± one standard deviation [61]. one of the three filamentous fungi that dominate the heterologous enzyme market as production hosts (the other two being Aspergillus niger and Trichoderma reesei) [88, 125].

CWFS solution was used for fungal cultivation in a 26-L bubble column bioreactor. On A. oryzae was cultivated in two food industry side streams: wheat starch wastewater and pea- average, 65–72 h was necessary for R. oryzae to completely assimilate the glucose processing byproduct. With the former as substrate, A. oryzae performed better than R. (approximately 20 g/L) and approximately 10 g/L of other sugars present in the CWFS oryzae, removing over 80% of COD after three days for ED1 and ED2 streams [61]. Using medium. In CWFS medium, R. oryzae first consumed glucose. When the glucose level in the the SS stream, the COD removal efficiency after five days was less than 30%. The presence medium was low, consumption of other sugars, e.g. fructose and sucrose, was initiated. of a high content of suspended solids is believed to negatively affect the mass transfer, rendering it difficult for oxygen to be dissolved, yielding an anaerobic environment. The concentration of lactic acid and ethanol remained low during fungal cultivation in the Consequently, a high production of ethanol and low removal of COD were observed. CWFS medium. Moreover, it was observed that the glycerol production was much higher than Moreover, 77.3% of the nitrogen initially present in the SS stream was recovered by the ethanol production in CWFS medium (34.5 g glycerol/g ethanol). The same observation harvesting the biomass using a common kitchen sieve. conducted in a semi-synthetic medium yielded a very different result (0.12 g glycerol/g ethanol). This behavior suggested that different metabolic pathways were used in these media Following cultivation of A. oryzae and R. oryzae in the ED1 and ED2 media in shake flasks, for glycerol production, which is consistent with the obtained composition of the biomasses there was no difference in the biomass yields within a 95% confidence level with regard to the grown in the different media. The biomass of R. oryzae cultivated in CWFS medium biomasses grown in the same sample and upon enzyme supplementation. A. oryzae, however,

32 33 performed better than R. oryzae when comparing the lag phases for both microorganisms. also constitutes an ascomycete whose biomass can be currently found in the markets of When growing A. oryzae in the clarified media (ED1 and ED2), the lag phase of substrate Europe and the USA as a healthy substitute for meat [110, 129]. consumption lasted only 3 h whereas, for R. oryzae, the same lag phase lasted 6 h (Figures 4.4 Biomass production after 36 h is shown in Figure 4.6. M. purpureus produced significantly and 4.5). Comparing the ethanol production of both strains supplemented or not with less biomass than the other strains, except F. venenatum. A. oryzae produced more biomass enzymes, only R. oryzae without enzyme supplementation produced significantly less ethanol. than R. oryzae and N. intermedia. The addition of α-amylase to the medium significantly The ethanol production of A. oryzae was not improved by enzyme addition, reaching values affected only A. oryzae and M. purpureus growth. When comparing the protein content of the not significantly different than those obtained by cultivation of R. oryzae with enzyme harvested biomass, cultivation of A. oryzae resulted in the highest yield of protein per gram of supplementation. pea-protein byproduct substrate, 0.14 g/g. Scaling-up of the process to a bench-scale airlift bioreactor increased the protein yield to 0.26 g/g because of the improved aeration.

3 A 3 B

300 2 2 without α-amylase

250 with α-amylase 1 1 200

Concentration (g/L) 0 0 150 0 6 12 18 24 0 6 12 18 24 100

16 C 16 D Biomass Biomass yield (mg/g) 50 12 12 0 M. purpureus A. oryzae F. venenatum N. intermedia R. oryzae 8 8

4 4 Figure 4.6. Biomass yield (mg dry fungal biomass per gram of pea-processing byproduct)

Concentration (g/L) 0 0 after 36 h of cultivation in airlift bioreactor and 2% w/v medium [5]. 0 6 12 18 24 0 6 12 18 24 Time (h) Time (h) Figure 4.5. Concentration profiles of glucose (X), other sugars (■), ethanol (●), and lactic acid (▲) using A. oryzae: (A) in ED1 medium; (B) in ED2 medium; (C) in ED1 medium with enzyme supplementation; and (D) in ED2 medium with enzyme supplementation [61].

Pea-processing byproduct suspended in water (2% w/v) was used as medium for the cultivation of different filamentous fungi including N. intermedia, A. oryzae, Monascus purpureus, Fusarium venenatum, and R. oryzae [5]. Similar to A. oryzae and N. intermedia, M. purpureus constitutes a strain of ascomycete that is used in oriental cuisine as coloring and flavoring agent in the production of e.g., red yeast rice and rice wine [126-128]. F. venenatum

34 35

Chapter 5

Bioplastics

The nature of the biopolymers present in the fungal biomass has aroused interest regarding whether the biomass could be used for the production of bioplastic films. Bioplastics are considered one of the most promising alternatives to conventional plastics and the hazards they present to the environment. Plastics comprise cheap, lightweight, strong, durable, and corrosion-resistant materials formed by synthetic or semi-synthetic organic polymers. Therefore, they are considered to be the most utilized and versatile material of the modern age [130]. Approximately one third of all the plastic produced annually is designed to be non- reusable and is usually discarded up to 12 months after manufacture [131]. However, the management of plastic waste faces some drawbacks such as recalcitrance toward biodegradation, toxicity after incineration, and massive waste accumulation in landfills and the marine environment [132].

Biopolymers are molecules made of renewable resources, synthesized by microorganisms, or synthesized from petroleum-based chemicals, which can be naturally recycled by biological processes [133]. However, high cost and performance limitations still prevent the complete substitution of plastics made from crude oil [134]. Global production of bioplastics in 2017 reached 2.05 million tons, 58% of which is used in the packaging industry. Other applications with minor strength requirements that also make use of bioplastics include textiles (11% of the total), consumer goods (7%), automotive and transport (7%), agriculture (5%), building and construction (4%), and electrics and electronics (2%) [135].

36 37

5.1 Pectin biofilms subjected to effective mixing at high temperature and pressure (up to several hundred tons per hour) with short processing times [150]. One biopolymer that offers promise as a bioplastic is pectin, which constitutes a polysaccharide naturally found in the cell wall of vascular plants. Pectin is present in Generally, compression molding of sheets is studied to demonstrate material flowability and considerable amounts in the peel of citrus fruit and plays an important role in cell-to-cell fusion prior to extrusion, in order to identify suitable extrusion conditions [151]. Similar to adhesion. Representing the most complex polysaccharide forming the structure of the plant the extrusion method, compression molding is considered to be economically and cell wall, pectin is mainly formed by monomers of α-(1–4)-D-galacturonic acid with some environmentally suitable for the production of rigid materials as it is rapid and requires no methyl ester and L-rhamnose groups and has a negative charge (Figure 5.1). Pectin chains act solvent [152]. Compression molding is performed by placing a material between two hot as a cementing matrix, forming crosslinks with the cellulose microfibrils [136, 137]. In plates and applying a pressure to the material. In general, biopolymers require the addition of particular, the crosslinking ability of pectin has encouraged its use for the production of a plasticizer to control the brittleness of the materials and to lower the shaping temperature biomaterials with applications in food, biomedical, and pharmaceutical industries [138]. [153]. Usually, the plasticizer comprises a small molecule of low volatility. Added to the polymeric material, it modifies the three-dimensional organization, decreases attractive

intermolecular forces, and increases free volume and chain mobility. The results of these changes are increased extensibility, dispensability, and flexibility of the developed film together with decreased film cohesion and rigidity [154].

Citrus pectin was investigated for the production of biofilms using the compression molding technique with glycerol as a plasticizer. Pectin and glycerol were tested at different ratios and subjected to a pressure of 1.33 MPa for 10 min at 120 °C. A 70:30 pectin-glycerol ratio was chosen because, for higher glycerol content, the obtained films were too thin to allow testing of their mechanical properties, whereas for lower glycerol content, the amount of plasticizer was not sufficient to completely wet the pectin.

Figure 5.1. Representative chemical structure of pectin showing typical repeating groups. Reprinted with permission [139].

Numerous studies have reported the use of pectin for the production of bioplastics either by solution casting [140-145] or extrusion methods [146-149]. Solution casting film production involves the preparation of a relatively diluted water solution disposed onto inert surfaces, on which the solution is air dried. Although casting is a practical method at the laboratorial scale and is utilized by a few commercial processes, the method is considered slower and more energy-consuming compared to extrusion [146], in which the material is continuously

Figure 5.2. Optical photo of a pectin/glycerol biofilm produced by compression molding.

38 39

In general, pectin bioplastic has poor mechanical properties and blending with other polymers Table 5.1. Mechanical properties and thickness of films with and without incorporation of is usually recommended [155]. Blending with starch, for example, yields edible films with fungal biomass. good mechanical and oxygen barrier properties [146]. Composites formed by pectin and Pectin Biomass Cultivation Tensile strength Elongation Thickness WVPC -1 -1 -1 protein (e.g., bovine serum albumin, chicken egg albumin, bovine skin gelatin, porcine skin (%) (%) medium (MPa) at break (%) (mm) (kg.s .m .Pa ) 70 0 - 15.7 ± 0.53 5.5 ± 1.68 1.03 6.92 × 10-13 gelatin, fish skin gelatin, and soybean flour protein) exhibit improved storage moduli and loss CWFS 15.9 ± 0.84 3.5 ± 1.57 0.42 1.68 × 10-13 moduli, and better moisture resistance compared to those of pectin films [156]. 60 10 Semi-synthetic 19.3 ± 3.5 1.6 ± 0.88 0.41 3.5 × 10-13 CWFS 7.6 ± 0.64 3.8 ± 1.12 0.35 7.22 × 10-13 50 20 Semi-synthetic 16.1 ± 2.3 4.3 ± 2.71 0.38 3.56 × 10-13 5.2 Biomass for bioplastic production CWFS 6.5 ± 2.01 1.4 ± 0.39 0.25 1.35 × 10-13 40 30 Semi-synthetic 11.2 ± 2.12 4.5 ± 0.98 0.32 3.58 × 10-13 Pectin was blended with lyophilized and milled fungal biomass cultivated either in the CWFS as described in section 4.3 or in semi-synthetic medium. The different media resulted in biomasses containing low protein and high fat content and high protein and low fat content, Scanning electron microscopic (SEM) images of the films showed that the incorporation of respectively. The films were prepared using the same parameters as the pectin-glycerol films. biomass improved the homogeneity of the films by reducing the amount of pectin debris on In average, addition of biomass resulted in reduction of the thickness of the films as well as the surface of the films as well as reducing the gaps observed in the cross-sectional area of the the elongation at break (Table 5.1). Overall, the incorporation of protein-rich fungal biomass films (Figure 5.3). In comparison, agglomerates present in the biomass-containing films yielded films with increased tensile strength compared to those from the incorporation of suggested that some of the biomass components were incompatible with the pectin, leading to protein-poor biomass. It has been suggested that the amide bonds of the proteins and the the formation of these heterogeneous structures. carboxylic group of pectin molecules form complex interactions that improve the strength of the films [157]. The addition of biomass to the pectin-glycerol matrix also improved the impermeability of the films to water vapor. In particular, tests of water vapor permeability coefficient (WVPC) showed that the films containing fungal biomass exhibited lower values of WVPC than those of the pectin-glycerol film.

40 41

Chapter 6

Circular economy approach

A major societal challenge is to meet the increasing demand of food caused by the growing global population while minimizing the associated environmental impact [28]. The whole food supply chain (from agriculture through industrial manufacturing and processing, retail, and household) generates environmental impacts in various aspects, such as greenhouse gas emission and eutrophication [84, 158]. A possible solution to this problem is to improve the management of food waste, as approximately 40% of the food intended for human consumption ends up as waste at different stages of the food system. This represents a substantial loss of productivity, energy, and natural resources [158].

Figure 5.3. SEM images of the surface (left) and cross-section (right) of the pectin-based In this context, growing attention has been paid to the new concept and development model of bioplastic films. A) Pectin film without biomass; B) pectin film with 20% protein-rich circular economy [159]. In the circular economy, what was once considered as waste becomes biomass; C) pectin film with 20% protein-poor biomass (Magnification: surface = 1000X, instead a resource. The concept of circular economy involves reuse, repair, refurbishing, and cross-section = 1500X). recycling of existing materials and products [158]. In particular, the primary goal of a circular economy is to minimize waste and excessive resource utilization by turning goods at the end of their lifespan, along with waste generated during the manufacturing and use of goods, into resources for the manufacturing of other products [16]. In contrast, the dominant current economic development model, known as the linear model, focuses mainly on the efficient allocation of resources in the market but fails to provide analytical tools that take into account the limitations and exhaustion of natural resources [159]. With respect to the food system, circular economy involves reduction of the amount of waste generated, re-use of food, utilization of by-products and food waste, nutrient recycling, and changes in diet toward more diverse and more efficient food profiles [158].

42 43

For example, 77% of the 1.3 g/L of nitrogen lost in wheat starch wastewater [61] can be EVAPORATION SCENARIO Steam recovered through the cultivation of filamentous fungi in the form of edible fungal biomass.

Both tested strains (A. oryzae and R. oryzae) were effective in the recovery of the nitrogen Heat Treated Potato Liquor Pump Heat Potato Protein present in the SS albeit at different conditions: the best result for A. oryzae was obtained at pH 19.64 ton/h exchanger Liquor Evaporator To farmers

3.85 whereas R. oryzae yielded its best performance at pH 5.5. Both, however, benefited from Storage tank supplementation of the medium with cellulase and amyloglucosidase enzymes. These

FUNGUS SCENARIO enzymes released sugars by hydrolyzing the cellulose and starch polymers, therefore reducing Air + CO2 Wet air the amount of suspended solids in the medium. These suspended solids otherwise would have HTPL 19.64 ton/h Filter been encompassed by the fungal mycelium rather than being consumed. Thus, the fungi Enzyme Biomass efficiently converted the extra sugar into fungal biomass. Defoamer Dryer Airlift bioractor It must be noted, however, that circular economy does not just represent an approach for more Pumps Air Air appropriate waste management. This can be considered a very limiting point of view that may Air blower Air sterilizer Wastewater cause the circular economy to fail. For example, in some contexts, certain recycling, reuse or Compressor recovery options may not be appropriate, whereas in other situations they may comprise the BIOGAS SCENARIO best option. Additionally, some biotechnological alternatives may be much more expensive Biogas and impactful than the conventional technology. Therefore, circular economy should be Condenser Compressor viewed as a new economic pattern that aims to increase the efficiency of production and HTPL 19.64 ton/h consumption through the appropriate use, reuse, and exchange of resources. Pump Anaerobic Storage Wastewater digestor tank

Figure 6.1. Process flow diagram describing the three scenarios studied for the 6.1 Economic evaluation of alternative biotreatment of industrial waste management of potato liquor [39]. A techno-economic analysis of the treatment and use of the wastewater of the potato starch manufacturing was performed considering three scenarios. The current strategy of PPL production was compared with the use of HTPL for i) the cultivation of filamentous fungus R. The technical analysis indicated a reduction of 90.8% in the energy consumption when oryzae for the production of biomass or ii) the production of biogas [39]. The scenarios are shifting the waste management from the current scenario to the fungal alternative. Anaerobic presented in Figure 6.1. PPL was obtained by evaporation of HTPL, increasing the total solids digestion for production of biogas consumed only 4.5% of the energy currently required. from approximately 3% to 40% (w/v). Previously it was determined that pellet-growth of R. Moreover, the production of R. oryzae biomass reduced the initial cost by more than half oryzae in PPL requires 10-fold dilution; i.e., with a composition similar to that of HTPL. compared to that required for the production of PPL. The initial capital cost dropped from Therefore, the economic analysis of the fungal cultivation was performed using HTPL as the €16.5 million to €7.5 million. Biogas production capital cost was predicted to be €14.2 substrate. million.

Comparing the Net Present Value (NPV) of the three scenarios (Figure 6.2), fungal cultivation resulted in a NPV that was less negative than that of the biogas scenario. Considering a

44 45 project lifetime of 15 years, the NPV of the biogas scenario became similar to the NPV of the programs and public campaigns as well as seminars [161]. However, the social aspects of evaporation scenario and, considering 20 years of lifetime for the plant, the NPV of the circular economy appear to be attracting only little attention. The gains in this area are usually evaporation scenario was less negative than that of the fungus scenario. This was caused by implicit and the most cited benefit is the associated job creation. In contrast, the subjective the large capital cost (€16.5 million) and low operational cost (€1.7 million per operating contribution of the circular economy for general well-being is not as clear [162, 163]. For period) of the evaporation scenario, as opposed to the low capital cost (€7.5 million) and large example, the new businesses and job opportunities created by the circular economic model operational cost (€1.84 million per operating period) of the fungus scenario. The contrasting might be focused on industries and academia, especially those in the agro, food, chemical, characteristics of the scenarios led to a shift in the NPV when increasing the plant’s lifetime healthcare, pharma, and logistics sectors. However, other consequences of the circular from 15 to 20 years. economy include the promotion of economic growth by saving material costs, dampened price volatility, and improved security of supplies [160].

Food waste has experienced a quick ascendancy in its consideration through public policy. -35 The statistics of food waste are shocking in themselves and an escalating urgency is observed -30 when issues such as climate change, peak oil, and global food security are considered [164]. -25 The use of food waste as a resource for the production of new products, including more food, -20 can contribute to solving the growing pressure of the food security problem. Furthermore, it -15 raises questions regarding the amount of food wasted that could otherwise have been used to -10 feed people [165]. Therefore, the first problem to address in terms of food waste is the -5 reduction and prevention thereof. This attitude, together with recovery of the nutrients of Net Present Value (million Euro)Value (million Net Present 0 10 15 20 unavoidable food waste, might markedly increase food security. Years Evaporation Scenario Fungus Scenario Biogas Scenario In addition, the production of animal protein demands grains that, in another scenario, could

be used directly for humans. Moreover, this process of converting plant protein into animal Figure 6.2. Net present value for the different scenarios considering different project life-times [39]. protein is considered to be inefficient in terms of environmental resources. Filamentous fungal biomass produced from agro industrial waste and used as feed is presented as an alternative to the use of grains for animal feed. As human food, mycoprotein, besides addressing the

6.2 Social and ethical aspects environmental issues associated with the food security debate, also addresses the ethical concerns related to the cruelty of certain treatments applied to animals. Furthermore, meat- In the light of the previous considerations, the movement from linear to circular economy alternative products offer the potential to positively affect some public health issues. For appears inevitable. This movement can be facilitated by the promotion by biorefineries of the example, some studies have associated red meat consumption with the risk of developing switch from the consumption of fossil reserves to that of renewable resources [160]. The full diseases such as cardiovascular disease, type II diabetes, and colorectal and other carcinomas. implementation of circular economy involves long-term socio-technical transformations, Furthermore, millions of episodes of illness are caused annually by food-borne pathogens which includes changes in infrastructure and technologies along with citizen competences, found in meat such as Salmonella, Campylobacter, and Escherichia coli [166, 167]. practices, and world-views [158]. In this aspect, social awareness is vital for the successful implementation of the circular economy, as citizens (customers) comprise a central part of the economy. Progress toward social awareness is ongoing through the utilization of educational

46 47

Chapter 7

Concluding remarks and future work

7.1 Concluding remarks

In this thesis, the discarding of resources by the food industry was addressed using filamentous fungi as biocatalysts for the conversion of nutrients into fungal biomass, with application in the food, feed, and bioplastic production industries. Different byproducts of the food and drink industry with diverse characteristics were studied. The major results are as follows:

 Environmental impact of waste streams could be reduced by providing new applications for the waste: removal of half of the initial COD load of PPL and over 80% of the initial COD of clarified wheat starch wastewater was achieved by fungal cultivation;

 Economic analysis suggested a reduction in the initial capital cost: compared to the traditional method used in the potato starch industry, cultivation of filamentous fungus for the production of biomass might afford a reduction of over half of the initial cost. Moreover, only 9.2% of the energy used in the current scenario would be required to produce the fungal biomass.

 Fungi could return the nutrients of the waste back to the food chain: up to 77% of the nitrogen present in the solid-rich stream of the wheat starch wastewater was recovered in the form of mycoprotein at the conclusion of filamentous fungi cultivation. Ease of collection of the recovered nitrogen, via a simple sieving process, was also notable.

 Fungi could improve low-quality food processing side-streams: a mycoprotein concentrate could be produced using the pea-processing byproduct, which exhibits poor functional properties for the direct use in food. Fungal biomass with approximately 50% protein

48 49

content was obtained from the pea-processing byproduct using edible strains of filamentous Acknowledgements fungi.

 Citrus waste represented a potential substrate for the production of single cell oils: fungal biomass containing 20% of lipids was obtained upon cultivation in a nutrient solution This thesis would not have been possible without the help and support of many people to extracted from citrus waste. whom I wish to express my most sincere gratitude.

 Fungal biomass improved the characteristics of pectin films: compression molding was Firstly, I would like to thank my supervisor, Prof. Mohammad Taherzadeh, for his generous successfully used to produce pectin-based bioplastics, yielding films with promising advice and support during these four years. Thank you for your availability, discussions, mechanical characteristics. Moreover, the incorporation of fungal biomass improved the suggestions, and encouragement. Your guidance was most crucial to the development of this structural and mechanical properties of the pectin films. thesis.

I would also like to thank my co-supervisor, Dr. Akram Zamani, for all the help and comments. They were very valuable for this thesis and for my research journey. 7.2 Future work I also want to address my thanks to my examiner, Prof. Kim Bolton, for his availability and To assess the potential of filamentous fungi for the management of co-products of the food support. and drink industry, some areas appear to be worthy of further exploration. I extend my acknowledgement to the Directors of Studies Dr. Tomas Wahnström and Dr. Dan  Scaling up the cultivation processes, including operation in a continuous mode. Åkesson, and the Head of the Department, Dr. Peter Therning.  Production of lipids by R. oryzae during growth in a nutrient solution extracted from citrus Many thanks are also offered to the laboratory managers Marlén Kilberg, Kristina Laurila, waste might be optimized by varying operational parameters. The lipids should be tested for Haike Hilke, Thomas Södergren, and Sofie Svensson for their constant effort to maintain a use as food or feed supplements or for biodiesel production. good working environment.  The effect of the biomass components in the pectin biofilms should be investigated in I express my sincere gratitude to the staff of the University of Borås, especially Sari Sarhamo, further detail, including the utilization of other bioplastic production techniques. Susanne Borg, Jonas Edberg, Irene Lammassaari, and Louise Holmgren for their assistance

 Recovery by fungi of other nutrients that also have marked environmental impact, such as with all the administrative duties. phosphorus, needs to be investigated. I would also like to manifest my gratitude to Dr. Patrik Lennartsson, Dr. Ilona Sárvári Horváth, Dr. Anita Pettersson and Dr. Magnus Lundin for everything they have taught me.

My heartfelt thanks to my fellow doctoral student of the Resource Recovery group: Steven, Veronika, Amir, Rebecca, Lukitawesa, Konstantinos, Anette, Madumita, Eboh, Mostafa, Andreas, Supriyanto, David, Adib, Johan, Sabina, Moein, and Sajjad. My acknowledgements are extended to the visiting Ph.Ds from other institutions who worked in our group during certain periods: Mohsen, Zohre, Forough, Behzad, Sindor and Tomás.

50 51

I would also like to acknowledge the Ph.Ds that have been part of the Resource Recovery References group and with whom I have worked together: Jorge, Julius, Regina, Alex, Ram, Farzad, Abas, Kehinde, Sunil, Fatimat, Behnaz, Päivi, Swarnima, Sunanda, Mofoluwake, and Kamram. 1. C. R. Kagan, At the nexus of food security and safety: opportunities for nanoscience Many thanks also to former M.Sc. students Eshna, Vamsi, Rahul, Gülru, Mohsen, and Dahn, and nanotechnology. 2016, ACS Publications. and B.Sc. student Sam. 2. FAO, IFAD, and WFP, The State of Food Insecurity in the World 2015. Meeting the 2015 international hunger targets: taking stock of uneven progress. 2015, FAO: Special thanks to M.Sc. Rajesh and B.Sc. Frida, Karolin, and Sara. Our collaboration was of Rome. indescribable value. 3. M. A. Augustin, M. Riley, R. Stockmann, L. Bennett, A. Kahl, T. Lockett, M. I would like to extend many thanks to Pedro and Ann for their friendship and all the fun we Osmond, P. Sanguansri, W. Stonehouse, I. Zajac, and L. Cobiac, Role of food have had in the last several years. processing in food and nutrition security. Trends in Food Science & Technology, 2016. 56: p. 115-125. Lastly, I want to express my gratitude in this thesis to my parents. Their love and support kept 4. FAO, Social network analysis for territorial assessment and mapping of Food Security me motivated to pursue this Ph.D. Thank you. and Nutrition Systems (FSNS) - A methodological approach. 2018, FAO: Rome. 5. P. F. Souza Filho, R. B. Nair, D. Andersson, P. R. Lennartsson, and M. J. Taherzadeh, Vegan-mycoprotein concentrate from pea-processing industry byproduct using edible filamentous fungi. Fungal Biology and Biotechnology, 2018. 5(1): p. 5. 6. L. A. Garibaldi, B. Gemmill-Herren, R. D’Annolfo, B. E. Graeub, S. A. Cunningham, and T. D. Breeze, Farming Approaches for Greater Biodiversity, Livelihoods, and Food Security. Trends in Ecology & Evolution, 2017. 32(1): p. 68-80. 7. T. Garnett, Three perspectives on sustainable food security: efficiency, demand restraint, food system transformation. What role for life cycle assessment? Journal of Cleaner Production, 2014. 73: p. 10-18. 8. C. A. Monteiro, J. C. Moubarac, G. Cannon, S. W. Ng, and B. Popkin, Ultra

processed products are becoming dominant in the global food system. Obesity‐ Reviews, 2013. 14(S2): p. 21-28. 9. E. Comission, Food and drink industry - Growth - European Commission. 2018. 10. D. Somsen and A. Capelle, Introduction to production yield analysis—a new tool for improvement of raw material yield. Trends in Food Science & Technology, 2002. 13(4): p. 136-145. 11. K. Waldron, Waste minimization, management and co-product recovery in food processing: an introduction, in Handbook of waste management and co-product

52 53

recovery in food processing, K. Waldron, Editor. 2007, Woodhead Publishing 23. A. José Miguel, Food engineering into the XXI century. AIChE Journal, 2018. 64(1): Limited: Cambridge, UK. p. 2-11. 12. E. C.-E. Commission, Directive 2008/98/EC of the European Parliament and of the 24. M. Nestle, Food politics: How the food industry influences nutrition and health. Vol. Council of 19 November 2008 on waste. Official Journal of the European Union L, 3. 2013: Univ of California Press. 2008. 312(13): p. 22.11. 25. FoodDrinkEurope. Data & Trends of the European Food and Drink Industry 2017. 13. M. Eriksson, I. Strid, and P.-A. Hansson, Carbon footprint of food waste management 2017 [cited 06/01/2018]; Available from: options in the waste hierarchy – a Swedish case study. Journal of Cleaner Production, http://www.fooddrinkeurope.eu/publication/data-trends-of-the-european-food-and- 2015. 93: p. 115-125. drink-industry-2017/. 14. K. S. H.-Y. Hashim, A. H. Mohamed, and H. Z. M. S. Redza, Developing a waste 26. E. F. Beitzen-Heineke, N. Balta-Ozkan, and H. Reefke, The prospects of zero- minimization awareness model through community based movement: A case study of packaging grocery stores to improve the social and environmental impacts of the food the IIUM Green Team. Geografia-Malaysian Journal of Society and Space, 2017. 8(5). supply chain. Journal of Cleaner Production, 2017. 140: p. 1528-1541. 15. S. Van Ewijk and J. A. Stegemann, Limitations of the waste hierarchy for achieving 27. G. Garcia-Garcia, E. Woolley, S. Rahimifard, J. Colwill, R. White, and L. Needham, A absolute reductions in material throughput. Journal of Cleaner Production, 2016. 132: Methodology for Sustainable Management of Food Waste. Waste and Biomass p. 122-128. Valorization, 2016. 16. C. Ingrao, N. Faccilongo, L. Di Gioia, and A. Messineo, Food waste recovery into 28. R. Ravindran and A. K. Jaiswal, Exploitation of Food Industry Waste for High-Value energy in a circular economy perspective: A comprehensive review of aspects related Products. Trends in Biotechnology, 2016. 34(1): p. 58-69. to plant operation and environmental assessment. Journal of Cleaner Production, 29. S. Arvanitoyannis Ioannis and D. Ladas, Meat waste treatment methods and potential 2018. 184: p. 869-892. uses. International Journal of Food Science & Technology, 2008. 43(3): p. 543-559. 17. M. Eriksson, R. Ghosh, L. Mattsson, and A. Ismatov, Take-back agreements in the 30. B. Mahro and M. Timm, Potential of Biowaste from the Food Industry as a Biomass perspective of food waste generation at the supplier-retailer interface. Resources, Resource. Engineering in Life Sciences, 2007. 7(5): p. 457-468. Conservation and Recycling, 2017. 122: p. 83-93. 31. L. A. Pfaltzgraff, E. C. Cooper, V. Budarin, and J. H. Clark, Food waste biomass: a 18. D. Somsen, A. Capelle, and J. Tramper, Production yield analysis—a new systematic resource for high-value chemicals. Green Chemistry, 2013. 15(2): p. 307-314. method for improvement of raw material yield. Trends in Food Science & Technology, 32. S. Corrado, F. Ardente, S. Sala, and E. Saouter, Modelling of food loss within life cycle 2004. 15(5): p. 267-275. assessment: From current practice towards a systematisation. Journal of Cleaner 19. E. C.-E. Commission, Directive 2010/75/EU of the European Parliament and of the Production, 2017. 140: p. 847-859. Council of 24 November 2010 on industrial emissions (integrated pollution prevention 33. M. S. Benhabiles, N. Abdi, N. Drouiche, H. Lounici, A. Pauss, M. F. A. Goosen, and and control). OJ EU, L, 2010. 334(17.12): p. 2010. N. Mameri, Fish protein hydrolysate production from sardine solid waste by crude 20. A. Stefan and E. Lars, Regulation via the Polluter pays Principle. The Economic pepsin enzymatic hydrolysis in a bioreactor coupled to an ultrafiltration unit.

Journal, 2016. 126(593): p. 884-906. ‐ Materials Science and Engineering: C, 2012. 32(4): p. 922-928. 21. P. Glavič and R. Lukman, Review of sustainability terms and their definitions. Journal 34. FAOSTAT. [cited 03/16/2018]; Available from: of cleaner production, 2007. 15(18): p. 1875-1885. http://www.fao.org/faostat/en/#data/QC. 22. M. Nasr, A. Tawfik, M. Suzuki, and S. Ookawara, Mathematical modeling of bio- 35. J. E. Bradshaw and G. Ramsay, Potato origin and production. Advances in potato hydrogen production from starch wastewater via up-flow anaerobic staged reactor. chemistry and technology, 2009: p. 1-26. Desalination and Water Treatment, 2015. 54(1): p. 50-58.

54 55

36. J. J. G. van Soest, H. Tournois, D. de Wit, and J. F. G. Vliegenthart, Short-range 49. J. Bárta, V. Heřmanová, and J. Diviš, Effect of low-molecular additives on structure in (partially) crystalline potato starch determined with attenuated total precipitation of potato fruit juice proteins under different temperature regimes. reflectance Fourier-transform IR spectroscopy. Carbohydrate Research, 1995. 279: p. Journal of Food Process Engineering, 2008. 31(4): p. 533-547. 201-214. 50. M. Vikelouda and V. Kiosseoglou, The use of carboxymethylcellulose to recover 37. Starch.eu. http://www.starch.eu/. 2015 [cited 2015 20/10/2015]. potato proteins and control their functional properties. Food Hydrocolloids, 2004. 38. A. Laval, Solutions for Processing Starch, A. L. C. AB, Editor. 2016: Sweden. 18(1): p. 21-27. 39. P. Souza Filho, P. Brancoli, K. Bolton, A. Zamani, and M. Taherzadeh, Techno- 51. V. Bártová and J. A. N. Bárta, Chemical composition and nutritional value of protein Economic and Life Cycle Assessment of Wastewater Management from Potato Starch concentrates isolated from potato (Solanum tuberosum L.) fruit juice by precipitation Production: Present Status and Alternative Biotreatments. Fermentation, 2017. 3(4): with ethanol or ferric chloride. Journal of Agricultural and Food Chemistry, 2009. p. 56. 57(19): p. 9028-9034. 40. C. Kamnerdpetch, M. Weiss, C. Kasper, and T. Scheper, An improvement of potato 52. G. Lisinska and W. Leszczynski, Potato science and technology. 1989: Springer pulp protein hydrolyzation process by the combination of protease enzyme systems. Science & Business Media. Enzyme and Microbial Technology, 2007. 40(4): p. 508-514. 53. U. Klingspohn, P. Vijai Papsupuleti, and K. Schügerl, Production of enzymes from 41. F. Mayer, Potato pulp: properties, physical modification and applications. Polymer potato pulp using batch operation of a bioreactor. Journal of chemical technology and Degradation and Stability, 1998. 59(1): p. 231-235. biotechnology, 1993. 58(1): p. 19-25. 42. H. N. Lærke, A. S. Meyer, K. V. Kaack, and T. Larsen, Soluble fiber extracted from 54. A. Sluiter, B. Hames, R. Ruiz, C. Scarlata, J. Sluiter, and D. Templeton, potato pulp is highly fermentable but has no effect on risk markers of diabetes and Determination of sugars, byproducts, and degradation products in liquid fraction cardiovascular disease in Goto-Kakizaki rats. Nutrition Research, 2007. 27(3): p. 152- process samples. National Renewable Energy Laboratory, Golden, CO, 2006. 160. 55. FAO, Save and Grow in practice: maize, rice, wheat - A guide to sustainable cereal 43. J.-S. Yang, T.-H. Mu, and M.-M. Ma, Extraction, structure, and emulsifying production. 2016: Rome. p. 124. properties of pectin from potato pulp. Food Chemistry, 2018. 244: p. 197-205. 56. F. Fava, G. Zanaroli, L. Vannini, E. Guerzoni, A. Bordoni, D. Viaggi, J. Robertson, K. 44. P. F. Souza Filho, A. Zamani, and M. J. Taherzadeh, Production of Edible Fungi from Waldron, C. Bald, and A. Esturo, New advances in the integrated management of food Potato Protein Liquor (PPL) in Airlift Bioreactor. Fermentation, 2017. 3(1): p. 12. processing by-products in Europe: sustainable exploitation of fruit and cereal 45. G. A. van Koningsveld, H. Gruppen, H. H. J. de Jongh, G. Wijngaards, M. A. J. S. van processing by-products with the production of new food products (NAMASTE EU). Boekel, P. Walstra, and A. G. J. Voragen, Effects of pH and Heat Treatments on the New biotechnology, 2013. 30(6): p. 647-655. Structure and Solubility of Potato Proteins in Different Preparations. Journal of 57. A. Van Der Borght, H. Goesaert, W. S. Veraverbeke, and J. A. Delcour, Fractionation Agricultural and Food Chemistry, 2001. 49(10): p. 4889-4897. of wheat and wheat flour into starch and gluten: overview of the main processes and 46. A. Waglay, S. Karboune, and I. Alli, Potato protein isolates: Recovery and the factors involved. Journal of Cereal Science, 2005. 41(3): p. 221-237. characterization of their properties. Food Chemistry, 2014. 142: p. 373-382. 58. S.-y. Pu, L.-l. Qin, J.-p. Che, B.-r. Zhang, and M. Xu, Preparation and application of 47. K. Schügerl, Agriculture wastes. A source of bulk products? Chemical engineering & a novel bioflocculant by two strains of Rhizopus sp. using potato starch wastewater as technology, 1994. 17(5): p. 291-300. nutrilite. Bioresource technology, 2014. 162: p. 184-191. 48. U. Klingspohn, J. Bader, B. Kruse, P. V. Kishore, K. Schuegerl, H. A. Kracke-Helm, 59. D. Yu, C. Li, L. Wang, J. Zhang, J. Liu, and Y. Wei, Multiple effects of trace elements and Z. Likidis, Utilization of potato pulp from potato starch processing. Process on methanogenesis in a two-phase anaerobic membrane bioreactor treating starch Biochemistry, 1993. 28(2): p. 91-98. wastewater. Applied microbiology and biotechnology, 2016: p. 1-12.

56 57

60. S. Deng, R. Bai, X. Hu, and Q. Luo, Characteristics of a bioflocculant produced by 70. R. Wikandari, H. Nguyen, R. Millati, C. Niklasson, and M. J. Taherzadeh, Bacillus mucilaginosus and its use in starch wastewater treatment. Applied Improvement of Biogas Production from Orange Peel Waste by Leaching of microbiology and biotechnology, 2003. 60(5): p. 588-593. Limonene. BioMed Research International, 2015. 2015: p. 6. 61. P. F. Souza Filho, A. Zamani, and M. J. Taherzadeh, Edible Protein Production by 71. V. Bátori, M. Jabbari, D. Åkesson, P. R. Lennartsson, M. J. Taherzadeh, and A. Filamentous Fungi using Starch Plant Wastewater. Waste and Biomass Valorization, Zamani, Production of Pectin-Cellulose Biofilms: A New Approach for Citrus Waste 2018. Recycling. International Journal of Polymer Science, 2017. 2017: p. 9. 62. K. Shevkani, N. Singh, A. Kaur, and J. C. Rana, Structural and functional 72. M. Boukroufa, C. Boutekedjiret, L. Petigny, N. Rakotomanomana, and F. Chemat, characterization of kidney bean and field pea protein isolates: A comparative study. Bio-refinery of orange peels waste: A new concept based on integrated green and Food Hydrocolloids, 2015. 43: p. 679-689. solvent free extraction processes using ultrasound and microwave techniques to 63. A. K. Stone, N. A. Avarmenko, T. D. Warkentin, and M. T. Nickerson, Functional obtain essential oil, polyphenols and pectin. Ultrasonics Sonochemistry, 2015. 24: p. properties of protein isolates from different pea cultivars. Food Science and 72-79. Biotechnology, 2015. 24(3): p. 827-833. 73. R. Ciriminna, N. Chavarría Hernández, A. Inés Rodríguez Hernández, and M.

64. Y. Shimomura, T. Murakami, N. Nakai, M. Nagasaki, and R. A. Harris, Exercise Pagliaro, Pectin: A new perspective‐ from the biorefinery standpoint. Biofuels, Promotes BCAA Catabolism: Effects of BCAA Supplementation on Skeletal Muscle Bioproducts and Biorefining, 2015. 9(4): p. 368-377. during Exercise. The Journal of Nutrition, 2004. 134(6): p. 1583S-1587S. 74. B. Satari, K. Karimi, M. J. Taherzadeh, and A. Zamani, Co-production of fungal 65. N. Babault, C. Païzis, G. Deley, L. Guérin-Deremaux, M.-H. Saniez, C. Lefranc- biomass derived constituents and ethanol from citrus wastes free sugars without Millot, and F. A. Allaert, Pea proteins oral supplementation promotes muscle auxiliary nutrients in airlift bioreactor. International journal of molecular sciences, thickness gains during resistance training: a double-blind, randomized, Placebo- 2016. 17(3): p. 302. controlled clinical trial vs. Whey protein. Journal of the International Society of Sports 75. F. Cherubini, The biorefinery concept: using biomass instead of oil for producing Nutrition, 2015. 12(1): p. 3. energy and chemicals. Energy Conversion and Management, 2010. 51(7): p. 1412- 66. L. Day, Proteins from land plants – Potential resources for human nutrition and food 1421. security. Trends in Food Science & Technology, 2013. 32(1): p. 25-42. 76. G. C. Hegerl, F. W. Zwiers, P. Braconnot, N. P. Gillett, Y. Luo, J. Marengo, N. 67. J. I. Boye, S. Aksay, S. Roufik, S. Ribéreau, M. Mondor, E. Farnworth, and S. H. Nicholls, J. E. Penner, and P. A. Stott, Understanding and attributing climate change. Rajamohamed, Comparison of the functional properties of pea, chickpea and lentil Climate Change 2007: The Physical Science Basis. Working Group 1 Contribution to protein concentrates processed using ultrafiltration and isoelectric precipitation the Fourth Assessment Report of the Intergovernmental Panell on Climate Change techniques. Food Research International, 2010. 43(2): p. 537-546. (IPCC) Volume Chapter, 2007. 9. 68. N. A. McCarthy, D. Kennedy, S. A. Hogan, P. M. Kelly, K. Thapa, K. M. Murphy, 77. M. Gavrilescu, V. K. Gupta, M. G. Tuohy, C. P. Kubicek, J. Saddler, and F. Xu, and M. A. Fenelon, Emulsification properties of pea protein isolate using Chapter 14 - Biorefinery Systems: An Overview, in Bioenergy Research: Advances homogenization, microfluidization and ultrasonication. Food Research International, and Applications. 2014, Elsevier: Amsterdam. p. 219-241. 2016. 89, Part 1: p. 415-421. 78. S. V. Mohan, G. N. Nikhil, P. Chiranjeevi, C. N. Reddy, M. V. Rohit, A. N. Kumar, 69. W. S. Ratnayake, R. Hoover, and T. Warkentin, Pea starch: composition, structure and O. Sarkar, Waste biorefinery models towards sustainable circular bioeconomy: and properties—a review. Starch Stärke, 2002. 54(6): p. 217-234. critical review and future perspectives. Bioresource technology, 2016.

‐ 79. G. Rødsrud, M. Lersch, and A. Sjöde, History and future of world's most advanced biorefinery in operation. Biomass and Bioenergy, 2012. 46: p. 46-59.

58 59

80. D. K. S. Ng, Automated targeting for the synthesis of an integrated biorefinery. Hendriks, The future supply of animal-derived protein for human consumption. Trends Chemical Engineering Journal, 2010. 162(1): p. 67-74. in Food Science & Technology, 2013. 29(1): p. 62-73. 81. A. Yadav and V. K. Garg, Feasibility of nutrient recovery from industrial sludge by 93. M. J. Sadler, Meat alternatives—market developments and health benefits. Trends in vermicomposting technology. Journal of Hazardous Materials, 2009. 168(1): p. 262- Food Science & Technology, 2004. 15(5): p. 250-260. 268. 94. R. Molina, T. O'Dell, D. Luoma, M. Amaranthus, M. Castellano, and K. Russell, 82. R. Parajuli, T. Dalgaard, U. Jørgensen, A. P. S. Adamsen, M. T. Knudsen, M. Birkved, Biology, ecology, and social aspects of wild edible mushrooms in the forests of the M. Gylling, and J. K. Schjørring, Biorefining in the prevailing energy and materials Pacific Northwest: a preface to managing commercial harvest. Gen. Tech. Rep. PNW- crisis: a review of sustainable pathways for biorefinery value chains and sustainability GTR-309. Portland, OR: US Department of Agriculture, Forest Service, Pacific assessment methodologies. Renewable and Sustainable Energy Reviews, 2015. 43: p. Northwest Research Station. 42 p, 1993. 309. 244-263. 95. V. Meyer, M. R. Andersen, A. A. Brakhage, G. H. Braus, M. X. Caddick, T. C. 83. P. McKendry, Energy production from biomass (part 2): conversion technologies. Cairns, R. P. de Vries, T. Haarmann, K. Hansen, C. Hertz-Fowler, S. Krappmann, U. Bioresource technology, 2002. 83(1): p. 47-54. H. Mortensen, M. A. Peñalva, A. F. J. Ram, and R. M. Head, Current challenges of 84. N. Mirabella, V. Castellani, and S. Sala, Current options for the valorization of food research on filamentous fungi in relation to human welfare and a sustainable bio- manufacturing waste: a review. Journal of Cleaner Production, 2014. 65: p. 28-41. economy: a white paper. Fungal Biology and Biotechnology, 2016. 3(1): p. 6. 85. V. L. Budarin, P. S. Shuttleworth, J. R. Dodson, A. J. Hunt, B. Lanigan, R. Marriott, 96. F. Alberti, G. D. Foster, and A. M. Bailey, Natural products from filamentous fungi K. J. Milkowski, A. J. Wilson, S. W. Breeden, and J. Fan, Use of green chemical and production by heterologous expression. Applied Microbiology and technologies in an integrated biorefinery. Energy & Environmental Science, 2011. Biotechnology, 2017. 101(2): p. 493-500. 4(2): p. 471-479. 97. J. E. Galagan, M. R. Henn, L.-J. Ma, C. A. Cuomo, and B. Birren, Genomics of the 86. L. Dufossé, M. Fouillaud, Y. Caro, S. A. S. Mapari, and N. Sutthiwong, Filamentous fungal kingdom: Insights into eukaryotic biology. Genome Research, 2005. 15(12): p. fungi are large-scale producers of pigments and colorants for the food industry. 1620-1631. Current Opinion in Biotechnology, 2014. 26: p. 56-61. 98. M. Riquelme, J. Aguirre, S. Bartnicki-García, G. H. Braus, M. Feldbrügge, U. Fleig, 87. J. H Clark, F. Ei Deswarte, and T. J Farmer, The integration of green chemistry into W. Hansberg, A. Herrera-Estrella, J. Kämper, and U. Kück, Fungal Morphogenesis, future biorefineries. Biofuels, Bioproducts and Biorefining, 2009. 3(1): p. 72-90. from the Polarized Growth of Hyphae to Complex Reproduction and Infection 88. J. A. Ferreira, A. Mahboubi, P. R. Lennartsson, and M. J. Taherzadeh, Waste Structures. Microbiology and Molecular Biology Reviews, 2018. 82(2): p. e00068-17. biorefineries using filamentous ascomycetes fungi: Present status and future 99. W. Liao, Y. Liu, C. Frear, and S. Chen, A new approach of pellet formation of a prospects. Bioresource technology, 2016. 215: p. 334-345. filamentous fungus – Rhizopus oryzae. Bioresource Technology, 2007. 98(18): p. 89. B. Kendrick, The fifth kingdom. 2017: Hackett Publishing. 3415-3423. 90. S. Wakai, T. Arazoe, C. Ogino, and A. Kondo, Future insights in fungal metabolic 100. J. A. Ferreira, P. R. Lennartsson, L. Edebo, and M. J. Taherzadeh, Zygomycetes-based engineering. Bioresource Technology, 2017. 245: p. 1314-1326. biorefinery: Present status and future prospects. Bioresource technology, 2013. 135: 91. M. R. Goddard and D. Greig, Saccharomyces cerevisiae: a nomadic yeast with no p. 523-532. niche? FEMS Yeast Research, 2015. 15(3): p. fov009-fov009. 101. M. Papagianni, Fungal morphology and metabolite production in submerged mycelial 92. M. J. Boland, A. N. Rae, J. M. Vereijken, M. P. M. Meuwissen, A. R. H. Fischer, M. processes. Biotechnology advances, 2004. 22(3): p. 189-259. A. J. S. van Boekel, S. M. Rutherfurd, H. Gruppen, P. J. Moughan, and W. H.

60 61

102. Y.-Q. Fu, L.-F. Yin, R. Jiang, H.-Y. Zhu, and Q.-C. Ruan, Effects of Calcium on the 113. O. P. Malav, S. Talukder, P. Gokulakrishnan, and S. Chand, Meat Analog: A Review. Morphology of Rhizopus oryzae and L-lactic Acid Production, in Advances in Applied Critical Reviews in Food Science and Nutrition, 2015. 55(9): p. 1241-1245. Biotechnology. 2015, Springer. p. 233-243. 114. A. Denny, B. Aisbitt, and J. Lunn, Mycoprotein and health. Nutrition Bulletin, 2008. 103. Y. Teng, Y. Xu, and D. Wang, Changes in morphology of Rhizopus chinensis in 33(4): p. 298-310. submerged fermentation and their effect on production of mycelium-bound lipase. 115. M. V. Dunlop, S. P. Kilroe, J. L. Bowtell, T. J. A. Finnigan, D. L. Salmon, and B. T. Bioprocess and biosystems engineering, 2009. 32(3): p. 397-405. Wall, Mycoprotein represents a bioavailable and insulinotropic non-animal-derived 104. H. Smith, S. Doyle, and R. Murphy, Filamentous fungi as a source of natural dietary protein source: a dose–response study. British Journal of Nutrition, 2017. antioxidants. Food Chemistry, 2015. 185: p. 389-397. 118(9): p. 673-685. 105. N. Kotwaliwale, P. Bakane, and A. Verma, Changes in textural and optical properties 116. D. A. Williamson, P. J. Geiselman, J. Lovejoy, F. Greenway, J. Volaufova, C. K. of oyster mushroom during hot air drying. Journal of Food Engineering, 2007. 78(4): Martin, C. Arnett, and L. Ortego, Effects of consuming mycoprotein, tofu or chicken p. 1207-1211. upon subsequent eating behaviour, hunger and safety. Appetite, 2006. 46(1): p. 41-48. 106. P. Bobek, L. Ozdín, and L. Kuniak, Effect of oyster mushroom (Pleurotus Ostreatus) 117. S. Matassa, N. Boon, I. Pikaar, and W. Verstraete, Microbial protein: future and its ethanolic extract in diet on absorption and turnover of cholesterol in sustainable food supply route with low environmental footprint. Microbial hypercholesterolemic rat. Food / Nahrung, 1996. 40(4): p. 222-224. Biotechnology, 2016. 9(5): p. 568-575. 107. E. Baysal, H. Peker, M. K. Yalinkiliç, and A. Temiz, Cultivation of oyster mushroom 118. Indexmundi, Fishmeal - Monthly Price - Commodity Prices - Price Charts, Data, and on waste paper with some added supplementary materials. Bioresource Technology, News - IndexMundi. 2018. 2003. 89(1): p. 95-97. 119. V. Prudêncio da Silva, H. M. G. van der Werf, A. Spies, and S. R. Soares, Variability 108. M. J. R. Nout and K. E. Aidoo, Asian Fungal Fermented Food, in Industrial in environmental impacts of Brazilian soybean according to crop production and Applications, H. D. Osiewacz, Editor. 2002, Springer Berlin Heidelberg: Berlin, transport scenarios. Journal of Environmental Management, 2010. 91(9): p. 1831- Heidelberg. p. 23-47. 1839. 109. J. A. Ferreira, P. R. Lennartsson, C. Niklasson, M. Lundin, L. Edebo, and M. J. 120. M. F. Simon and F. L. Garagorry, The expansion of agriculture in the Brazilian Taherzadeh, Spent sulphite liquor for cultivation of an edible Rhizopus sp. Amazon. Environmental Conservation, 2005. 32(3): p. 203-212. BioResources, 2012. 7(1): p. 173-188. 121. J. J. Olli, Å. Krogdahl, T. S. G. A. M. van den Ingh, and L. E. Brattås, Nutritive Value 110. D. Moore and S. W. Chiu, Fungal products as food. Bio-exploitation of filamentous of Four Soybean Products in Diets for Atlantic Salmon (Salmo salar, L.). Acta fungi. Fungal Diversity Press, Hong Kong, 2001: p. 223-251. Agriculturae Scandinavica, Section A — Animal Science, 1994. 44(1): p. 50-60. 111. J. H. Bottin, J. R. Swann, E. Cropp, E. S. Chambers, H. E. Ford, M. A. Ghatei, and G. 122. L. B. Edebo, Zygomycetes for fish feed. 2008, Google Patents. S. Frost, Mycoprotein reduces energy intake and postprandial insulin release without 123. D.-M. Bai, M.-Z. Jia, X.-M. Zhao, R. Ban, F. Shen, X.-G. Li, and S.-M. Xu, L (+)- altering glucagon-like peptide-1 and peptide tyrosine-tyrosine concentrations in lactic acid production by pellet-form Rhizopus oryzae R1021 in a stirred tank healthy overweight and obese adults: a randomised-controlled trial. The British fermentor. Chemical Engineering Science, 2003. 58(3): p. 785-791. Journal of Nutrition, 2016. 116(2): p. 360-374. 124. M. Jin, P. J. Slininger, B. S. Dien, S. Waghmode, B. R. Moser, A. Orjuela, L. da Costa 112. J. A. Ferreira, P. R. Lennartsson, and M. J. Taherzadeh, Production of ethanol and Sousa, and V. Balan, Microbial lipid-based lignocellulosic biorefinery: feasibility and biomass from thin stillage by Neurospora intermedia: a pilot study for process challenges. Trends in biotechnology, 2015. 33(1): p. 43-54. diversification. Engineering in Life Sciences, 2015. 15(8): p. 751-759. 125. K. M. H. Nevalainen, V. S. J. Te'o, and P. L. Bergquist, Heterologous protein expression in filamentous fungi. Trends in Biotechnology, 2005. 23(9): p. 468-474.

62 63

126. M.-H. Chen and M. R. Johns, Effect of pH and nitrogen source on pigment production 139. G. W. White, T. Katona, and J. P. Zodda, The use of high-performance size exclusion by Monascus purpureus. Applied Microbiology and Biotechnology, 1993. 40(1): p. chromatography (HPSEC) as a molecular weight screening technique for 132-138. polygalacturonic acid for use in pharmaceutical applications. Journal of 127. Z. Huang, Y. Xu, L. Li, and Y. Li, Two new Monascus metabolites with strong blue Pharmaceutical and Biomedical Analysis, 1999. 20(6): p. 905-912. fluorescence isolated from red yeast rice. Journal of agricultural and food chemistry, 140. A. M. Pandele, C. Andronescu, E. Vasile, I. C. Radu, P. Stanescu, and H. Iovu, Non- 2007. 56(1): p. 112-118. covalent functionalization of GO for improved mechanical performances of pectin 128. M. Tsukahara, N. Shinzato, Y. Tamaki, T. Namihira, and T. Matsui, Red yeast rice composite films. Composites Part A: Applied Science and Manufacturing, 2017. 103: fermentation by selected Monascus sp. with deep-red color, lovastatin production but p. 188-195. no citrinin, and effect of temperature-shift cultivation on lovastatin production. 141. S. Tripathi, G. K. Mehrotra, and P. K. Dutta, Preparation and physicochemical Applied biochemistry and biotechnology, 2009. 158(2): p. 476-482. evaluation of chitosan/poly(vinyl alcohol)/pectin ternary film for food-packaging 129. M. G. Wiebe, Quorn™ Myco-protein—Overview of a successful fungal product. applications. Carbohydrate Polymers, 2010. 79(3): p. 711-716. Mycologist, 2004. 18(1): p. 17-20. 142. C. Jo, H. Kang, N. Y. Lee, J. H. Kwon, and M. W. Byun, Pectin- and gelatin-based 130. W. C. Li, H. F. Tse, and L. Fok, Plastic waste in the marine environment: A review of film: effect of gamma irradiation on the mechanical properties and biodegradation. sources, occurrence and effects. Science of The Total Environment, 2016. 566-567: p. Radiation Physics and Chemistry, 2005. 72(6): p. 745-750. 333-349. 143. K. Ofori-Kwakye and J. T. Fell, Biphasic drug release: the permeability of films 131. C. B. Crawford and B. Quinn, 3 - Plastic production, waste and legislation, in containing pectin, chitosan and HPMC. International Journal of Pharmaceutics, 2001. Microplastic Pollutants. 2017, Elsevier. p. 39-56. 226(1): p. 139-145. 132. T. Keshavarz and I. Roy, Polyhydroxyalkanoates: bioplastics with a green agenda. 144. P. D. Hoagland and N. Parris, Chitosan/Pectin Laminated Films. Journal of Current Opinion in Microbiology, 2010. 13(3): p. 321-326. Agricultural and Food Chemistry, 1996. 44(7): p. 1915-1919. 133. S. Pilla, Handbook of bioplastics and biocomposites engineering applications. Vol. 145. R. Coffin David, L. Fishman Marshall, and H. Cooke Peter, Mechanical and 81. 2011: John Wiley & Sons. microstructural properties of pectin/starch films. Journal of Applied Polymer Science, 134. A. K. Mohanty, M. Misra, and L. T. Drzal, Sustainable Bio-Composites from 1995. 57(6): p. 663-670. Renewable Resources: Opportunities and Challenges in the Green Materials World. 146. M. L. Fishman, D. R. Coffin, R. P. Konstance, and C. I. Onwulata, Extrusion of Journal of Polymers and the Environment, 2002. 10(1): p. 19-26. pectin/starch blends plasticized with glycerol. Carbohydrate Polymers, 2000. 41(4): p. 135. E. Bioplastics, Bioplastics market data. 2018. 317-325. 136. P. Chomto and J. Nunthanid, Physicochemical and powder characteristics of various 147. M. L. Fishman, D. R. Coffin, C. I. Onwulata, and R. P. Konstance, Extrusion of pectin citrus pectins and their application for oral pharmaceutical tablets. Carbohydrate and glycerol with various combinations of orange albedo and starch. Carbohydrate Polymers, 2017. 174: p. 25-31. Polymers, 2004. 57(4): p. 401-413. 137. M. Ochoa-Villarreal, E. Aispuro-Hernández, I. Vargas-Arispuro, and M. Á. Martínez- 148. M. L. Fishman, D. R. Coffin, C. I. Onwulata, and J. L. Willett, Two stage extrusion of Téllez, Plant cell wall polymers: function, structure and biological activity of their plasticized pectin/poly(vinyl alcohol) blends. Carbohydrate Polymers, 2006. 65(4): p. derivatives, in Polymerization. 2012, InTech. 421-429. 138. X. Ye, Y. Zhan, T. Li, X. Shi, H. Deng, and Y. Du, Pectin based composite 149. A. L. Da Róz, P. Veiga-Santos, A. M. Ferreira, T. C. R. Antunes, F. d. L. Leite, F. M. nanofabrics incorporated with layered silicate and their cytotoxicity. International Yamaji, and A. J. F. d. Carvalho, Water Susceptibility and Mechanical Properties of Journal of Biological Macromolecules, 2016. 93: p. 123-130.

64 65

Thermoplastic Starch Pectin Blends Reactively Extruded with Edible Citric Acid. 160. S. Venkata Mohan, G. N. Nikhil, P. Chiranjeevi, C. Nagendranatha Reddy, M. V. Materials Research, 2016. 19: p. 138-142. Rohit, A. N. Kumar, and O. Sarkar, Waste biorefinery models towards sustainable 150. L. Koch, M. A. Emin, and H. P. Schuchmann, Influence of processing conditions on circular bioeconomy: Critical review and future perspectives. Bioresource the formation of whey protein-citrus pectin conjugates in extrusion. Journal of Food Technology, 2016. 215: p. 2-12. Engineering, 2017. 193: p. 1-9. 161. M. Lieder and A. Rashid, Towards circular economy implementation: a 151. R. Sothornvit, C. W. Olsen, T. H. McHugh, and J. M. Krochta, Tensile properties of comprehensive review in context of manufacturing industry. Journal of Cleaner compression-molded whey protein sheets: Determination of molding condition and Production, 2016. 115: p. 36-51. glycerol-content effects and comparison with solution-cast films. Journal of Food 162. M. Geissdoerfer, P. Savaget, N. M. P. Bocken, and E. J. Hultink, The Circular Engineering, 2007. 78(3): p. 855-860. Economy – A new sustainability paradigm? Journal of Cleaner Production, 2017. 143: 152. K. J. A. Jansens, N. Vo Hong, L. Telen, K. Brijs, B. Lagrain, A. W. Van Vuure, K. p. 757-768. Van Acker, I. Verpoest, P. Van Puyvelde, B. Goderis, M. Smet, and J. A. Delcour, 163. A. Murray, K. Skene, and K. Haynes, The Circular Economy: An Interdisciplinary Effect of molding conditions and moisture content on the mechanical properties of Exploration of the Concept and Application in a Global Context. Journal of Business compression molded glassy, wheat gluten bioplastics. Industrial Crops and Products, Ethics, 2017. 140(3): p. 369-380. 2013. 44: p. 480-487. 164. M. Watson and A. Meah, Food, waste and safety: negotiating conflicting social 153. F. Zubeldía, M. R. Ansorena, and N. E. Marcovich, Wheat gluten films obtained by anxieties into the practices of domestic provisioning. The Sociological Review, 2012. compression molding. Polymer Testing, 2015. 43: p. 68-77. 60(S2): p. 102-120. 154. P. J. P. Espitia, W.-X. Du, R. d. J. Avena-Bustillos, N. d. F. F. Soares, and T. H. 165. E. Papargyropoulou, R. Lozano, J. K. Steinberger, N. Wright, and Z. b. Ujang, The McHugh, Edible films from pectin: Physical-mechanical and antimicrobial properties food waste hierarchy as a framework for the management of food surplus and food - A review. Food Hydrocolloids, 2014. 35: p. 287-296. waste. Journal of Cleaner Production, 2014. 76: p. 106-115. 155. N. Ninan, M. Muthiah, I.-K. Park, A. Elain, S. Thomas, and Y. Grohens, 166. P. Kumar, M. K. Chatli, N. Mehta, P. Singh, O. P. Malav, and A. K. Verma, Meat Pectin/carboxymethyl cellulose/microfibrillated cellulose composite scaffolds for analogues: Health promising sustainable meat substitutes. Critical Reviews in Food tissue engineering. Carbohydrate Polymers, 2013. 98(1): p. 877-885. Science and Nutrition, 2017. 57(5): p. 923-932. 156. L. Liu, L. Liu, C.-K. Liu, M. L. Fishman, and K. B. Hicks, Composite films from 167. F. Alisson-Silva, K. Kawanishi, and A. Varki, Human risk of diseases associated with pectin and fish skin gelatin or soybean flour protein. Journal of agricultural and food red meat intake: Analysis of current theories and proposed role for metabolic chemistry, 2007. 55(6): p. 2349-2355. incorporation of a non-human sialic acid. Molecular Aspects of Medicine, 2016. 51: 157. L. Bonnaillie, H. Zhang, S. Akkurt, K. Yam, and P. Tomasula, Casein Films: The p. 16-30. Effects of Formulation, Environmental Conditions and the Addition of Citric Pectin on

the Structure and Mechanical Properties. Polymers, 2014. 6(7): p. 2018. 158. A. Jurgilevich, T. Birge, J. Kentala-Lehtonen, K. Korhonen-Kurki, J. Pietikäinen, L. Saikku, and H. Schösler, Transition towards Circular Economy in the Food System. Sustainability, 2016. 8(1): p. 69. 159. P. Ghisellini, C. Cialani, and S. Ulgiati, A review on circular economy: the expected transition to a balanced interplay of environmental and economic systems. Journal of Cleaner Production, 2016. 114: p. 11-32.

66 67

Paper 1 fermentation

Article Production of Edible Fungi from Potato Protein Liquor (PPL) in Airlift Bioreactor

Pedro F. Souza Filho *, Akram Zamani and Mohammad J. Taherzadeh Swedish Centre for Resource Recovery, University of Borås, 501 90 Borås, Sweden; [email protected] (A.Z.); [email protected] (M.J.T.) * Correspondence: pedro.ferreira_de_souza_fi[email protected]; Tel.: +46-70-006-6572; Fax: +46-33-435-4003 Academic Editor: Ronnie G. Willaert Received: 13 February 2017; Accepted: 16 March 2017; Published: 21 March 2017

Abstract: Potato protein liquor (PPL), a side stream from the potato starch industry, is normally used as fertilizer. However, with more than 100 g/L of sugars, 20 g/L of Kjeldahl nitrogen and Chemical Oxigen Demand (COD) of 300 g/L, it represents serious environmental challenges. The use of PPL for fungal cultivation is a promising solution to convert this waste into valuable products. In this study, PPL was characterized and used to cultivate edible zygomycete Rhizopus oryzae, which is widely used in Southeast Asian cuisine to prepare e.g., tempeh. Moreover, it can be potentially used as a protein source in animal feed worldwide. Under the best conditions, 65.47 2.91 g of fungal biomass per ± litre of PPL was obtained in airlift bioreactors. The total Kjeldahl nitrogen content of the biomass was above 70 g/kg dry biomass. The best results showed 51% reduction of COD and 98.7% reduction in the total sugar content of PPL.

Keywords: airlift bioreactor; filamentous fungal biomass; fungal pellets; potato protein liquor; Rhizopus oryzae

1. Introduction Potato, a tuber originally from South America, is now one of the most important food crops in the world [1]. It produces more dry matter and protein per hectare than other major cereal crops [2]. More than 50% of the potatoes harvested in North America and Europe are processed to produce, e.g., French fries, chips, and starch [1,3]. Starch is a polysaccharide that is the main energy reserve of higher plants and has applications in food, feed and paper industries [4]. According to the European Starch Industry Association, the starch market in Europe had a turnover of 8.3 billion € in 2014. Potato is the third most used raw material, accounting for 13.3% of starch production [5]. Potato liquor, containing the soluble residues from potato starch production, is treated with steam at 110 ◦C to remove proteins. The remaining liquor is concentrated to potato protein liquor (PPL) and consists of approximately 40% solids [6]. PPL is a severe challenge to wastewater treatment plants because of its high biological oxygen demand [7]. Currently, the side stream is stocked during the year to be used as a fertilizer during planting. The strategy, however, requires large storage space in industrial facilities, contaminates ground water, and annoys adjacent areas with its bad odour. Additionally, PPL can be used as a supplement to cattle feed. However, it is not considered suitable because, despite its high nitrogen content, the quality of the proteins is low since they are oxidized during the steam treatment [6,8]. Alternative treatments have been suggested, attempting to find a destination for PPL, including using it as a substrate in different bioprocesses to produce yeast biomass [9], fungal protein [10], enzymes [7], and other metabolites [11]. However, the cultivation of filamentous fungi in this side stream has never been reported.

Fermentation 2017, 3, 12; doi:10.3390/fermentation3010012 www.mdpi.com/journal/fermentation Fermentation 2017, 3, 12 2 of 12 Fermentation 2017, 3, 12 3 of 12

Filamentous fungi have been extensively used in several biotechnological applications to produce 2.2. Analytical Methods food, feed, and metabolites (e.g., enzymes, organic acids, antibiotics, cholesterol-lowering statins, D-Glucose, fructose, sucrose, D-galactose, raffinose, starch, L-asparagine, L-glutamine, and biofuels). They are commonly cultivated in submerged cultures, in which they can grow in and ammonia were determined in liquid samples using the respective enzymatic assay kits from three morphologies, namely, suspended mycelium, pellet, and clump. Some fungi are also able to Megazyme (Bray, Ireland) following the company’s instructions. Liquid samples of PPL or the grow in a yeast-like form. The morphology is influenced by the strain characteristics as well as the fermentation broth were centrifuged at 3000 g for 10 min prior to analysis. When necessary, medium composition and cultivation conditions [12–15]. Growth in pellet form has been reported as × the samples were sufficiently diluted with distilled water to reach the concentration recommended a favourable alternative for fungal cultivation, because it enables repeated-batch fermentation and in the kit. Kjeldahl nitrogen was determined using the InKjel P digestor and the Behrotest S1 improves the rheology of the broth, resulting in better mass and oxygen transfer in the reactor with distiller (Behr Labor-Technik, Dusseldorf, Germany). Acetic acid, lactic acid, glycerol, and ethanol, lower energy consumption [12–15]. Morphology of the fungi can affect the production of primary and as the main metabolites, were analysed using high-performance liquid chromatography (HPLC) in a secondary metabolites [16]. hydrogen-based ion-exchange column (Aminex HPX-87H, Bio-Rad, Hercules, CA, USA) at 60 ◦C using Zygomycetes are receiving increased attention for both their capacity to produce a large variety 0.6 mL/min of 0.5 mM H2SO4 solution as the eluent. The National Renewable Energy Laboratory of metabolic products and their valuable biomass, which contains several useful components, such as (NREL) analytical procedure was used to determine total carbohydrates [31] based on the hydrolysis proteins, lipids, specific amino acids, chitosan, and chitin, making it eligible to be used for the of the polysaccharides using sulfuric acid and analysing the final solution on HPLC. A lead (II)-based production of animal feed, human food, and chitosan. Economic exploration of zygomycetous column (Aminex HPX-87P, Bio-Rad) operating at 85 ◦C using 0.6 mL/min of ultrapure water as the biomass is believed to be responsible for making the future establishment of a zygomycete-based eluent was used to detect the sugars that were released in the NREL procedure. Components were biorefinery viable [13,17–21]. identified and quantified in the HPLC by a refractive index detector (Waters 2414, Milford, MA, USA). Rhizopus oryzae is one of the most widely used strains of zygomycetous fungi. Among the different The total solids were determined by drying a specified sample volume (10 mL) at 105 ◦Ctoa bio-products produced by R. oryzae, fumaric acid, lactic acid, pectinase, amyloglucosidase, α-amylase, constant weight. The remaining solids were further placed in a muffle furnace (Gallenkamp, London, UK) and ethanol can be named [12,22]. Moreover, it can produce almost pure L(+)-lactic acid without at 550 ◦C to determine the ash content. The amount of insoluble solids was determined by centrifuging any rigorous nutritional requirements [23]. Several investigations were performed to study the effect the PPL at 3000 g for 10 min and drying the obtained solid at the same condition that was previously of culture medium on the morphology of Rhizopus species and the relationship with biomass and × mentioned. An AQUANAL™—professional tube test (1000–15,000 mg/L) (Sigma-Aldrich, St. Louis, metabolite productivity [12,16,24–29]. The use of R. oryzae as fish feed has recently been noted as MO, USA) was used to determine the chemical oxygen demand (COD) in liquid samples. All the a potential replacement for fishmeal in fish feed composition because of the high protein content, waste generated from the experiments was sent to the appropriate treatment, following the Swedish comparable amino acid composition to fishmeal, and the proportion of C18 unsaturated fatty acids. guidelines. Residues of R. oryzae were sent to incineration. In addition, several R. oryzae strains are used in human food and are generally regarded as safe (GRAS) [13]. Stirred-tank bioreactors are not common for the cultivation of filamentous fungi. On the 2.3. Fungal Cultivation in Shake Flasks other hand, airlift bioreactors are considered to be suitable choices for aerobic fungal cultivation, PPL was obtained from Lyckeby Starch AB (Kristianstad, Sweden). All the experiments described due to design simplicity, which favours increasing the scale and decreasing the processing costs and in this paper used the same batch of liquor. R. oryzae was cultivated in 250 mL baffled Erlenmeyer flasks contamination risks [30]. containing 50 mL of culture medium, composed of PPL diluted with tap water. Different PPL–water This study aimed to present a solution for the treatment of PPL and to use R. oryzae to reduce the volume ratios (1:39, 1:19, 1:9, 1:4, 1:1.5) were examined. The medium was sterilized at 121 C for 20 min organic load of potato starch side stream, while producing biomass, a potential product source. Firstly, ◦ before inoculation with 5 mL of the spore suspension at room temperature. The cultivation was carried PPL was characterized in terms of microorganism nutritional requirements such as carbohydrate and out in a water bath shaker at 35 C and 120 rpm for 48 h. The pH of the sample was not adjusted protein content. Secondly, the effect of liquor’s dilution on the morphology and yield of the fungal ◦ during the cultivation. Biomass morphologies and yields were observed in order to choose the best biomass was studied. At last, scaling up of the process to airlift reactors was performed at different conditions for scaling up the cultivation to bench-scale airlift reactors. At the end of the cultivation, aeration rates, aimed at obtaining the best conditions for pelleted fungal growth. the fungal biomass was collected by passing the broth through a kitchen sieve, manually pressing the 2. Materials and Methods collected biomass, and then freeze-drying (0.3 mbar, condenser at 50 ◦C). − As a reference, cultivation of R. oryzae in a synthetic medium containing either sucrose (9.5 g/L) or 2.1. Microorganism asparagine (9.5 g/L) as the carbon source was performed to test the consumption of these two substances. The medium also contained K HPO (3.5 g/L), CaCl 2H O (1.0 g/L), MgSO 7H O (0.75 g/L), Rhizopus oryzae, CCUG 28958, was obtained from the Culture Collection of the University 2 4 2· 2 4· 2 of Gothenburg (Gothenburg, Sweden). The strain was originally isolated from Indonesian food and (NH4)2SO4 (7.5 g/L), and the pH was brought to 5.5 with2MH2SO4. After sterilization production (tempeh), and therefore, it is Generally Recognized As Safe (GRAS). The fungus was grown (121 ◦C, 20 min), the medium was supplemented with a vitamin solution (1 mL/L) and trace metals solution (10 mL/L) [32] which were sterilized by filtration. in 25-mL glass bottles for 5 days at 30 ◦C on a potato dextrose agar (PDA) medium containing (in g/L) potato extract 4; dextrose 20; agar 15. Then, the sporulated slants were stored at 5 C (up to 30 days) ◦ 2.4. Fungal Cultivation on PPL in Airlift Reactor until use. For inoculation, 12 mL of sterile tap water was poured on each slant. The slants were closed and manually agitated to release the spores. The spore suspension was then added to the liquid Fungal cultivation was carried out in a 4-L airlift reactor (Belach Bioteknik AB, Skogås, Sweden) medium to reach a concentration between 2 105 and 5 105 spores/mL in the final medium. filled with 3.5 L of PPL–water solution. Inoculation was made using fungal pellets grown in × × PPL–water solution for 24 h in Erlenmeyer flasks under the same conditions as described in Section 2.3. Cultivation lasted for 54 h with regular sampling in order to analyse the sucrose, glucose, fructose, asparagine, ammonia, ethanol, glycerol, acetic acid, and lactic acid. In order to improve the sucrose Fermentation 2017, 3, 12 4 of 12 Fermentation 2017, 3, 12 5 of 12 consumption, 1 mL of enzyme invertase (from yeast, in 50% glycerol, Megazyme, Ireland) was added Table 1. Potato protein liquor (PPL) characterization. to the reactor. The cultivation was carried out at 35 ◦C and the pH was controlled at 5.5 or below by Composition Average (g/L) addition of 2 M H2SO4. Biomass was harvested at the end of cultivation and freeze-dried. Different aeration rates were examined (0.57, 0.86, and 1.14 vvm) and the consumption of the substrate and the Sugars: biomass yield were measured. Antifoam 204 (Sigma-Aldrich) was used to avoid foaming. The volume Glucose 72.63 6.93 * ± of antifoam needed was determined by aerating the medium before autoclaving at the same rate as Present as monosaccharide 17.49 0.34 ± As component of: Sucrose 24.97 3.20 it would be used in the test. The medium was sterilized inside the reactor, in autoclave, at 121 ◦C ± Raffinose 1.66 0.04 for 20 min. ± Starch † 0.75 0.06 ± 2.5. Statistical Analysis Fructose * Present as monosaccharide 16.17 0.28 ± All the characterization tests and Erlenmeyer cultivations were performed in triplicate. The airlift As component of: Sucrose 24.97 3.20 ± Raffinose 1.66 0.04 cultivations were carried out in duplicate, and the presented results are the mean of the observed ± values and the standard errors of the means. Galactose 10.00 2.43 * ± ® Present as monosaccharide 0.08 0.01 The software MINITAB (version 17.1.0, Minitab Inc., State College, PA, USA) was used to ± As component of: Raffinose 1.66 0.04 statistically analyse the data collected from the experiments. Analyses of variance (ANOVA) were ± Arabinose 1.30 0.82 * performed using general linear models and considering a confidence level of 95%. ± Total 126.71 8.05 ± 3. Results Nitrogen compounds: Total Kjeldahl Nitrogen 20.25 0.86 In this study, potato protein liquor (PPL), a by-product of the potato starch industry, was used for ± Ammonia 1.86 0.12 the cultivation of R. oryzae. PPL consists of a solution with the soluble residues of the potato starch ± Asparagine 20.98 1.26 production after the steam treatment to remove proteins. Its treatment and disposal is a challenge ± Total Solids 438.77 3.98 because of its high organic content. PPL was firstly characterized and then employed for preparation ± Insolubles 9.09 0.41 ± of fungal biomass. The yield and morphology of biomass as well as consumption of PPL ingredients Ashes 114.19 12.82 ± during the course of fermentation were investigated. COD 300 26 ± pH 5.35 3.1. PPL Characterization * Total monosaccharides determined using the NREL methodology. Fructose was not determined by this methodology. † Insoluble in ethanol 95% (v/v). PPL is described in the literature as a source of sugar and nitrogen for microorganisms [6,7,9,10]. The PPL used in this work was characterized by its concentration of sugars, nitrogen compounds, 3.2. Cultivation of Fungi on Different PPL Concentrations total solids, ashes, and COD. The results are presented in Table 1. The pH was determined to be 5.35. The Total Kjeldahl Nitrogen (TKN) was above 20 g/L. As the main amino acid, L-asparagine As a first step, R. oryzae was cultivated using PPL in Erlenmeyer flasks in order to determine the concentration was 20.98 1.26 g/L. The density of the PPL was determined to be 1.15 g/mL. effect of its dilution on fungal growth. PPL was previously reported as containing all the nutrients ± Determination of total carbohydrates in the PPL by the NREL methodology showed a high necessary for microbial growth [6,7,9,10]. Therefore, no nutritional supplementation was used in carbohydrate content in this side stream. The total amount of glucose (including polysaccharides) any cultivation. was determined to be 72.63 6.93 g/L PPL. Some glucose-containing compounds were determined ± 3.2.1. Biomass Yield by enzymatic kits (monomeric glucose, sucrose, raffinose and starch) which represent 62% of the total glucose measured in NREL methodology. Galactose amounted to 10.00 2.43 g/L, and the The yield of R. oryzae biomass was affected by the degree of PPL dilution in the medium with ± concentration of arabinose was 1.30 0.82 g/L. Mannose was not detected. Complete hydrolysis of water (Table 2). Up to dilution 1:4, the increase in the PPL concentration resulted in an increase in the ± sucrose, whose concentration was determined to be 47.43 6.08 g/L, would yield 24.97 3.20 g/L yield of biomass regarding the volume of PPL. A linear (R2 = 0.998) relation of the biomass yield and ± ± of glucose and the same amount of fructose. Total carbohydrates amounted to more than 120 g/L the volume of PPL in the culture was obtained. (considering total glucose, galactose and arabinose, and the determined forms of fructose). Biomass yield = 0.953 C + 0.254 (1) × where C stands for PPL concentration in the mixture (volume of PPL/volume of medium, e.g., for the 1:39 ratio, C = 1/(39 + 1) = 0.025) and the biomass yield is given in grams of dry biomass per gram of carbohydrates. Fermentation 2017, 3, 12 6 of 12 Fermentation 2017, 3, 12 7 of 12

Table 2. Biomass yield and morphology at different dilutions of PPL. Table 3. Concentration of sugars a and amino acid before and after cultivation.

PPL:Water Biomass Yield Biomass Yield Glucose Sucrose Ammonia Asparagine Morphology Final pH PPL–Water Ratio (g/L Undiluted PPL) (g/g Sugar) Ratio 0hb 48 h 0hb 48 h 0hb 48 h 0hb 48 h 1:39 28.11 3.54 0.28 0.04 Filaments 7–8 ± ± 1:39 0.44 0.03 0.01 1.19 1.31 0.07 0.05 0.18 0.01 0.53 0.01 0.02 1:19 29.75 1.95 0.30 0.02 Filaments and Pellets 7–8 ± ± ± ± ± ± 1:19 0.87 0.03 0.02 2.37 2.50 0.07 0.09 0.45 0.02 1.05 0.05 0.12 1:9 34.99 1.06 0.35 0.01 Pellets 7–8 ± ± ± ± ± ± 1:9 1.75 0.09 0.03 4.74 5.20 0.22 0.19 0.81 0.08 2.10 0.13 0.10 1:4 44.51 0.09 0.45 0.00 Clumps 5–6 ± ± ± ± ± ± 1:4 3.50 0.07 0.02 9.49 9.95 0.10 0.37 0.92 0.34 4.20 2.32 1.29 1:1.5 - - No growth 4–5 ± ± ± ± 1:1.5 6.99 6.14 0.25 18.97 17.20 0.36 0.75 0.92 0.14 8.39 8.23 0.65 ± ± ± ± a Fructose was not found after autoclaving. b Initial concentrations were estimated from characterization with the 3.2.2. Biomass Morphology same standard deviation. Increasing the PPL concentration in the mixture changed the R. oryzae morphology from filaments 3.3. Sucrose and Asparagine Consumption to pellets and later to clumps. The different morphologies are shown in Figure 1. When PPL was diluted by 1:39 ratio, R. oryzae grew sparsely as filaments. At 1:19 dilution, more biomass was produced, The results presented in Table 3 indicate some challenges in the consumption of sucrose by the resulting in entangled hyphae with a few small pellets. At 1:9 dilution, only pellets were obtained; strain and the use of an amino acid as a carbon source. Asparagine, as well as glutamine, are the and a dilution of 1:4 resulted in a large and single clump of the fungi. For each PPL–water solution, main free amino acids present in potato [33]. Thus, an experiment was designed to investigate the the fungus presented the same morphology since the beginning of its growth. role of these two compounds in R. oryzae growth. Approximately 32% of the sucrose was hydrolysed to glucose and fructose during autoclavation as a result of the low pH of the medium (5.5). Table 4 presents the concentration of the substances as well as the biomass collected and pH after 48 h of the cultivation. Less than 5% of the non-hydrolysed sucrose and 11.8% of the asparagine were consumed by the fungus after 48 h. The consumption of sugars decreased the pH to 2.8; while the asparagine consumption increased the pH to 7.2. The biomass grew as pellets in both cases. The biomass yield was 0.50 0.07 g dry biomass/g of sugars consumed and 1.09 0.04 g dry biomass/g of asparagine. ± ± Table 4. Cultivation of R. oryzae in sucrose and asparagine media for 48 h. Consumption of substrates and effect on pH.

Carbon Source Sucrose Asparagine Time (h) 0 48 0 48 pH 5.50 2.81 5.50 7.23 Glucose (g/L) 1.48 0.15 0.01 0.00 - - ± ± Fructose (g/L) 1.51 0.08 0.01 0.00 - - ± ± Sucrose (g/L) 6.45 0.20 6.17 0.19 - - ± ± Asparagine (g/L) - - 9.77 1.15 8.61 0.73 Figure 1. Biomass morphologies obtained when growing R. oryzae in Erlenmeyer flasks using different ± ± Ammonia (g/L) - - 2.08 0.82 2.73 0.53 ± ± PPL–water ratios: (a) 1:39; (b) 1:19; (c) 1:9 and; (d) 1:4. Biomass (g/L) - 1.65 0.23 - 1.26 0.04 ± ± 3.2.3. Metabolites Formation 3.4. Fungal Cultivation in Airlift Bioreactors Analysis of the metabolites in the medium after 48 h of cultivation of the fungus (Table 3) showed As cultivation of the fungus at a PPL:water ratio of 1:9 resulted in fungal growth with pellet an increase in the ammonia concentration. Glucose and asparagine concentrations decreased. At the morphology as well as a high yield, this ratio was chosen for further investigation of the cultivations end of the cultivation, raffinose, a trisaccharide composed of a glucose linked to a galactose and a to airlift bioreactors. The aeration rate was chosen to vary from 0.57 to 1.14 vvm (volume of gas per fructose, disappeared and sucrose concentration increased. The pH went up to the alkaline region culture medium per minute), i.e., 1 to 3 NL/min. Pellets were only obtained at the highest aeration (between 7 and 8) in the diluted 1:39, 1:19, and 1:9 PPL:water media (Table 2). Lactic acid, ethanol and rate examined, while at other aeration rates the fungus grew as clumps. The biomass yield increased glycerol were only formed at PPL–water ratio 1:1.5, reaching average concentrations of 2.21 0.01, ± from 41.63 5.43 to 65.47 2.91 g/L undiluted PPL when aeration increased from 0.57 to 1.14 vvm. 0.31 0.15, and 0.85 0.06 g/L, respectively. ± ± ± ± Although high aeration was required and the medium content of the protein was high, no problems with foam were observed during the cultivation, provided that the antifoam had been added at the beginning of cultivation. The maximum antifoam volume that was necessary was 0.002 mL antifoam/mL medium. The Kjeldahl nitrogen content of the biomass was also analysed, and the results are presented in Table 5. The protein content can be estimated by multiplying the nitrogen content by 6.25 [30]. Protein content is an important factor when fungal biomass is intended to be destined for feed applications. The maximum TKN was obtained at 0.86 vvm, representing 50% protein content (w/w). Fermentation 2017, 3, 12 8 of 12 Fermentation 2017, 3, 12 9 of 12

Table 5. Biomass yield, morphology, and nitrogen content of the fungal biomass cultivated on 1:9 An increased aeration rate, as expected, reduced the ethanol production (Figure 2D–F). In all of the water–diluted PPL in airlift bioreactors at different aeration rates. cases, ethanol had a maximum concentration after 24 h. Lactic acid was also simultaneously produced. After the sugar was depleted, both ethanol and lactic acid were assimilated by the fungus. Glycerol was Biomass Yield Total Kjeldahl Nitrogen Aeration (vvm) Morphology formed together with ethanol, reaching the highest concentration (1.13 0.01 g/L) at 24 h in the lower (g/L Undiluted PPL) (TKN) (g/kg Dry Biomass) ± aeration case and decreasing over a longer time. The value of the concentration of some components 0.57 41.63 5.43 74.98 5.08 Clumps ± ± varied for each cultivation at time zero, e.g., lactic acid and glycerol. The difference may have been 0.86 50.07 0.58 79.78 4.63 Clumps ± ± 1.14 65.47 2.91 70.97 13.58 Pellets caused by e.g., the variable concentration of these components in the used inoculum. Different observed ± ± Erlenmeyer 34.99 1.06 61.96 5.74 Pellets sucrose concentrations at time zero are the result of its hydrolysis during autoclavation as well as the ± ± action of the invertase enzyme in the period between the sampling and the sugar analysis. The acetic acid concentration, however, decreased from approximately 0.15 to 0 g/L during the first day (data not During the first 24 h, the sugars were fully consumed. The sucrose was partially hydrolysed shown). This behaviour was expected because acetate is used in the process of aerobic respiration during the autoclavation of the medium. The remainder of this disaccharide was fully hydrolysed (e.g., for the formation of acetyl-CoA enzyme). during the first 12 h of cultivation by the enzyme invertase added at the beginning of the cultivation. In all of the airlift bioreactor experiments in this work, the final COD of the medium was Asparagine was entirely consumed after 48 h, but the ammonium concentration, different from what found to be 14.8 0.1 g/L, against an initial value of 30 g COD/L (300 g/L of PPL 10-fold diluted). was observed in shake flasks, did not increase (Figure 2A–C). The pH levels were monitored and the ± This represents more than a 50% reduction in COD, leading to a lower load for the wastewater plant profile is presented for the lower aeration case (Figure 2D). The pH decreased from 5.4 in the beginning and resulting in cost reduction in addition to biomass production, which can be used for animal feed. to 5.2 during the first day, together with the sugar consumption. When the sugar was finished, the pH started to increase and, after 46 h, it reached 5.5 and sulfuric acid started being added to the medium. 4. Discussion The PPL characterization results presented in Table 1 are comparable with the literature; Schürgerl and Rosen reported a total sugar content between 4.0% and 14% (w/w) and a raw protein content between 16% and 24% (w/w)[10]. The total sugar concentration of 8.7% (w/w) was in agreement with cited data, and Kjeldahl protein content (multiplying TKN by 6.25) of 11% (w/w) was found to be below that reported in literature. Davies showed asparagine and glutamine as the main free amino acids in potatoes [33], while Heisler identified both amino acids in the composition of potato starch factory soluble waste [34]. However, in the PPL sample used in this study, L-glutamine was not found. Using 1:4 water–diluted PPL for fungal cultivation resulted in 0.45 g of biomass/g of sugar consumed. Ferreira et al. obtained a maximum fungal biomass yield of 0.34 g/g of sugar using spent sulphite liquor [30]. This might suggest either a lower inhibition in PPL, which results in lower consumption of the substrate for maintenance and can be further submitted to investigation, or the presence of other substances in the medium that could be used by the fungus as a substrate such as asparagine. Asparagine has long been considered to be a good source of carbon and nitrogen for microorganisms [35]. Two enzymes were found to be involved in asparagine uptake by microorganisms: asparaginase, which catalyses the hydrolysis of asparagine to aspartate and ammonium, and aspartase ammonia-lyase, which catalyses the deamination of aspartate to fumarate and ammonium [36]. The results of the experiments using asparagine as a substrate confirm the role of the asparagine as a carbon source and its contribution to a higher biomass yield. The final pH in the alkaline region, together with the increased concentration of ammonia, contributed to stop the growth of R. oryzae. The pH, however, did not increase in the cultivation with 1:4 PPL:water ratio due to the buffering capacity of the liquor. The raffinose can be hydrolysed to its monosaccharides by the enzymes α-galactosidase (acting in the glucose–galactose linkage) and β-fructosidase (cleaving the glucose–fructose bond) [37]. The disappearance of raffinose with an increased sucrose concentration suggests the fungal ability to break the linkage between galactose and glucose but not the glucose–fructose linkage. The result is in agreement with Bulut et al., who have also described the poor metabolization of sucrose by R. oryzae strain [38]. In the cultivation using airlift bioreactors, the aeration rate showed an influence on fungal morphology. The fungal morphology during the microbial growth altered from clumps, when the Figure 2. Cultivation of R. oryzae in 1:9 water–diluted PPL in airlift bioreactors: concentration profiles of glucose ( ); sucrose ( ); fructose ( ); ammonia ( ); asparagine ( ); ethanol (-); lactic acid (ɳ); tested aeration rates were 0.57 and 0.86 vvm, to pellets, when the aeration rate was 1.14 vvm.  ×  and glycerol (+) at different aeration rates: (A,D) 0.57 vvm; (B,E) 0.86 vvm; (C,F) 1.14 vvm. pH profile Increasing the aeration rate in the airlift bioreactors enlarges the superficial gas velocities, improving shown for the first case (D) in the dashed line. the volumetric mass transfer coefficient [30], the gas holdup, and the liquid circulation velocity (shear force) [39]. The cultivation showed a significantly better performance than that obtained in Fermentation 2017, 3, 12 10 of 12 Fermentation 2017, 3, 12 11 of 12

Erlenmeyer flasks (Table 5) comparing the obtained biomass yields. For 0.86 and 1.14 vvm of aeration, 9. Lotz, M.; Fröhlich, R.; Matthes, R.; Schügerl, K.; Seekamp, M. Bakers’ yeast cultivation on by-products and the biomass yield in relation to the PPL used was significantly (p < 0.025) higher than that obtained in wastes of potato and wheat starch production on a laboratory and pilot-plant scale. Process Biochem. 1991, 26, shake flasks. The best condition tested was cultivation in a 1:9 diluted PPL medium and 1.14 vvm at 301–311. [CrossRef] 35 ◦C, given that these conditions resulted in pellet growth of the fungus. Sucrose was hydrolysed by 10. Schügerl, K.; Rosen, W. Investigation of the use of agricultural byproducts for fungal protein production. the invertase enzyme added at the beginning of the cultivation. The glucose and fructose left in the Process Biochem. 1997, 32, 705–714. [CrossRef] medium after the cultivation represent 1.3% of the same sugars present at the beginning. 11. Kumar, K.R.; Singh, A.; Schügerl, K. Fed-batch culture for the direct conversion of cellulosic substrates to An interesting result from the airlift cultivation was that the ammonia concentration (in the form acetic acid/ethanol by Fusarium oxysporum. Process Biochem. 1991, 26, 209–216. [CrossRef] of ammonium salts) did not increase as it did in the Erlenmeyer flasks. One of the reasons for this 12. Liao, W.; Liu, Y.; Frear, C.; Chen, S. A new approach of pellet formation of a filamentous different behaviour may be that, compared to the Erlenmeyer cultivation, more nitrogen was necessary fungus—Rhizopus oryzae. Bioresour. Technol. 2007, 98, 3415–3423. [CrossRef][PubMed] 13. Ferreira, J.A.; Lennartsson, P.R.; Edebo, L.; Taherzadeh, M.J. Zygomycetes-based biorefinery: Present status to form more biomass with higher TKN concentration in the airlift reactor. Therefore, more ammonia and future prospects. Bioresour. Technol. 2013, 135, 523–532. [CrossRef][PubMed] was consumed. In Section 3.3, it was determined that asparagine consumption results in a higher 14. Papagianni, M. Fungal morphology and metabolite production in submerged mycelial processes. biomass yield than the one from sugar consumption. Thus, it may be an interesting strategy to wait for Bioresour. Technol. 2004, 22, 189–259. [CrossRef] asparagine consumption in order to obtain more biomass. 15. Fu, Y.-Q.; Yin, L.-F.; Jiang, R.; Zhu, H.-Y.; Ruan, Q.-C. Effects of Calcium on the Morphology of Rhizopus oryzae Using S. cerevisiae, Lotz et al. reduced the COD of a PPL solution by 44% (1:9 PPL–water ratio), and L-lactic Acid Production. In Advances in Applied Biotechnology; Zhang, T.-C., Nakajima, M., Eds.; Springer: from 32.2 to 18 g/L, using a continuously stirred-tank reactor [9]. R. oryzae was able to reduce it to less Berlin/Heidelberg, Germany, 2015; pp. 233–243. than half of the initial values in the present study, proving its efficiency to reduce the organic load of 16. Teng, Y.; Xu, Y.; Wang, D. Changes in morphology of Rhizopus chinensis in submerged fermentation and the side stream while producing valuable biomass. their effect on production of mycelium-bound lipase. Bioprocess Biosyst. Eng. 2009, 32, 397–405. [CrossRef] [PubMed] 5. Conclusions 17. Zamani, A.; Taherzadeh, M.J. Effects of partial dehydration and freezing temperature on the morphology Potato protein liquor was proven to be a suitable medium for the cultivation of R. oryzae in an and water binding capacity of carboxymethyl chitosan-based superabsorbents. Ind. Eng. Chem. Res. 2010, 49, airlift bioreactor with pellet morphology producing high protein content (almost 50%) fungal biomass. 8094–8099. [CrossRef] The fungus could not consume the PPL directly, and the dilution with water was necessary. In addition, 18. Bankefors, J.; Kaszowska, M.; Schelechtriem, C.; Pickova, J.; Brännäs, E.; Edebo, L.; Kiessling, A.; the aeration rate was found to play an important role in the fungal morphology. The best condition Sandström, C. A comparison of the metabolic profile on intact tissue and extracts of muscle and liver of juvenile Atlantic salmon (Salmo salar L.)–Application to a short feeding study. Food Chem. 2011, 129, tested was cultivation in 1:9 dilution of PPL and 1.14 vvm at 35 ◦C. Sugars were completely assimilated within the first 24 h of growth, and as the cultivation continued, other substances were consumed. 1397–1405. [CrossRef] COD reduction of the liquor proved the fungus efficiency in reducing the organic load. 19. Lennartsson, P. Zygomycetes and Cellulose Residuals: Hydrolysis, Cultivation and Applications; School of Engineering, University of Borås: Borås, Sweden; Department of Chemical and Biological Engineering, Acknowledgments: This work was financed by Coordination for the Improvement of Higher Education Personnel Chalmers University of Technology: Göteborg, Sweden, 2012. (CAPES-Brazil) and Vinnova (Sweden) in collaboration with Cewatech AB, and Lyckeby Starch AB. 20. Karimi, K.; Zamani, A. Mucor indicus: Biology and industrial application perspectives: A review. Author Contributions: Pedro F. Souza Filho performed the experiments; Pedro F. Souza Filho, Akram Zamani, Biotechnol. Adv. 2013, 31, 466–481. [CrossRef][PubMed] and Mohammad J. Taherzadeh were involved in the conception and design of the experiments. All the authors 21. Zamani, A. Superabsorbent Polymers from the Cell Wall of Zygomycetes Fungi; Chalmers University of Technology: analysed the data and contributed to writing the paper. Göteborg, Sweden, 2010. Conflicts of Interest: The authors declare no conflicts of interest. 22. Edebo, L. Zygomycetes for Fish Feed. Patent WO2008/002231, 3 January 2008. 23. Bai, D.-M.; Jia, M.-Z.; Zhao, X.-M.; Ban, R.; Shen, F.; Li, X.-G.; Xu, S.-M. L(+)-lactic acid production by References pellet-form Rhizopus oryzae R1021 in a stirred tank fermentor. Chem. Eng. Sci. 2003, 58, 785–791. [CrossRef] 1. Bradshaw, J.E.; Ramsay, G. Potato origin and production. In Advances in Potato Chemistry and Technology, 24. Liao, W.; Liu, Y.; Chen, S. Studying pellet formation of a filamentous fungus Rhizopus oryzae to enhance 1st ed.; Singh, J., Kaur, L, Eds.; Elsevier: Amsterdam, The Netherlands, 2009; pp. 1–26. organic acid production. In Applied Biochemistry and Biotecnology; Springer: Berlin/Heidelberg, Germany, 2. Storey, R.M.J. The harvested crop. In Potato Biology and Biotechnology Advanced Perspectives, 1st ed.; 2007; pp. 689–701. Vreugdenhil, D., Ed.; Elsevier: Amsterdam, The Netherlands, 2007; pp. 441–470. 25. Byrne, G.S.; Ward, O.P. Effect of nutrition on pellet formation by Rhizopus arrhizus. Biotechnol. Bioeng. 1989, 3. Van Soest, J.J.G.; Tournois, H.; de Wit, D.; Vliegenthart, J.F.G. Short-range structure in (partially) crystalline 33, 912–914. [CrossRef][PubMed] potato starch determined with attenuated total reflectance Fourier-transform IR spectroscopy. Carbohydr. Res. 26. Byrne, G.S.; Ward, O.P. Growth of Rhizopus arrhizus in fermentation media. J. Ind. Microbiol. 1989, 4, 1995, 279, 201–214. [CrossRef] 155–161. [CrossRef] 4. Jobling, S. Improving starch for food and industrial applications. Curr. Opin. Plant Biol. 2004, 7, 210–218. 27. Liu, Y.; Liao, W.; Chen, S. Study of pellet formation of filamentous fungi Rhizopus oryzae using a multiple [CrossRef][PubMed] logistic regression model. Biotechnol. Bioeng. 2008, 99, 117–128. [CrossRef][PubMed] 5. Starch.eu. Available online: http://www.starch.eu/ (accessed on 20 December 2015). 28. Žnidaršiˇc, P.; Komel, R.; Pavko, A. Influence of some environmental factors on Rhizopus nigricans submerged 6. Schügerl, K. Agriculture wastes. A source of bulk products? Chem. Eng. Technol. 1994, 17, 291–300. [CrossRef] growth in the form of pellets. World J. Microbiol. Biotechnol. 2000, 16, 589–593. [CrossRef] 7. Klingspohn, U.; Papsupuleti, P.V.; Schügerl, K. Production of enzymes from potato pulp using batch 29. Zhou, Y.; Du, J.; Tsao, G.T. Mycelial pellet formation by Rhizopus oryzae ATCC 20344. operation of a bioreactor. J. Chem. Technol. Biotechnol. 1993, 58, 19–25. [CrossRef] Appl. Biochem. Biotechnol. 2000, 84, 779–789. [CrossRef] 8. Klingspohn, U.; Bader, J.; Kruse, B.; Kishore, P.V.; Schügerl, K. Utilization of potato pulp from potato starch 30. Ferreira, J.A.; Lennartsson, P.R.; Niklasson, C.; Lundin, M.; Edebo, L.; Taherzadeh, M.J. Spent sulphite liquor processing. Process Biochem. 1993, 28, 91–98. [CrossRef] for cultivation of an edible Rhizopus sp. BioResources 2012, 7, 173–188. Fermentation 2017, 3, 12 12 of 12

31. Sluiter, A.; Hames, B.; Ruiz, R.; Scarlata, C.; Sluiter, J.; Templeton, D. Determination of Sugars, Byproducts, and Degradation Products in Liquid Fraction Process Samples; National Renewable Energy Laboratory: Golden, CO, USA, 2006. 32. Sues, A.; Millati, R.; Edebo, L.; Taherzadeh, M.J. Ethanol production from hexoses, pentoses, and dilute-acid hydrolyzate by Mucor indicus. FEMS Yeast Res. 2005, 5, 669–676. [CrossRef][PubMed] 33. Davies, A.M.C. The free amino acids of tubers of potato varieties grown in England and Ireland. Potato Res. 1977, 20, 9–21. [CrossRef] 34. Heisler, E.G.; Siciliano, J.; Treadway, R.H.; Woodward, C.F. Recovery of free amino compounds from potato starch processing water by use of ion exchange. Am. J. Potato Res. 1959, 36, 1–11. [CrossRef] Paper 2 35. Arima, K.; Sakamoto, T.; Araki, C.; Tamura, G. Production of extracellular L-asparaginases by microorganisms. Agric. Biol. Chem. 1972, 36, 356–361. [CrossRef] 36. Huerta-Zepeda, A.; Durán, S.; Du Pont, G.; Calderón, J. Asparagine degradation in Rhizobium etli. Microbiology 1996, 142, 1071–1076. [CrossRef] 37. Rehms, H.; Barz, W. Degradation of stachyose, raffinose, melibiose and sucrose by different tempe-producing Rhizopus fungi. Appl. Microbiol. Biotechnol. 1995, 44, 47–52. [CrossRef][PubMed] 38. Bulut, S.; Elibol, M.; Ozer, D. Effect of different carbon sources on L(+)-lactic acid production by Rhizopus oryzae. Biochem. Eng. J. 2004, 21, 33–37. [CrossRef] 39. Nitayavardhana, S.; Issarapayup, K.; Pavasant, P.; Khanal, S.K. Production of protein-rich fungal biomass in an airlift bioreactor using vinasse as substrate. Bioresour. Technol. 2013, 133, 301–306. [CrossRef][PubMed]

© 2017 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/). fermentation

Article Techno-Economic and Life Cycle Assessment of Wastewater Management from Potato Starch Production: Present Status and Alternative Biotreatments

Pedro F. Souza Filho * ID , Pedro Brancoli ID , Kim Bolton, Akram Zamani and Mohammad J. Taherzadeh ID Swedish Centre for Resource Recovery, University of Borås, 501 90 Borås, Sweden; [email protected] (P.B.); [email protected] (K.B.); [email protected] (A.Z.); [email protected] (M.J.T.) * Correspondence: pedro.ferreira_de_souza_fi[email protected]; Tel.: +46-70-006-6572; Fax: +46-33-435-4003 Received: 13 September 2017; Accepted: 16 October 2017; Published: 23 October 2017

Abstract: Potato liquor, a byproduct of potato starch production, is steam-treated to produce protein isolate. The heat treated potato liquor (HTPL), containing significant amounts of organic compounds, still needs to be further treated before it is discarded. Presently, the most common strategy for HTPL management is concentrating it via evaporation before using it as a fertilizer. In this study, this scenario was compared with two biotreatments: (1) fermentation using filamentous fungus R. oryzae to produce a protein-rich biomass, and (2) anaerobic digestion of the HTPL to produce biogas. Technical, economic and environmental analyses were performed via computational simulation to determine potential benefits of the proposed scenarios to a plant discarding 19.64 ton/h of HTPL. Fungal cultivation was found to be the preferred scenario with respect to the economic aspects. This scenario needed only 46% of the investment needed for the evaporation scenario. In terms of the environmental impacts, fungal cultivation yielded the lowest impacts in the acidification, terrestrial eutrophication, freshwater eutrophication, marine eutrophication and freshwater ecotoxicity impact categories. The lowest impact in the climate change category was obtained when using the HTPL for anaerobic digestion.

Keywords: potato liquor; techno-economic analysis; life cycle assessment; filamentous fungus; anaerobic digestion

1. Introduction Potato is one of the most important food crops in the world, and it accounts for 13.3% of the starch produced in the European Union (EU). The processing of potato to produce starch results in two major byproducts: potato pulp (PP) and potato liquor (PL). PP contains the insoluble polysaccharides cellulose, hemicellulose, pectin and residual starch. Proteins, minerals and trace elements in high concentrations, are the major ingredients of PL [1]. Each metric ton of processed potato yields approximately 200 kg of starch and generates ca 700 kg PL [2] containing 30–41% protein per total solid (TS) [3]. The proteins present in the PL are of good quality, similar to those of whole eggs [3]. Different methods to recover these proteins have been reported. They include thermal coagulation, acid precipitation, salting out, isoelectric precipitation, complexing with carboxymethylcellulose or bentonite, ultrafiltration, expanded bed adsorption and dry separation [3–6]. However, the only method that is presently being used for industrial recovery of protein from PL is heat coagulation [3–7]. In this method, steam is injected into the liquor at a pH of 5.5 to increase the temperature to 99 ◦C. This coagulates the proteins, which are then precipitated and collected [8]. However, the proteins

Fermentation 2017, 3, 56; doi:10.3390/fermentation3040056 www.mdpi.com/journal/fermentation Fermentation 2017, 3, 56 2 of 15 Fermentation 2017, 3, 56 3 of 15

obtained by this method lose their functional properties and are mostly used as an additive to cattle feed. the dissolved proteins are coagulated by injecting steam with a temperature of 140 ◦C. The HTPL (i.e., Moreover, they have a salty, bitter taste, which prevents them from being used as food additives [3,7,9]. the remaining liquid after protein removal) was characterized by Fang et al. (2011) [2]. After the proteins have been removed from PL, the residual liquid—called HTPL—is further evaporated 2.1.1. Concentration of HTPL by Evaporation (Evaporation Scenario) at 140 ◦C to produce potato protein liquor (PPL), containing 40% (w/w)TS[1]. The most common method to manage PPL is to use it as fertilizer. However, most potato HTPL is evaporated to produce steam that is used at the same facility. After protein coagulation, processing occurs during the winter. The reduced biological activity in the soil during this period approximately 84,848 metric tons of HTPL (per six month operational period) containing 3.3% (w/w) prevents the uptake of the liquor, which forces the potato starch facilities to store large volumes TS is sent to a boiler to be concentrated to PPL containing 40% (w/w) TS. The PPL was characterized by of PPL for several months [9]. Additionally, the use of PPL as fertilizer causes contamination of Souza Filho et al. (2017) [1]. Evaporation occurs at 122 ◦C and 2 bar and at a flow rate of 19.64 ton/h of groundwater and emits a bad odor that can disturb citizens living in neighboring areas [10]. However, HTPL, producing 18.1 ton/h of steam and 1.5 ton/h of PPL. Part of the steam (15%) is used to preheat as mentioned above, despite the high content of residual proteins in the PPL, they are of poor quality, the HTPL and the remaining part is used for other operations at the same facility. The process flow hence preventing them from being used as a supplement for cattle feed [1]. diagram (PFD) of this scenario is presented in Figure 1. A few alternative management strategies for PPL have been presented in the literature.

The biodegradable nature of the byproduct has stimulated its use in bioprocesses to produce EVAPORATION SCENARIO Steam yeast biomass [11], acetate and ethanol [12], enzymes [10] and fungal protein [13]. Additionally, the filamentous zygomycete fungus Rhizopus oryzae has been used to reduce the chemical oxygen Heat Treated Potato Liquor Pump Heat Potato Protein demand (COD) of the PPL and produce a protein-rich biomass with a potential application as fish 19.64 ton/h exchanger Liquor Evaporator To farmers feed [1]. Storage tank One of the reasons for the growing interest in producing supplements for fish feed is that aquaculture has been growing at about 8% per year since the late 1970s, which is higher than the FUNGUS SCENARIO Air + CO2 rate of human population growth. It has been suggested that the reason for this large growth is Wet air HTPL 19.64 ton/h the widespread knowledge of the importance of fish for a healthy lifestyle, mainly because fish Filter Enzyme contains ω-3 polyunsaturated fatty acids. Fish feed is a major cost in intensive fish farming [14], and Biomass zygomycete fungi can be used as a substitute for fish feed [15]. These are filamentous fungi that Defoamer Dryer Airlift bioractor contain large amounts of polyunsaturated fatty acids and protein, resulting in an increased content of Pumps Air

Air these components in the fish feed [16]. Ferreira et al. (2016) [17] reported that ascomycete biomass Air blower Air sterilizer Wastewater (which has protein and fatty acid compositions comparable to zygomycete) is a good substitute for Compressor soybean-based feeds in the diet of animals like poultry, cattle, chicken and fish. BIOGAS SCENARIO Among the zygomycete strains that have been investigated for fish feed, R. oryzae is a promising Biogas microorganism. R. oryzae has been used for many centuries in Asian cuisine to prepare fermented food, Condenser Compressor such as tempeh. Therefore, it is Generally Regarded As Safe (GRAS), which is a very favorable property HTPL when investigating its potential use as animal feed [18]. Due to these arguments, the possibility of 19.64 ton/h Pump using HTPL to cultivate R. oryzae is studied in this work. Alternatively, HTPL can be used in anaerobic Anaerobic Storage Wastewater digestor tank digestion (AD) to produce biogas. AD is considered to be a sustainable form of treating industrial waste while simultaneously producing energy in the form of biogas [2]. Production of biogas from waste can have several benefits, including reduction in the costs of waste treatment, contribution to global Figure 1. Process flow diagram of the three scenarios studied in this work. energy needs using relatively cheap feedstock, and lower environmental impact than conventional types of energy [19]. 2.1.2. Cultivation of Filamentous Fungus in HTPL (Fungus Scenario) In the present study, techno-economic and life cycle assessments of the treatment and use of Experimental studies by Souza Filho et al. (2017) [1] indicate that R. oryzae grow best in the PPL HTPL are performed for three scenarios. The current strategy of PPL production is compared with two waste stream when it is diluted back to 1:9 (i.e., 1 volume of PPL to 9 volumes of water). Therefore, alternative scenarios where the HTPL is used for (i) the cultivation of filamentous fungus R. oryzae the stream before concentration (i.e., the HTPL stream) is used in this modelling to cultivate R. oryzae biomass or (ii) the production of biogas. The data used in the analyses are obtained from experiment, under aerobic conditions using an airlift bioreactor. This type of bioreactor uses air, which is sparged the literature and industrial potato starch production plants. in the medium, as the sole source of agitation. The aeration rate used in the reactor was calculated to keep the same gas holdup used by Souza Filho et al. (2017) [1]. The terminal velocity of a spherical 2. Materials and Methods bubble (vt) in a bubble column is given by the Stokes’ law: 2.1. Process Description g d2 ρ ρ · · l − g This study is based on a typical plant producing potato starch that operates for a period of vt = , (1) 18 μl  six months per year. It processes 300,000 metric tons of potato, produces 62,000 metric tons of starch · and discards 210,000 m3 of PL per six month period. The pH of the liquor is adjusted to 5.3 before where g is the gravitational acceleration, d is the diameter of the bubble, ρl is the density of the medium, ρg is the density of the bubble and μl is the dynamic viscosity of the medium. The time it takes for a bubble to leave the liquid (Δt) is: Fermentation 2017, 3, 56 4 of 15 Fermentation 2017, 3, 56 5 of 15

Table 2. Technical values used for simulation of the fungus scenario 1. H Δt = , (2) vt Type Assumption where H is the height of the liquid column. During this time the volume of gas injected into the reactor Reactor type EGSB (Vair) is: Hydraulic retention time (HRT) 8 days Organic loading rate (OLR) 3.2 g COD/Lreactor.day Vair = Qair Δt, (3) · Temperature 37 ◦C where Qair is the air flow rate. The gas holdup (ε), defined as the volume of air divided by the volume Methane production rate 1420 mL CH4/Lreactor.day of liquid present in the reactor, is: Methane concentration 58% (v/v) VFA content in the bioreactor 1 mM Biogas pressure in the distribution pipeline 5 bar Vair Qair Δt Qair H ε = = · = · , (4) 1 V V v V Data based on [2]. liquid liquid t· liquid Assuming that the properties of the liquid and gas phases are the same in the experimental work 2.2. Energy, Equipment, and Economic Analyses and in the simulation (i.e., vt is constant), and that the gas holdup is the same in experiment and in The energy, equipment and economic aspects were studied using Aspen Plus® V9 (Aspentech, the simulation, then the air flow rate in the simulated reactor can be calculated from the experimental Burlington, MA, USA) integrated with Aspen Energy Analyzer. The simulated data was exported to the data using: Aspen Process Economic Analyzer software, where economic assumptions were entered. All economic Qair1 H1 Vliquid2 assumptions used in this study are listed in Table 3. A modified version of the activity coefficient Qair2 = · , (5) Vliquid1 · H2 model NRTL (i.e., ELECNRTL) was used in all of the scenarios to include the effect of the electrolytes The subscript 2 represents the properties in the simulated reactor and the subscript 1 the properties present in the HTPL. All of the equipment was made of stainless steel or carbon steel. The simulations in the reactor used in the previous work. The height:diameter ratio of the simulated reactor was kept included the purchase of one back-up pump identical to the original pump for all pumps. No further the same as the one used in the experimental work by Souza Filho et al. (2017) [1]. sterilization of the HTPL was considered in the fungus scenario, since the HTPL passes a heat treatment The proteins present in the HTPL induce the formation of foam [7]. Therefore, 0.2% (v/v of process in the starch plant. Contamination risks were not considered during the economic evaluation HTPL) defoamer is used in the reactor. Moreover, invertase is added at a proportion of 32.6 U per g of the Fungus Scenario. of HTPL to assist the hydrolysis of the sucrose in the medium. After cultivation, the broth is sent to Table 3. Economic evaluation inputs and operational cost. filters to separate the fermented broth from the fungal biomass. The broth containing low COD is sent to a wastewater treatment plant. The collected biomass is dried using hot air and used as fish feed. Type Assumption The operational conditions used in this simulation are presented in Table 1. Annual operating time 4368 h (26 weeks) Depreciation method Straight line Table 1. Technical values used for the Fungus Scenario 1. Working capital 1 15%/period Tax rate 1 33%/period Type Assumption Interest rate 1 6%/period Reactor type Airlift Lifetime of the plant 1 20 years 1 1 Dilution rate 0.1 h− Salvage value 20% of initial capital cost Temperature 35 ◦C Operator labor 20 €/h Biomass yield 4.6 g/L HTPL Supervisor labor 35 €/h Nitrogen content in biomass 7.456% (w/w) Electricity 1 0.0775 €/kW h 1 · 1 Data based on [1]. Steam 0.01 €/kg Wastewater treatment 1 0.001 €/m3 2 2.1.3. Anaerobic Digestion of HTPL for Biogas Production (Biogas Scenario) Fish meal 0.929 €/kg Digestible crude protein content in fish meal 3 65.6% (DM) Fang et al. (2011) [2] have investigated the production of biogas from HTPL using an expanded Digestible crude protein content in fungal biomass 4 44.1% (DM) granular sludge bed (EGSB) reactor. They found that the bioreactor can be operated continuously at a Price conversion rate fungal biomass/fish meal 0.672 Invertase 5 2.25 10 5 €/U hydraulic retention time of 8 days, removing 87% of the COD in the form of biogas. The specifications × − Defoamer 6 2.3 €/L that were used to simulate the biogas digester are presented in Table 2. The digestate resulting from Biogas 7 33 €/MW h · the biogas production was kept in a storage tank for 20 h to remove the residual methane dissolved Low heat value Biogas (58% CH ) 5.47 kW h/Nm3 4 · before being sent to a wastewater treatment plant. It was assumed that the biogas that is produced is 1 Data based on [20]; 2 [21]; 3 [22]; 4 [16]; 5 [23]; 6 [24]; 7 [25]. DM: dry matter. used directly, without upgrading, in a combined heat and power (CHP) plant in the vicinity of the facility that produces the potato starch. The biogas is compressed to 5 bar before being sent to the The price of fungal biomass was estimated using the price of fish meal adjusted by a factor based gas grid. on the digestible protein content of both materials (see Table 3). The biogas price was calculated according to the market price of biogas in Sweden and the low heat value of the produced biogas. The revenue from the steam produced in the evaporation scenario is calculated from the regular steam price (see Table 3), even though it is used in the same plant. Economic calculations using the Aspen Fermentation 2017, 3, 56 6 of 15 Fermentation 2017, 3, 56 7 of 15

Process Economic Analyzer (Aspentech, Burlington, MA, USA) were performed based on the prices from the first quarter of 2015. Capital costs, operating costs, product sales and net present value (NPV) Energy were calculated considering a lifetime of the plant of 20 years.

1 ton HTPL PPL Storage and land Evaporator 2.3. Life Cycle Assessment (LCA) application This study uses a Consequential Life Cycle Assessment (CLCA) approach. CLCA assesses the environmental impact of products and yields information regarding the consequences as a result of Avoided mineral fertilizer production 1.5 kg N marginal changes [26]. Therefore, it includes activities that are directly or indirectly affected by a (Calcium ammonium nitrate and Potassium chloride) 5.3 kg K marginal change in the level of output of a product [27]. Within the CLCA approach, system expansion is used to handle coproducts. In this method, the boundaries of the system are expanded to include the environmental impacts of alternative processes that produce the same products or functions as the studied coproducts [28]. The main product in this study is the supply of a treatment service, i.e., Energy, chemicals, heat the service of treating the HTPL for further use. The coproducts are considered to substitute products 1 ton HTPL Effluent Fermentation that are already available on the market [29]. The products that are substituted are shown in Table 4. treatment

Table 4. Coproducts obtained from the three scenarios studied here and the alternative products that are replaced by the coproducts. Avoided fish meal production 6.0 kg fungal biomass

Scenario Coproduct Replaced Product for Coproduct Evaporation PPL Fertilizers 1 2 Fungus Fungal biomass Fishmeal 1 ton HTPL Biogas firing in CHP Anaerobic digestion Plant Electricity Marginal market for electricity in Sweden 3 Biogas Heat Biomass in CHP plant 3 1 aaEfluentaa Effluent Marginal fertilizers: Calcium ammonium nitrate and potassium chloride [30]. Inventory data for fertilizer treatment production retrieved from consequential life cycle assessment (CLCA) EcoInvent database [31]; 2 Data from CLCA EcoInvent database [31]; 3 Inventory data for fishmeal production retrieved from Fréon et al. [32].

Avoided marginal energy production 123 MJ electricity 2.3.1. System Boundaries, Functional Unit, and Environmental Impact Categories 170 MJ heat

The system boundaries for the three scenarios are show in Figure 2. It is assumed that the geographical boundaries for the systems are within Sweden. The functional unit is the treatment of Figure 2. System boundaries and reference flows for the evaporation, fungus and biogas scenarios. one ton of HTPL residue. It is assumed that the waste material enters the system burden free, i.e., The dotted lines show the avoided products in the system expansion. without any environmental impact associated with it. The selected environmental impact categories were global warming potential, acidification, fresh water ecotoxicity, as well as terrestrial, marine and Table 5. Biochemical composition of the fungal biomass and the substituted products. freshwater eutrophication. The life cycle impact assessment (LCIA) methodologies recommended by Biochemical Parameter Unit Fish Meal 1 Fungal Biomass 2 the International Reference Life Cycle Data System (ILCD) were used [33]. The selection of the impact categories allows comparison of the systems for different environmental burdens and geographical Gross energy MJ/kg DM 20.4 20.2 Digestible energy MJ/kg DM 16.7 3 16.34 scales (global and regional), in order to be able to identify and avoid solutions that could decrease local Crude protein % DM 70.6 47.5 impacts but increase global burdens or vice versa [34]. The calculations were done using SimaPro v.8.3 1 [22]; 2 [16]; 3 Salmonid digestible energy. (PRé Sustainability: Amersfoort, The Netherlands). Treatment of Wastewater 2.3.2. Basic Assumptions and Data Sources The effluent after the fungi cultivation (fungus scenario) or the AD (biogas scenario) requires Fungal Biomass further treatment. The wastewater treatment was adapted from the process “Wastewater from It was assumed that the fungal biomass is used to replace conventional fish meal in the potato starch production” from EcoInvent Consequential database [31] based on the wastewater market [35]. Table 5 shows the values for energy and protein content of the biomass and the fish meal. composition from Souza Filho et al. (2017) [1] and Fang et al. (2011) [2] for the fungus and biogas The substitution rate of the fish meal by the fungal biomass used the same factor previously mentioned scenarios, respectively. for the economic analysis and that is shown in Table 3. Nutrient Recovery The PPL in the evaporation scenario is used as organic fertilizer for nitrogen and potassium. It was assumed that this organic fertilizer substitutes—and hence avoids the production of—the Fermentation 2017, 3, 56 8 of 15 Fermentation 2017, 3, 56 9 of 15 mineral fertilizers calcium ammonium nitrate and potassium chloride [30]. The emissions to air and HTPL. The equipment, size and construction material used in the simulation, as well as the individual water when using nitrogen for fertilizer on land used data from Tonini, Hamelin [30], which are prices, are presented in Table 6. average values in the literature regarding emissions to air and water from organic residues used as fertilizer [36–38]. Table 6. Equipment costs for the different scenarios.

Transportation of Coproducts Scenario Equipment Capacity/Size 1 Material Cost (thousand €) The PPL produced in the evaporation scenario is collected by farmers and transported an average HTPL pump 6.2 L/s SS 8.2 Heat exchanger 114 m2 SS 61.5 of 100 km. The fungal biomass produced in the fungus scenario was considered to be transported Evaporation Evaporator 650 m2 CS 5450.5 300 km to be used as feed in aquaculture production in western Sweden, where 10% of the national Storage tank 7000 m3 CS 107.8 fish production occurs [39]. HTPL pump 6.2 L/s SS 8.2 Air compressor for fermenter 467 m3/h CS 153.0 3. Results and Discussion Sterile air filter 467 m3/h 2 2.0 Production of potato starch generates a protein-rich side stream which is exploited by the industry Pump for defoamer 4.4 mL/s CS 4.1 Fungus Pump for enzyme 98 mL/s CS 4.1 to produce protein isolate. The residual wastewater is given (without charge) to the farmers to be used Airlift fermenter 200 m3 SS 1478.5 as fertilizer. This scenario was compared to other scenarios in which the wastewater is used to produce Biomass filter 9.3 m2 CS 109.0 fungal biomass for use as fish feed or to produce biogas. Biomass dryer 9.3 m2 CS 47.5 Air blower for dryer 3095 m3/h CS 10.9 3.1. Technical Analysis HTPL pump 6.2 L/s SS 8.2 3 In the evaporation scenario, 19,641 kg/h of HTPL are pumped to a boiler operating at 2 bar to EGSB digester 4800 m CS 3952.2 Biogas Storage tank 480 m3 CS 29.7 concentrate the HTPL to PPL, which contains approximately 40% (w/w) solids. The boiler produces Water condenser 1.6 m2 SS 10.3 18,126 kg/h of steam which is used to preheat the HTPL before it enters the boiler. Approximately Biogas compressor 260 m3/h SS 879.9 267 MW h/day of energy are consumed in this process. · 1 Heat exchangers and filters defined by the surface area. Pumps and compressors defined by the flow rate. 2 Filter The fungus scenario, involving the production of R. oryzae biomass to be used as fish feed, was material cannot be adjusted in Aspen Process Economic Analyzer. evaluated for the same flow rate used for the evaporation scenario (19,641 kg/h of HTPL during six months a year). The cultivation of R. oryzae yielded a biomass production of 2475 kg/day Production of fungal biomass (fungus scenario) demands much less energy. Only 9.2% of the (445 metric tons for an operational period of six months) containing 46.6% crude protein. The energy energy presently used in the evaporation scenario would be required to produce the fungal biomass consumption in this process is 24.5 MW h/day, which is primarily due to the aeration of the bioreactor from HTPL. Cultivating the fungus on the potato starch wastewater has a capital cost of about · 3 (95% of the energy consumed in the scenario is for aeration). A daily volume of 473.8 m of wastewater €7.5 million. The cost of the airlift bioreactor was considered to be the same as the cost of a jacketed is discarded by the plant. An airlift bioreactor was chosen because of the improved agitation achieved vertical tank. A rotary drum filter was used to collect the biomass after fermentation. Drying the in this design without the use of internal parts (e.g., impellers and baffles) in which the fungus can biomass was achieved using a direct contact rotary dryer. The compressor used to provide air to the grow around interfering in the mass transfer [40]. bioreactor was designed to provide an aeration of 0.04 vvm, and was calculated using Equation (4). 3 In the biogas scenario, 279.1 Nm /h of biogas containing 58.8% (v/v) of methane is produced. The costs of the equipment are presented in Table 6. The operational cost of the plant was estimated to At this concentration, the biogas contains a low heat value of 5.47 kW h/Nm3. The biogas production · be €1.84 million per operating period, and the fungal biomass obtained during this period was sold as is equivalent to 36.6 MW h/day, while the energy consumption is 11.9 MW h/day, i.e., less than half · · fish feed supplement for €282,150. of the energy needed for the fungus scenario. This is because the anaerobic digester uses mechanical The biogas scenario has the lowest energy consumption (11.9 MW h/day). The capital investment · agitation to create homogeneous conditions inside the reactor, as opposed to the airlift bioreactor for this scenario is about €14.2 million. Operational costs are €1.4 million/period (including sending used in the fungus scenario, which uses aeration as the source of agitation. This decreases the energy wastewater to a municipal treatment plant), and €216,785/period would be obtained from selling the 3 demand in the biogas scenario. 474.0 m of wastewater is generated each day. Compared to the biogas. Compared to the fungus scenario, the biogas scenario demands 89% more capital investment evaporation scenario, which is presently the most common alternative in potato starch plants, both and the operational cost is 24% lower. The capital cost, operating cost and product sales for the bioprocess scenarios reduce the energy consumption. proposed scenarios are presented in Figure 3. The digestate from the AD still contains nutrients which can be recovered in the form of fertilizer. However, the low concentration of such components in the 3.2. Economic Analysis digestate would require processes that have high energy demands, e.g., evaporation or centrifugation, Treatment of HTPL to PPL in the evaporation scenario requires an operating cost of €1.7 million or the transportation of large volumes of liquid. This would increase the costs associated with biogas per operating period (6 months). No income when using the PPL as fertilizer was accounted for in scenario. Therefore, the wastewater produced in fungus and biogas scenarios is sent to the municipal this scenario, since PPL is given without charge to the farmers for use as fertilizer. The excess steam wastewater treatment plant. produced and not used to preheat the HTPL represents an income of €671,414. The capital cost for this The NPV diagram after 10, 15 and 20 years is presented in Figure 4. No scenario returns the scenario is approximately €16.5 million. The evaporation of HTPL, containing as little as 3.3% (w/w) investment made. Fungal cultivation (fungus scenario) results in a NPV that is less negative than of TS, which is required to obtain the highly-concentrated PPL, demands the highest amount of heat AD (biogas scenario). After 15 years, the NPV of the biogas scenario becomes similar to the NPV when compared to the other scenarios. A standard vertical vessel was used to estimate the cost of the of the evaporation scenario and, at the end of the lifetime of the plant, evaporation and fungus boiler for direct steam injection. Also, a shell and tube heat exchanger was designed to preheat the scenarios have comparable NPV. This is caused by the large capital cost and low operational cost of Fermentation 2017, 3, 56 10 of 15 Fermentation 2017, 3, 56 11 of 15

the evaporation scenario, opposed to the low capital cost and large operational cost of the fungus 60 0.40 scenario. The contrasting characteristics of the scenarios lead to a shift in the NPV during the last 40 0.30 five years of the plant’s lifetime. Rajendran et al. [20] estimated that the capital cost for a municipal 0.20 solid waste (MSW) AD plant is about 40 million USD with operating costs of about 3 million USD 20 per year. The plant was designed to treat 55,000 m3 of MSW and to produce compressed biogas (CBG) 0.10 for the transport sector. The MSW, which has a high TS content, yields 64 Nm3 of raw biogas per m3 0 0.00 of MSW versus 14.2 Nm3 of raw biogas per m3 of HTPL. This creates a situation where a plant can -0.10 -20 make a profit from waste treatment. In the case of the HTPL, the dilute nature of the waste stream Evaporation Fungus Sc. Biogas Sc. Evaporation Fungus Sc. Biogas Sc. Sc. requires larger equipment and higher energy consumption, and returns lower quantities of biogas, Sc. Acidification (molc H+ eq) hence making it difficult to operate the treatment processes with a positive economic balance. Climate change (kg CO2 eq) 0.30 2.5 0.25 2.0 15 0.20 4 1.5 0.15 1.0 0.10 0.5 0.05 10 0.0 0.00 -0.5 -0.05 2 Evaporation Fungus Sc. Biogas Sc. Evaporation Fungus Sc. Biogas Sc. 5 Sc. Sc.

(MillionEuro/period) Terrestrial eutrophication (molc N eq) Marine eutrophication (kg N eq)

0.016 300 Capital cost (Million Euro) (Million cost Capital Operating cost and product sales 0 0 0.012 200 Evaporation Fungus Biogas Scenarios 0.008 100 Capital cost Operating cost Product sales 0.004 0 Figure 3. Results from the economic evaluation for the different scenarios considered in this study. 0.000 -100 The period is one year with six operational months. -0.004 -200 Evaporation Fungus Sc. Biogas Sc. Evaporation Fungus Sc. Biogas Sc. Sc. Sc. -35 Freshwater eutrophication (kg P eq) Freshwater ecotoxicity (CTUe) -30

-25

-20

-15

-10 Figure 5. Environmental impacts of the three scenarios studied in this work. -5

0 The fungus scenario has lower impacts than the evaporation scenario in all of the impact categories. Net Present Value Euro) NetPresent (Million 10 15 20 It also has a lower impact than the biogas scenario in all of the impact categories except climate change. Years The lower impact of the biogas scenario on climate change is mainly due to the avoided marginal Evaporation Scenario Fungus Scenario Biogas Scenario energy production, which is a result from the biogas firing in a CHP plant. It can also be noted Figure 4. Net present value for the different scenarios after 10, 15 and 20 years. that biogas scenario has lower impacts than the evaporation scenario in all impact categories except freshwater ecotoxicity. 3.3. Life Cycle Assessment The trends seen in the acidification, terrestrial eutrophication, marine eutrophication and freshwater eutrophication impact categories are the same, with the fungus scenario having the lowest Figure 5 shows the environmental impacts of the three scenarios for HTPL treatment and use. impact and the evaporation scenario the highest. The impact of fungus scenario on freshwater The results show that the evaporation scenario has the largest impact in all of the impact categories eutrophication is 77% and 55% lower compared to the evaporation and biogas scenarios, respectively except freshwater ecotoxicity. This is primarily due to the impacts related to the large amount of heat (Figure 5). The results show that for the freshwater ecotoxicity category, the fungus scenario has the required for the evaporation of HTPL to PPL, which is part of the process emissions seen in the figure. best performance with impacts that are 48% and 51% lower compared to the evaporation and biogas Due to this heat requirement, the evaporation scenario has a large environmental impact despite the scenarios, respectively (Figure 5). abatement from the avoided production of mineral fertilizer (seen as nutrients recovery in Figure 5). Fermentation 2017, 3, 56 12 of 15 Fermentation 2017, 3, 56 13 of 15

3.4. Comparison of the Different Scenarios EU European Union GRAS Generally regarded as safe The fungus scenario has the lowest impact in five out of the six environmental impact categories HRT Hydraulic retention time that were analyzed. This may indicate that it is the preferred option. However, it must be emphasized HTPL Heat treated potato liquor that this scenario has a larger impact than the biogas scenario in the climate change category. Since this LCA Life cycle assessment impact category is considered by the United Nations “the single biggest threat to development” [41], LCIA Life cycle impact assessment this result may have a central role when selecting the preferred scenario. MSW Municipal solid waste The fungus scenario is also the preferred scenario according to the economic analysis, since it has NPV Net present value the lowest capital cost and the best NPV during the first fifteen years of operation. In contrast, at the OLR Organic loading rate end of the plant’s lifetime, the evaporation scenario becomes economically more viable. However, PFD Process flow diagram the difference between the two scenarios is only 1% of the evaporation scenario’s NPV. PL Potato liquor The evaporation scenario has the largest impact in five out of the six environmental impact PP Potato pulp categories, in addition to having the largest capital investment of all scenarios. PPL Potato protein liquor Since none of the scenarios was best in all of the analyzed parameters, it is not possible to draw a SS Stainless steel TS Total solids simple conclusion regarding the preferred scenario. A decision would ultimately be made according U Unity of enzyme activity to the political, environmental or economic agenda of the decision-makers and identifying the crucial USD United States dollar factors can be difficult [42]. The most important contribution of this study is to highlight the trade-offs VFA Volatile fatty acid inherently involved in the decision process. References 4. Conclusions 1. Souza Filho, P.F.; Zamani, A.; Taherzadeh, M.J. Production of edible fungi from potato protein liquor (PPL) Technical, economic and environmental analyses were performed to determine potential benefits in airlift bioreactor. Fermentation 2017, 3, 12. [CrossRef] of two proposed scenarios to a plant discarding 19.64 ton/h of HTPL. The two proposed scenarios are 2. Fang, C.; Boe, K.; Angelidaki, I. Biogas production from potato-juice, a by-product from potato-starch to use the HTPL (i) to cultivate filamentous fungus R. oryzae to produce a protein-rich biomass (fungus processing, in upflow anaerobic sludge blanket (UASB) and expanded granular sludge bed (EGSB) reactors. scenario) and (ii) to produce biogas via AD (biogas scenario). These two scenarios are compared to the Bioresour. Technol. 2011, 102, 5734–5741. [CrossRef][PubMed] most commonly used treatment method, which is concentrating the HTPL before using it as a fertilizer. 3. Ralet, M.-C.; Guéguen, J. Fractionation of potato proteins: Solubility, thermal coagulation and emulsifying Both proposed scenarios reduce the capital cost and the energy consumption of the wastewater properties. LWT-Food Sci. Technol. 2000, 33, 380–387. [CrossRef] treatment. Moreover, the current study highlights the environmental benefits of cultivating fungi in 4. Zhang, D.-Q.; Mu, T.-H.; Sun, H.-N.; Chen, J.-W.; Zhang, M. Comparative study of potato protein concentrates the HTPL (fungus scenario), since it has the lowest impact in acidification, freshwater ecotoxicity as extracted using ammonium sulfate and isoelectric precipitation. Int. J. Food Prop. 2017, 20, 2113–2127. well as the terrestrial, freshwater, and marine eutrophication categories. In contrast, the greenhouse [CrossRef] gas emissions were higher from fungus scenario compared to biogas scenario, where the residue was 5. Waglay, A.; Karboune, S.; Alli, I. Potato protein isolates: Recovery and characterization of their properties. anaerobically digested. The results show that the substituted products in the system expansion, such as Food Chem. 2014, 142, 373–382. [CrossRef][PubMed] mineral fertilizers, electricity and heat, substantially reduce the environmental footprints of fungus 6. Strætkvern, K.O.; Schwarz, J.G. Recovery of native potato protein comparing expanded bed adsorption and and biogas Scenarios. This study presents the techno-economic and environmental trade-offs that are ultrafiltration. Food Bioprocess Technol. 2012, 5, 1939–1949. [CrossRef] necessary to take into account when selecting one of the scenarios in preference to the others. 7. Bárta, J.; Heˇrmanová, V.; Diviš, J. Effect of low-molecular additives on precipitation of potato fruit juice proteins under different temperature regimes. J. Food Process Eng. 2008, 31, 533–547. [CrossRef] Acknowledgments: The authors would like to acknowledge the Coordination for the Improvement of Higher 8. Klingspohn, U.; Bader, J.; Kruse, B.; Kishore, P.V.; Schuegerl, K.; Kracke-Helm, H.A.; Likidis, Z. Utilization of Education Personel (CAPES-Brazil), and the Gunnar Ivarsson foundation for financing this work. potato pulp from potato starch processing. Process Biochem. 1993, 28, 91–98. [CrossRef] Author Contributions: Pedro F Souza Filho, Akram Zamani, and Mohammad J. Taherzadeh conceived 9. Zwijnenberg, H.J.; Kemperman, A.J.B.; Boerrigter, M.E.; Lotz, M.; Dijksterhuis, J.F.; Poulsen, P.E.; Koops, G.-H. and designed the simulation scenarios. Pedro Brancoli and Kim Bolton were responsible for the life cycle Native protein recovery from potato fruit juice by ultrafiltration. 2002, , 331–334. [CrossRef] analysis. Pedro F Souza Filho and Pedro Brancoli wrote the paper. Kim Bolton, Akram Zamani, and Desalination 144 Mohammad J. Taherzadeh reviewed and edited the manuscript. 10. Klingspohn, U.; Vijai Papsupuleti, P.; Schügerl, K. Production of enzymes from potato pulp using batch operation of a bioreactor. J. Chem. Technol. Biotechnol. 1993, 58, 19–25. [CrossRef] Conflicts of Interest: The authors declare no conflict of interest. 11. Lotz, M.; Fröhlich, R.; Matthes, R.; Schügerl, K.; Seekamp, M. Bakers’ yeast cultivation on by-products and Abbreviations wastes of potato and wheat starch production on a laboratory and pilot-plant scale. Process Biochem. 1991, 26, 301–311. [CrossRef] AD Anaerobic digestion 12. Kumar, P.K.R.; Singh, A.; Schügerl, K. Fed-batch culture for the direct conversion of cellulosic substrates to CBG Compressed biogas acetic acid/ethanol by Fusarium oxysporum. Process Biochem. 1991, 26, 209–216. [CrossRef] CHP Combined heat and power 13. Schügerl, K.; Rosen, W. Investigation of the use of agricultural byproducts for fungal protein production. CLCA Consequential life cycle assessment Process Biochem. 1997, 32, 705–714. [CrossRef] COD Chemical oxygen demand 14. Olli, J.J.; Krogdahl, Å.; van den Ingh, T.S.; Brattås, L.E. Nutritive value of four soybean products in diets for CS Carbon steel atlantic salmon (Salmo salar, L.). Acta Agric. Scand. Sect. A 1994, 44, 50–60. [CrossRef] DM Dry matter EGSB Expanded granular sludge bed Fermentation 2017, 3, 56 14 of 15 Fermentation 2017, 3, 56 15 of 15

15. Ferreira, J.A.; Lennartsson, P.R.; Niklasson, C.; Lundin, M.; Edebo, L.; Taherzadeh, M.J. Spent sulphite liquor 37. Hamelin, L.; Jørgensen, U.; Petersen, B.M.; Olesen, J.E.; Wenzel, H. Modelling environmental consequences for cultivation of an edible Rhizopus sp. BioResources 2012, 7, 173–188. of direct land use changes from energy crops in a self-sustained and fully renewable energy system in 16. Ferreira, J.A.; Lennartsson, P.R.; Taherzadeh, M.J. Production of ethanol and biomass from thin stillage by Denmark: Effect of crop types, soil, climate, residues management, initial carbon level and turnover Neurospora intermedia: A pilot study for process diversification. Eng. Life Sci. 2015, 15, 751–759. [CrossRef] time. In Proceedings of the Quantifying and Managing Land Use Effects of Bioenergy, Campinas, Brazil, 17. Ferreira, J.A.; Mahboubi, A.; Lennartsson, P.R.; Taherzadeh, M.J. Waste biorefineries using filamentous 19–21 September 2011. ascomycetes fungi: Present status and future prospects. Bioresour. Technol. 2016, 215, 334–345. [CrossRef] 38. Galloway, J.N.; Dentener, F.J.; Capone, D.G.; Boyer, E.W.; Howarth, R.W.; Seitzinger, S.P.; Asner, G.P.; [PubMed] Cleveland, C.C.; Green, P.A.; Holland, E.A.; et al. Nitrogen cycles: Past, present, and future. Biogeochemistry 18. Ferreira, J.A.; Lennartsson, P.R.; Edebo, L.; Taherzadeh, M.J. Zygomycetes-based biorefinery: Present status 2004, 70, 153–226. [CrossRef] and future prospects. Bioresour. Technol. 2013, 135, 523–532. [CrossRef][PubMed] 39. Statistiska Centralbyrån. Aquaculture in Sweden in 2014; Statistiska Centralbyrån: Stockholm, Sweden, 2014. 19. Mao, C.; Feng, Y.; Wang, X.; Ren, G. Review on research achievements of biogas from anaerobic digestion. 40. Ferreira, J.A.; Lennartsson, P.R.; Taherzadeh, M.J. Airlift bioreactors for fish feed fungal biomass production Renew. Sustain. Energy Rev. 2015, 45, 540–555. [CrossRef] using edible filamentous fungi. In Proceedings of the FFBiotech Symposium, Villeneuve d’Ascq, France, 20. Rajendran, K.; Kankanala, H.R.; Martinsson, R.; Taherzadeh, M.J. Uncertainty over techno-economic 15–16 May 2017. potentials of biogas from municipal solid waste (MSW): A case study on an industrial process. Appl. Energy 41. United Nations. Goal 13: Sustainable Development Knowledge Platform. Available online: https:// 2014, 125, 84–92. [CrossRef] sustainabledevelopment.un.org/sdg13 (accessed on 13 October 2017). 21. Indexmundi. Available online: http://www.indexmundi.com/commodities/?commodity=fish-meal 42. Eriksson, O.; Bisaillon, M.; Haraldsson, M.; Sundberg, J. Enhancement of biogas production from food waste (accessed on 12 September 2017). and sewage sludge—Environmental and economic life cycle performance. J. Environ. Manag. 2016, 175, 22. Feedipedia. Available online: https://www.feedipedia.org/node/208 (accessed on 12 September 2017). 33–39. [CrossRef][PubMed] 23. Invertase on alibaba.com. Available online: https://www.alibaba.com/product-detail/INVERTASE_ © 2017 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access 60590660279.html (accessed on 12 September 2017). article distributed under the terms and conditions of the Creative Commons Attribution 24. Defoamer for Fermentation on alibaba.com. Available online: https://www.alibaba.com/product-detail/ (CC BY) license (http://creativecommons.org/licenses/by/4.0/). Defoamer-for-Fermentation_1473403819.html (accessed on 12 September 2017). 25. Joelsson, E.; Dienes, D.; Kovacs, K.; Galbe, M.; Wallberg, O. Combined production of biogas and ethanol at high solids loading from wheat straw impregnated with acetic acid: Experimental study and techno-economic evaluation. Sustain. Chem. Processes 2016, 4, 14. [CrossRef] 26. Ekvall, T.; Weidema, B.P. System boundaries and input data in consequential life cycle inventory analysis. Int. J. Life Cycle Assess. 2004, 9, 161–171. [CrossRef] 27. Brander, M.; Tipper, R.; Hutchison, C.; Davis, G. Technical Paper: Consequential and Attributional Approaches to LCA: A Guide to Policy Makers with Specific Reference to Greenhouse Gas LCA of Biofuels; Ecometrica Press: London, UK, 2009. 28. Ekvall, T.; Finnveden, G. Allocation in ISO 14041—A critical review. J. Clean. Prod. 2001, 9, 197–208. [CrossRef] 29. By-products, Recycling and Waste. Available online: https://consequential-lca.org/clca/by-products- recycling-and-waste/ (accessed on 12 September 2017). 30. Tonini, D.; Hamelin, L.; Wenzel, H.; Astrup, T. Bioenergy production from perennial energy crops: A consequential LCA of 12 bioenergy scenarios including land use changes. Environ. Sci. Technol. 2012, 46, 13521–13530. [CrossRef][PubMed] 31. Wernet, G.; Bauer, C.; Steubing, B.; Reinhard, J.; Moreno-Ruiz, E.; Weidema, B. The Ecoinvent database version 3 (part I): Overview and methodology. Int. J. Life Cycle Assess. 2016, 21, 1218–1230. [CrossRef] 32. Fréon, P.; Durand, H.; Avadí, A.; Huaranca, S.; Moreyra, R.O. Life cycle assessment of three peruvian fishmeal plants: Toward a cleaner production. J. Clean. Prod. 2017, 145, 50–63. [CrossRef] 33. European Commission. International Reference Life Cycle Data System (ILCD) Handbook—General Guide for Life Cycle Assessment—Provisions and Action Steps, 1st ed.; Publications Office of the European Union: Ispra, Italy, 2011; ISBN 978-92-79-17451-3. 34. Van der Werf, H.M.G.; Petit, J. Evaluation of the environmental impact of agriculture at the farm level: A comparison and analysis of 12 indicator-based methods. Agric. Ecosyst. Environ. 2002, 93, 131–145. [CrossRef] 35. Edebo, L.B. Zygomycetes for Fish Feed. Patent WO2008/002231, 3 January 2008. 36. De Klein, C.; Novoa, R.S.A.; Ogle, S.; Smith, K.A.; Rochette, P.; Wirth, T.C.; McConkey, B.G.; Mosier, A.;

Rypdal, K.; Walsh, M.; et al. N2O emissions from managed soils, and CO2 emissions from lime and urea application. In 2006 IPCC Guidelines for National Greenhouse Gas Inventories; Eggleston, H.S., Buendia, L., Miwa, K., Ngara, T., Tanabe, K., Eds.; IPCC National Greenhouse Gas Inventories Programme: Kanagawa, Japan, 2006; Volume 4, pp. 1–54, ISBN 4-88788-032-4. Paper 3 Waste and Biomass Valorization https://doi.org/10.1007/s12649-018-0265-2

ORIGINAL PAPER

Edible Protein Production by Filamentous Fungi using Starch Plant Wastewater

Pedro F. Souza Filho1 · Akram Zamani1 · Mohammad J. Taherzadeh1

Received: 18 January 2018 / Accepted: 5 March 2018 © The Author(s) 2018. This article is an open access publication

Abstract The process to obtain starch from wheat requires high amounts of water, consequently generating large amounts of wastewater with very high environmental loading. This wastewater is traditionally sent to treatment facilities. This paper introduces an alternative method, where the wastewater of a wheat-starch plant is treated by edible filamentous fungi (Aspergillus oryzae and Rhizopus oryzae) to obtain a protein-rich biomass to be used as e.g. animal feed. The wastewater was taken from the clarified liquid of the first and second decanter (ED1 and ED2, respectively) and from the solid-rich stream (SS), whose carbohydrate and nitrogen concentrations ranged between 15 and 90 and 1.25–1.40 g/L, respectively. A. oryzae showed better performance than R. oryzae, removing more than 80% of COD after 3 days for ED1 and ED2 streams. Additionally, 12 g/L of dry biomass with protein content close to 35% (w/w) was collected, demonstrating the potential of filamentous fungi to be used in wastewater valorization. High content of fermentable solids in the SS sample led to high production of ethanol (10.91 g/L), which can be recovered and contribute to the economics of the process.

Keywords Filamentous fungi · Wastewater treatment · Bioethanol · Fungal biomass

Introduction demand (COD) of the wastewater, generally ranging between 6 and 10 g/L [7–10]. Wheat is one of the most important crops in human diet. In the EU, legislation governing emissions from indus- Globally, wheat supplies more protein than poultry, pig, and trial installations follows the “polluter pays” principle. These bovine meat together [1]. The European Union (EU) is the norms use market-based instruments so the industry takes most important producer of wheat in the world, accounting responsibility on the external damage caused by industrial for over 20% of the global production [2]. According to the activities. Such instruments include taxes and emission trad- European starch manufacture, wheat is the most processed ing schemes. The environmental policy requires industrial crop, accounting for 35.9% of the total 23.6 million tons plants to be equipped with the Best Available Techniques of crops processed annually. In 2015, wheat yielded 39.6% (BAT) to reach specific emissions limit values. To do so, of the 10.7 million tons of starch produced [3]. Starch is industries may have to implement new measures [11]. Cur- a naturally occurring polymer with numerous industrial rently, as a method to reduce the COD emissions, the sus- applications, including food, paper, cosmetic, pharmaceuti- pended solids present in the wastewater are collected and cal, textile, and detergent industries [4, 5]. Several methods sold as feed additive. Usually, the solid recovery is achieved of starch production from wheat were reviewed by Van Der by decantation, but this operation results in low efficiency Borght et al. [6]. These methods can demand up to 1.8 parts and requires long settling times. Furthermore, the use of of water per part of wheat flour [6]. This wastewater has flocculants is not recommended since it may have adverse to be treated at the end of the process, which represents an effects on animals [9]. environmental challenge due to the large chemical oxygen In general, starch wastewater contains a significant amount of carbohydrates, protein, and nutrients. Instead Pedro F. Souza Filho of being sent to a wastewater treatment plant for microbial pedro.ferreira_de_souza_fi[email protected] removal, these nutrients can be recovered either through purification processes or through the cultivation of micro- 1 Swedish Centre for Resource Recovery, University of Borås, organisms for the production of useful biomaterials [10]. 501 90 Borås, Sweden

Vol.:(0123456789)1 3 Waste and Biomass Valorization Waste and Biomass Valorization

Filamentous fungi, for instance, grown in industrial waste Materials and Methods Acetic acid, lactic acid, glycerol, ethanol, glucose, and and [COD] f is the COD concentration at the end of the materials presented as an interesting alternative to the use other sugars were analyzed using a high-performance liq- cultivation. of refined sugars [12]. Antibiotics, enzymes, organic acids, Industrial Wastewater uid chromatography (HPLC) by a hydrogen-ion based ion- Samples were taken during cultivation to monitor the and ethanol are among the assortment of chemicals which exchange column (Aminex HPX-87H, Bio-Rad, Hercules, consumption of sugars and production of metabolites. New can be produced by filamentous fungi. Moreover, the fungal Wastewater from wheat-starch production was kindly pro- CA, USA) at 60 °C, using 0.6 mL/min of 0.5 mM H2SO4 tests were carried out to study the influence of the addition biomass with high content of proteins and lipids is also a vided by Interstarch GmbH (Elsteraue, Germany). All the solution as the eluent. The NREL analytical procedure [17] of enzymes on fungal cultivation. Because of the high con- valuable product. It is easily harvested and can be used .g. wastewater from the starch plant is sent to two decanters was used to determine carbohydrates and lignin in the sus- tent of starch and fibers in the streams, 0.1 mL of cellulase animal feed [12, 13]. in series, before sending the clarified liquid to the munici- pended solids of the waste streams. The liquid containing the Cellic® CTec2 (94 FPU/mL, Novozymes, Denmark) and Among the variety of filamentous fungi, ascomycetes pal wastewater plant. A block flow diagram is presented in soluble solids was analyzed in terms of carbohydrates and 0.1 mL of amyloglucosidase from Aspergillus niger (300 U/ have been described in the literature as a group of versatile Fig. 1. Three samples were used in this study: ED1, col- their degradation products, using another NREL procedure mL, Sigma, USA) were added to the Erlenmeyer flasks at the microorganisms which can grow using different substrates, lected after the first decanter; ED2, collected after the sec- [18]. A HPLC lead (II)-based column (Aminex HPX-87P, start of the cultivation. All the other parameters were kept as including wheat straw and wheat bran. Aspergillus oryzae ond decanter; and SS, the mixture of the solid-rich streams Bio-Rad) operating at 85 °C and 0.6 mL/min of ultrapure previously mentioned. All the cultivation experiments were is an example of ascomycetes, and one of the most studied of both decanters. water as the eluent was used to detect the sugars which were conducted in triplicates. fungal species for industrial applications [12]. The ability to released in the NREL procedures. A refractive index detec- grow on low or negative cost substrates, such as agro-wastes Microorganisms tor (Waters 2414, USA) was used to identify and quantify Statistical Analysis and residues has also drawn attention to the zygomycetes the components. [14]. Rhizopus oryzae, a zygomycete fungus, can poten- Rhizopus oryzae CCUG 28958 (Culture Collection of the The sucrose/D-fructose/D-glucose assay kit, the total The data collected from the experiments were statistically tially replace fish feed, as reported in the literature [15]. Jin University of Gothenburg, Sweden) and Aspergillus oryzae starch assay kit, and the raffinose/D-galactose assay kit from analyzed using the software MINITAB® version 17.1.0. et al. [16] investigated different filamentous fungi, includ- var. oryzae CBS 819.72 (Centraalbureau voor Schimmelcul- Megazyme (Ireland) were used to obtain a soluble carbohy- Analyses of variance (ANOVA) were performed using gen- ing Fusarium spp., Trichoderma spp., Aspergillus spp., and tures, Utrecht, The Netherlands) were grown on a solid PDA drate profile of the wastewater. The InKjel P digestor and eral linear models at 95% of confidence level. The graphs Rhizopus spp., to produce biomass and clean water from (Potato Dextrose Agar) medium in 25-mL glass bottles for the Behrotest S1 distiller (Behr Labor-Technik, Germany) present the mean value of the measurements with an error the effluent of a wheat- and corn-starch plant, obtaining a 5 days at 30 °C without agitation. After the mycelial growth were used to determine the total Kjeldahl nitrogen (TKN) bar representing the range of standard deviations. removal of up to 92% of the initial COD. and sporulation, the slants were maintained at 5 °C (up to content for solid and liquid samples. Protein content in the The goal of the present paper is to use the wastewater 30 days) until use. Inoculation was started by pouring 12 mL fungal biomass was estimated by multiplying the nitrogen (including the suspended solids) of a wheat-starch plant as of sterile water on each slant. The slants were then closed, content found in the sample by a nitrogen-to-protein fac- Results and Discussion a substrate for filamentous fungi and collect the protein-rich and manually agitated to release the spores. The obtained tor (6.25) [19]. AQUANAL™—professional tube test COD biomass. This effluent has not been investigated before for spore suspension was used to inoculate the liquid medium (1000–15,000 mg/L) (Sigma-Aldrich, USA) was used to Starch plants produce large amounts of wastewater. Treat- the cultivation of A. oryzae and R. oryzae. Initially, three to reach a concentration between 2 × 105 and 5 × 105 spores/ determine the chemical oxygen demand in liquid samples. ment and disposal of this wastewater represents an extra streams from the wheat-starch wastewater treatment plant mL in the final medium. All experiments were conducted in duplicates. cost for the industry. Recently, more strict regulations have (Fig. 1) were characterized in terms of composition of car- led the sector to search for alternative methods to reduce bohydrates and nitrogen; the side streams were then tested as Analytical Methods Fungal Cultivation the impacts on the environment as well as the costs (or media for the growth of the two microorganisms. Addition even bring new incomes) to the business. In this work, the of enzymes (cellulase and amylase) to help the digestion of Total solids content of the wastewater samples was deter- Aspergillus oryzae and R. oryzae were cultivated in 250 mL wastewater of a wheat-starch plant was investigated as the complex carbohydrates was also investigated. The protein mined by drying 10 mL of the homogenized samples at baffled Erlenmeyer flasks, covered with a cotton plug, con- substrate for the cultivation of A. oryzae and R. oryzae. content of the collected biomass and the COD of the residual 105 °C until constant weight. To determine the ash content, taining 50 mL of culture medium (the wastewater without The microorganisms effectively used the nutrients of the liquid were determined and presented. the solids were placed in a muffle furnace (Gallenkamp, UK) dilution or nutritional supplementation). Sterilization of the agro-industrial residue, reducing the impact of the indus- at 575 °C for 5 h. Thereafter, the fraction of suspended solids media was carried out at 121 °C for 20 min before inocula- trial emission while producing value-added bio products. In was determined by centrifuging at 3000×g for 10 min and tion with 5 mL of the spore suspension at room temperature. this section, the results of the characterization of the waste drying the obtained solid at 105 °C until constant weight. The cultivation was carried out in a water bath shaker at water, fungal growth on wastewater, and COD removal are The difference between suspended and total solids yields the 35 °C and 120 rpm. The original pH of the ED1 and ED2 presented and discussed. amount of soluble solids. samples was used and not adjusted during the cultivation. The SS sample was tested at the original pH and pH adjusted Starch Plant’s Wastewater to 5.5 prior to cultivation. At the end of each cultivation, the produced fungal biomass was firstly collected using a Three samples of wastewater from a wheat-starch plant kitchen sieve and then freeze dried, and weighed. were used in this study: ED1 (clarified liquid from the first Fig. 1 Block flow diagram COD removal efficiency was also determined, according decanter), ED2 (clarified liquid from the second decanter), (BFD) representing the prelimi- to Eq. 1. and SS (the mixture of the solid-rich streams of both nary wastewater treatment plant and the streams in the wheat- decanters). The composition on solids, ash, carbohydrates, [COD] − [COD] starch facility o f 100 lignin, and nitrogen were determined for the three samples 𝜂𝜂COD = × % (1) [COD]o (Table 1). ED 1, ED 2, and SS had a total of carbohydrates of 17.29 ± 0.20, 20.30 ± 0.40, and 93.30 ± 4.72 g/L respec- where ηCOD is the percentage COD removal efficiency, tively. Monomeric sugars, however, represented only a [COD] 0 is the initial COD concentration of the medium, small fraction of this total (18% in ED1, 15% in ED2,

1 3 1 3 Waste and Biomass Valorization Waste and Biomass Valorization

Table 1 Characterization of the wheat-starch industry’s wastewater the same range as the content of the samples ED1 and ED2 Fig. 2 Concentration profiles 2 A 2 B streams (2.03 and 4.28 g/l, respectively). of glucose (cross), other sugars (filled square), ethanol (filled ED1 ED2 SS Within 95% of confidence level, the Total Kjeldahl nitro- circle), and lactic acid (filled gen (TKN) concentrations of the three samples were not triangle) in ED1 medium using: pH 5.28 5.29 3.85 1 1 a A. oryzae; b R. oryzae; c A. a a a significantly different (1.25–1.40 g/L). The COD of the Total Kjeldahl nitrogen 1.25 ± 0.06 1.34 ± 0.32 1.40 ± 0.04 SS sample (58.7 ± 2.1 g/L) was significantly (95% confi- oryzae and enzyme supplemen- (g/L) tation; and d R. oryzae with a b c dence level) higher than those of ED1 and ED2 (27.3 ± 1.2 Total solids (g/L) 28.52 ± 0.08 32.19 ± 0.43 96.68 ± 0.70 enzyme supplementation Concentration (g/L) and 36.7 ± 4.7 g/L, respectively). Comparatively, a maize 0 0 Ashes (g/L) 1.93 ± 0.02a 1.90 ± 0.11ab 1.80 ± 0.03b starch wastewater had a TKN content of 12.2 mg/L and 0 6 12 18 24 0 6 12 18 24 Suspended solids (g/L) 3.70 ± 0.11a 6.78 ± 0.02b 71.79 ± 0.14c a ab b a COD concentration of 4.39 g/L [10]. The COD and the Glucan (mg/g) 446 ± 9 600 ± 73 857 ± 90 12 C 12 D 1 a b c suspended solids concentrations of tapioca starch waste- Starch (mg/g) 53 ± 4 361 ± 2 771 ± 6 a a a water were determined as being in the range of 13.5–25.0 Xylan (mg/g) 103 ± 49 31 ± 0 65 ± 19 8 8 a and 2.2–4.2 g/L, respectively [21]. Comparable values were Galactan (mg/g) ND ND 3 ± 0 a found in this study for the ED1 wastewater. Arabinan (mg/g) ND ND 51 ± 8 4 Presence of some microbial metabolites such as ethanol, 4 Mannan (mg/g) ND ND ND a ab b acetic acid, lactic acid, and glycerol was also detected in the Lignin (mg/g) 102 ± 7 76 ± 18 49 ± 16 Concentration (g/L) 0 0 a a a tested side streams. This might be caused by the microbial Soluble solids (g/L) 24.82 ± 0.14 25.41 ± 0.43 24.88 ± 0.71 0 6 12 18 24 0 6 12 18 24 a a b activity in the wastewater since the effluent was not kept Glucose (g/L) 1.68 ± 0.02 1.62 ± 0.08 3.65 ± 0.29 Time (h) Time (h) a a a under sterile conditions in the facility. The high concentra- Fructose (g/L) 1.43 ± 0.02 1.44 ± 0.01 1.41 ± 0.30 a a a tion of metabolites, like lactic acid, in the SS sample may Galactose (g/L) 0.50 ± 0.03 0.53 ± 0.01 0.56 ± 0.01 a a also have contributed to the low pH of this sample (3.85). same trends for glucose consumption and ethanol produc- SS was examined in two conditions: at its original pH Raffinose (g/L) 0.24 ± 0.08 0.04 ± 0.03 ND a b c tion were observed (Fig. 3). Glucan content was signifi- (3.85) and at pH 5.5. SS contained the highest amount of Glucan (g/L) 7.22 ± 0.07 8.07 ± 0.17 10.44 ± 0.35 a a b Cultivation of Filamentous Fungi cantly higher in ED2 compared to ED1. When enzymes insoluble material among the three samples used in this Xylan (g/L) 2.32 ± 0.02 2.41 ± 0.03 3.91 ± 0.04 a a a on the Wastewaters were added in the ED2 medium, the glucose concentra- work. Most of the insoluble fraction was composed of starch Galactan (g/L) 0.08 ± 0.03 0.08 ± 0.01 0.07 ± 0.01 a a b tion reached values above 14 g/L. Ethanol and lactic acid (771 mg/g). A. oryzae is a well-known α-amylase enzyme Arabinan (g/L) 1.75 ± 0.07 1.81 ± 0.04 3.19 ± 0.04 a a Aspergillus oryzae R. oryzae maximum concentrations were obtained by cultivating R. producer [22]. The high amount of starch in the SS may Mannan (g/L) 0.04 ± 0.01 0.02 ± 0.00 ND and were cultivated in the three a a b oryzae with enzyme supplementation. In the ED2 medium, have stimulated the production of this enzyme by the fun- Lactic acid (g/L) 1.29 ± 0.02 1.29 ± 0.01 3.23 ± 0.14 effluents of a wheat-starch plant. ED1 and ED2 were used a a b the final lactic acid concentration (8.56 ± 0.53 g/L) was sig- gus. At pH 3.85, the rate of the hydrolysis of starch became Acetic acid (g/L) 1.06 ± 0.07 1.04 ± 0.05 1.48 ± 0.13 with no pH adjustment. Cultivation of the fungi in SS was Glycerol (g/L) 0.23 ± 0.02a 0.24 ± 0.01a 0.59 ± 0.00b performed both with and without pH adjustment (5.5 and nificantly (95% confidence interval) higher than in the ED1 higher than the uptake of glucose by the microorganism after Ethanol (g/L) 0.29 ± 0.01a 0.68 ± 0.03b 2.04 ± 0.15c 3.85, respectively). Samples were taken at 0, 3, 6, 12, and medium (5.85 ± 0.82 g/L). Considering the same confidence 12 h (Fig. 4a), which led to an increase in concentration A. oryzae COD (g/L) 27.3 ± 1.2a 36.7 ± 4.7a 58.7 ± 2.1b 24 h, and analyzed. When growing in ED1, the interval, the final ethanol concentration in the ED2 medium of this sugar. The low pH might have contributed to this lag phase of substrate consumption lasted only 3 h whereas, (3.17 ± 0.91 g/L, see Table 2) was not different from the one phenomenon by decreasing the metabolic activity of the Values in any line which are not followed by the same letter are sig- for R. oryzae, the same lag phase took 6 h. This might rep- obtained in ED1 stream (2.19 ± 0.23 g/L). microorganism. p nificantly different ( < 0.05) resent a better capacity of the Aspergillus strain to adapt to ND not detected Rhizopus the medium than the . Addition of cellulase and Fig. 3 Concentration profiles 1 Part of the measured glucans 3 A 3 B amyloglucosidase in the side stream increased the lag phase of glucose (cross), other sugars for both microorganisms, taking 6 and 12 h for A. oryzae and (filled square), ethanol (filled and 5% in SS). In the ED1 and ED2 samples, most of the R. oryzae, respectively. The longer lag phase observed when circle), and lactic acid (filled 2 2 triangle) in ED2 medium using: carbohydrates were present in the soluble form. The con- the enzymes were added may have been a consequence of a A. oryzae; b R. oryzae; c A. centration of soluble carbohydrates was in the same range the presence of preservatives in the enzyme cocktails. When oryzae with enzyme supple- 1 1 for the three samples. On the other hand, in the SS sample, supplemented with enzymes, the initial glucose concentra- mentation; and d R. oryzae with

enzyme supplementation Concentration (g/L) most of the carbohydrates were present in the insoluble tion increased to 10.37 ± 0.47 g/L. From the characterization 0 0 form and approximately 60% of the carbohydrates in the of this side stream, the maximum concentration of glucose 0 6 12 18 24 0 6 12 18 24 SS were suspended starch. The results also revealed that which could be obtained from the complete hydrolysis of the the content of hexoses was higher than the pentoses, with glucans was 11.47 g/L. In all the cases, the glucose was con- 18 C 18 D only a small proportion of xylose and arabinose among sumed until its near depletion within 24 h. The other sugars the carbohydrates. started being consumed after glucose. Moreover, R. oryzae 12 12 Nasr et al. [10] reported the wastewater of a maize starch cultivation yielded the highest ethanol and lactic acid con- plant containing 3.88 g/L of carbohydrates, with less than centration using ED1 (2.19 ± 0.23 g/L and 5.85 ± 0.82 g/L, 6 6 30% soluble, which represents a much more diluted efflu- respectively). The profile of metabolites during the cultiva- ent compared to the ones tested in this study. Jin et al. [20] tion of the ED1 sample is shown in Fig. 2. Concentration (g/L) 0 0 worked with starch processing wastewater with insoluble The results of the cultivation in ED2 stream were simi- 0 6 12 18 24 0 6 12 18 24 carbohydrate content of 2.36–3.55 g/L. These values are in lar to those obtained in ED1 for both microorganisms. The Time (h) Time (h)

1 3 1 3 Waste and Biomass Valorization Waste and Biomass Valorization

Table 2 Ethanol concentration at 24 h cultivation of the filamentous The glucose concentration significantly (95% confi- maximum ethanol concentration was obtained by A. ory- Fungal Biomass Production fungi in the ED1 and ED2 media dence level) increased during the first 6 h of cultivation of zae (5.45 ± 0.22 g/L) when using SS at pH 3.85. Medium Microorganism Enzyme sup- Ethanol concentra- R. oryzae at both pH conditions. This result indicates the When cellulase and amyloglucosidase were added to The biomass produced during cultivation was harvested and plementation tion at 24 h (g/L)* zygomycete, similarly to A. oryzae, produced and released the SS medium, glucose concentration reached a maxi- weighted. Protein content, as an important characteristic for enzymes in the medium to degrade the glucose oligomers. mum of 86.33 ± 1.09 g/L 12 h after the beginning of the ED1 A. oryzae No 1.77 ± 0.06a making the biomass eligible for formulation of animal feed, The enzyme activity, however, was not as intense as for the cultivation of R. oryzae at pH 3.85 (Fig. 4f). The aver- Yes 1.92 ± 0.09a was analyzed using the Kjeldahl method. Additional compo- ascomycete strain, given that the glucose concentration did age value was slightly higher than the theoretical yield R. oryzae No 0.70 ± 0.05b nents of the fungal biomass may include lipids and structural not increase as much and the hydrolysis rate did not over- from the complete glucan hydrolysis of the SS sample R. ory- Yes 2.19 ± 0.23ac carbohydrates (e.g. the chitosan and chitin present in come the glucose consumption by the microorganism after (83.61 ± 0.49 g/L), although the values are not signifi- zae ED2 A. oryzae No 2.65 ± 0.18c cell wall) [15]. The results are presented in Table 3. The c lag phase (Fig. 4b, d). cantly different (95% confidence interval). Moreover, dur- A. ory- Yes 2.97 ± 0.17 highest biomass production was achieved by growing d At pH 5.5, A. oryzae was not able to fully consume the ing the cultivation, no significant productions of ethanol zae R. oryzae No 1.36 ± 0.04 in SS at pH 3.85 (49.16 ± 5.96 g/L) while the biomass acd glucose in 24 h and 2.86 ± 0.39 g/L of glucose were left at and lactic acid were detected, indicating no fungal activity. R. oryzae Yes 3.17 ± 0.91 with the highest protein content was the grown in the end of the cultivation (Fig. 4c). This may be because The high concentration of sugars combined with the low the ED2 with enzyme supplementation (50.76 ± 0.71% w/w). *Values not followed by the same letter are significantly different of production of amylolytic enzymes by the fungus, which pH may have inhibited the growth of the strain. In the Although a very high amount of biomass was obtained using p ( < 0.05) kept hydrolyzing starch. Without enzyme supplementation, other cases which there was enzyme supplementation to SS, the protein content was very low (below 17% w/w). This the pH-adjusted SS medium, a reduction was observed may be explained by the high amount of suspended solids Fig. 4 Concentration profiles 6 A 6 B in the glucose concentration accompanied by the produc- in the sample, and the capacity of the filamentous fungi to of glucose (cross), other sugars tion of ethanol (both fungi) and lactic acid ( R. oryzae) at use the solids as a support to grow. The ability of filamen- (filled square), ethanol (filled the last 12 h of the cultivation (Fig. 4e, g, h). Most part tous fungi to improve the flocculation of activated sludge circle), and lactic acid (filled 4 4 triangle) in SS medium without of the glucose, however, remained in the medium. Addi- by the formation of a mycelial network has been previously enzyme supplementation using: tionally, for the same medium, although an upward trend reported [23]. These microorganisms reinforce the structure 2 2 a A. oryzae (pH 3.85); b R. ory- was detected for the concentration of the sugars other than of the floc, resulting in larger and stronger flocs. The con- zae (pH 3.85); A. oryzae (pH c glucose, the difference was not significant considering a sequence is better sludge settling and low turbidity of the 5.5); d R. oryzae (pH 5.5); and Concentration (g/L) 0 0 with enzyme supplementation e 95% confidence interval. sewage effluent. Thus, it can be inferred that the suspended A. oryzae (pH 3.85); f R. oryzae 0 6 12 18 24 0 6 12 18 24 Maximum ethanol concentration (10.91 ± 1.55 g/L) solids, with high content of carbohydrates, were wrapped (pH 3.85); g A. oryzae (pH 5.5); was obtained at pH 5.5 when cultivating R. oryzae. This by the fungal mycelium, resulting in a biomass with low R. oryzae (pH 5.5) 5 C 5 D h cultivation also resulted in the maximum glucose reduc- concentration of proteins. 4 4 tion (compared to the highest concentration achieved in Jin et al. [24] reported a dry biomass production of 3 3 the cultivation), yielding a 26% decrease in the glucose 6.05 g/L containing 36.58% protein from the cultivation of concentration. A. oryzae in an effluent of starch production from corn and 2 2 1 1

Concentration (g/L) Table 3 0 0 Biomass concentration at the end of the cultivation and its respective protein concentration 0 6 12 18 24 0 6 12 18 24 Medium Microorganism Enzyme supple- Dry biomass (g/L)* Protein (% w/w)* Nitrogen recovery (%)* mentation 8 E 8 F 80 80 ED1 A. oryzae No 9.58 ± 3.54ab 36.38 ± 0.75ab 44.75 ± 0.58ab 6 6 ab ab cd 60 60 Yes 10.19 ± 2.67 36.70 ± 0.85 48.02 ± 0.45 R. oryzae No 9.55 ± 0.73a 36.65 ± 0.77ab 44.94 ± 0.17a 4 4 40 40 Yes 8.25 ± 1.07a 43.88 ± 0.93c 46.49 ± 0.23e 2 20 2 20 ED2 A. oryzae No 14.39 ± 2.75abc 34.26 ± 0.79a 58.69 ± 0.46f Glucose (g/L) Yes 10.25 ± 1.54ab 38.72 ± 1.29b 47.25 ± 0.33cde Concentration (g/L) 0 0 0 0 R. oryzae b a g 0 6 12 18 24 0 6 12 18 24 No 13.33 ± 1.05 33.95 ± 0.93 53.88 ± 0.23 Yes 7.83 ± 0.23a 50.76 ± 0.71d 47.32 ± 0.13ch G H SS (pH 3.85) A. oryzae No 49.16 ± 5.96d 11.21 ± 0.31e 62.98 ± 0.95i 12 12 de fg j 80 80 Yes 43.47 ± 7.04 15.57 ± 0.72 77.35 ± 1.13 R. oryzae No 25.20 ± 0.21f 14.73 ± 0.64fg 42.42 ± 0.11k 8 60 8 60 ef fg b Yes 26.08 ± 1.87 14.70 ± 0.62 43.81 ± 0.32 40 40 SS (pH 5.5) A. oryzae No 44.13 ± 0.88d 12.48 ± 0.22h 62.94 ± 0.15i 4 4 20 20 Yes 32.56 ± 6.45cdef 13.39 ± 0.55fh 49.83 ± 1.04d Glucose (g/L) R. oryzae No 25.89 ± 0.85f 15.79 ± 0.59g 46.72 ± 0.17eh

Concentration (g/L) 0 0 0 0 d g j 0 6 12 18 24 0 6 12 18 24 Yes 38.69 ± 3.67 16.87 ± 0.66 74.59 ± 0.60 Time (h) Time (h) *Values in any column which are not followed by the same letter are significantly different (p < 0.05)

1 3 1 3 Waste and Biomass Valorization Waste and Biomass Valorization wheat. These values are in agreement with the ones obtained recovery (42.42 ± 0.11%). Maximum nitrogen recovery by consumed. Ethanol and lactic acid production followed the Conclusions for ED1 and ED2 samples in this paper. Cultivation of Rhiz- R. oryzae was obtained when using the same stream (i.e. SS) same trends; with little variations after 72 h. High lactic opus oligosporus in starch processing wastewater yielded but at higher pH (5.5). In this case, with the addition of the acid concentration (above 10 g/L) may have inhibited fungal The effluent of a wheat-starch plant was treated by A. oryzae around 5 g/L of dry biomass [20] while less than 2 g/L of R. enzymes, the nitrogen recovery was 74.59 ± 0.60%. activity, and stopped the fermentation. and R. oryzae. These fungi can convert the organic material arrhizhus dry biomass were obtained from the cultivation in After 5 days, the COD of the SS sample treated with A. to protein sources for e.g. animal feed. A large amount of wastewater of a potato chips plant (20–30 g COD/L) [25]. COD removal oryzae was 42.0 g/L, which represents a removal efficiency carbohydrates was identified in all the wastewater streams Within a 95% confidence level, A. oryzae and R. oryzae of only 28.4%. The high residual COD is the result of the (ED1, ED2, and SS). The microorganisms consumed the did not yield different biomass concentration when culti- After 24 h cultivation, less than 60% and 50% of the COD of high ethanol production which might be the result of poor sugars and hydrolyzed the long-chain carbohydrates in the vated in the ED1 and ED2 samples when comparing the the ED1 and ED2 were removed by A. oryzae, respectively. mass transfer in the SS medium, the last being a consequence wastewater, and produced a protein-rich biomass. A. oryzae biomasses grown in the same sample and enzyme supple- In both cases, A. oryzae was more efficient than R. oryzae of the high content of solids. The characteristics of the sam- showed high capacity to hydrolyze starch although a long mentation. All the values are in the range of 7.83–14.39 g/L. in removing COD of the samples. Cultivation of microor- ple make it difficult for oxygen to be dissolved, creating an lag phase was found when it was cultivated in a high sugar- The biomass production average was superior for A. oryzae, ganisms in the SS wastewater did not result in a detectable anaerobic environment. This leads to a high production of concentration medium. More than 80% of the initial COD of but in most of the cases the protein content of the R. oryzae removal of COD. In order to achieve a higher efficiency of ethanol, explaining the high COD. However, even though the the clarified streams were consumed by the microorganisms. biomass was numerically superior. A. oryzae did, however, COD removal, the cultivation was carried out for 72 h in all COD remains high, the considerable ethanol concentration The high concentration of ethanol obtained in the solid-rich produce more biomass than R. oryzae when the fungi were the samples, using A. oryzae and enzyme supplementation. in the fermented broth suggests purification of the alcohol sample represents an opportunity to produce another product grown in SS medium without enzyme supplementation (both The COD removal was determined to be 84.88 ± 1.25 and can be a worthwhile strategy. Therefore, the stream should from the same treatment process. pH). This result might be explained by a more entangled 84.33 ± 3.6% for ED1 and ED2, respectively. not be considered as a wastewater. growth of A. oryzae with the SS insoluble material com- For the SS wastewater (pH 3.85), after the 72 h cultiva- After 5 days, 11.95 ± 0.46 g/L of dry A. oryzae biomass Acknowledgements This work was financed by the Coordination for pared to R. oryzae. A. oryzae might have had higher affinity tion there was still no detected reduction in COD concentra- and 22.69 ± 1.93 g/L of dry R. oryzae biomass were obtained the Improvement of Higher Education Personnel (CAPES-Brazil) and Procon AB (Sweden) in collaboration with Interstarch GmbH. with the solid material, growing more attached to it. This can tion. Further investigation was then carried out by monitor- in the SS sample. The result for R. oryzae is similar to the A. oryzae also explain why was more efficient on degrad- ing the glucose concentration during the cultivation in the one obtained for 24 h cultivation. The lower amount of bio- Open Access This article is distributed under the terms of the Crea- ing the glucans and releasing glucose in the medium. The SS medium. The results are presented in Fig. 5, and show mass of the ascomycete compared to the 24 h experiment, tive Commons Attribution 4.0 International License (http://creativeco biomass concentrations at the end of the fermentation, when that, for A. oryzae, the lag phase lasted for 72 h, and between however, may have been caused by the degradation of the mmons.org/licen ses/by/4.0/ ), which permits unrestricted use, distribu- enzymes were added to the SS medium, are not statistically 72 and 120 h, the glucose concentration was reduced to fibers used as support for the growth of the microorganisms tion, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the different due to the large standard deviations obtained for 5.04 ± 0.04 g/L. Ethanol started being produced after 72 h, (The insoluble carbohydrates which were entrapped inside Creative Commons license, and indicate if changes were made. both microorganisms. This is a typical problem when work- and reached 20.04 ± 0.49 g/L at 120 h. This represents a the fungal biomass). The higher content of protein in the ing with high solid content substrates due to the huge impact very high concentration for an ethanol production plant. biomass supports this conclusion (35.72 ± 1.24% w/w). slight variations in factors e.g. agitation causes in the results. The R. oryzae cultivation followed a different profile, with For cultivation in the ED1 and the ED2 samples supple- the glucose being consumed between 24 and 72 h, and the References mented with enzymes, the collected biomass of R. oryzae remaining glucose (approximately 35 g/L) not being further Industrial Perspectives for the Wastewater contained significantly higher protein content than A. oryzae Treatment by Filamentous Fungi 1. FAO: Save and Grow in Practice: Maize, Rice, Wheat: A Guide biomass (respectively for ED1 and ED2, 43.88 ± 0.93 and to Sustainable Cereal Production. FAQ, Rome, 2016 50.76 ± 0.71 for R. oryzae, and 36.70 ± 0.85 and 38.72 ± 1.29 Further investigation on the production of ethanol and 2. Fava, F., Zanaroli, G., Vannini, L., Guerzoni, E., Bordoni, A., A Viaggi, D., Robertson, J., Waldron, K., Bald, C., Esturo, A., Tal- A. oryzae R. oryzae 20 80 for ). This indicates was more efficient organic acids using this wastewater seems to be promising. ens, C., Tueros, I., Cebrian, M., Sebök, A., Kuti, T., Broeze, J., on converting the released sugars to proteins than A. oryzae. 15 60 This would lead to a larger portfolio of products from the Macias, M., Brendle, H.-G.: New advances in the integrated man- Without enzyme supplementation, both strains produced 10 40 residue using the same microorganism, creating a flexible agement of food processing by-products in Europe: sustainable biomass with similar protein content. Moreover, addition of process which can be operated according to market demands. exploitation of fruit and cereal processing by-products with the 5 20 production of new food products (NAMASTE EU). New Biotech- enzyme did not change the protein content of the Aspergillus Currently, the studied wastewater is sent to the municipal 0 0 nol. 30(6), 647–655 (2013) biomass, and only R. oryzae effectively managed to con- wastewater treatment plant, which results in extra costs for 3. Starch.eu (2017) http://www.starch.eu/ Accessed 20 Oct 2017 Concentration (g/L) 0 20 40 60 80 100 120 vert the extra sugar into more protein. Protein content is an the industrial facility. Mahboubi et al. [26] have highlighted 4. Radley, J.A.: Industrial Uses of Starch and its Derivatives, 1st edn. Time (h) important factor to be considered if the biomass is intended how wastes can contribute to the income of a company and Applied Science Publishers LTD, Essex (1976) 5. Ahmadi-Abhari, S., Woortman, A.J.J., Hamer, R.J., Loos, K.: to be used as animal feed. B avoid the high cost related to the wastewater treatment. The 20 80 Rheological properties of wheat starch influenced by amylose- Supplementation with amyloglucosidase and cellulose production of ethanol, for instance, can be studied as part lysophosphatidylcholine complexation at different gelation phases. 15 only resulted in higher biomass production for the cultiva- 60 of an integrated process, in which the obtained broth can Carbohyd. Polym. 122, 197–201 (2015) tion of R. oryzae in ED2 and SS at pH 5.5 (compared to the 10 40 be filtered for the collection of the fungal biomass and the 6. Van Der Borght, A., Goesaert, H., Veraverbeke, W.S., Delcour, J.A.: Fractionation of wheat and wheat flour into starch and glu- cultivation without addition of enzymes). Similarly to the 5 20 liquid fraction can be processed for purification of ethanol.

Glucose (g/L) Glucose ten: overview of the main processes and the factors involved. J. previous discussion, R. oryzae again demonstrated higher 0 0 Since the solid fraction of the streams was found to be rich Cereal Sci. 41(3), 221–237 (2005) capacity to use the extra sugar and efficiently convert it to 0 20 40 60 80 100 120 in carbohydrates and easily hydrolyzed, the three streams 7. Pu, S.-Y., Qin, L.-L., Che, J.-P., Zhang, B.-R., Xu, M.: Preparation biomass. can be mixed together for cultivation of A. oryzae, yielding and application of a novel bioflocculant by two strains of Rhizopus Time (h) sp. using potato starch wastewater as nutrilite. Bioresour. Technol. Up to 77.35 ± 1.13% of the nitrogen initially present in fungal biomass and ethanol. Such scenario would work as 162, 184–191 (2014) the SS is converted into biomass by A. oryzae when growing a last step for the treatment of the wastewater. The clarified Fig. 5 Concentration profiles of glucose (cross), ethanol (filled circle) 8. Yu, D., Li, C., Wang, L., Zhang, J., Liu, J., Wei, Y.: Multiple effects of trace elements on methanogenesis in a two-phase at pH 3.85 and with addition of enzymes (see Table 3). At and lactic acid (filled triangle) from a A. oryzae and b R. oryzae culti- ethanol-free effluent can even be partially recycled to the these same conditions, R. oryzae yielded the lowest nitrogen vation in SS medium (pH 3.85) with enzyme supplementation beginning of the process.

1 3 1 3 Waste and Biomass Valorization

anaerobic membrane bioreactor treating starch wastewater. Appl. and Lignin in Biomass. National Renewable Energy Laboratory, Microbiol. Biotechnol. 100, 1–12 (2016) Golden (2008) 9. Deng, S., Bai, R., Hu, X., Luo, Q.: Characteristics of a biofloc- 18. Sluiter, A., Hames, B., Ruiz, R., Scarlata, C., Sluiter, J., Temple- culant produced by Bacillus mucilaginosus and its use in starch ton, D.: Determination of Sugars, Byproducts, and Degradation wastewater treatment. App. Microbiol. Biotechnol. 60(5), 588– Products in Liquid Fraction Process Samples. National Renewable 593 (2003) Energy Laboratory, Golden (2006) 10. Nasr, M., Tawfik, A., Suzuki, M., Ookawara, S.: Mathematical 19. Souza Filho, P.F., Zamani, A., Taherzadeh, M.J.: Production of modeling of bio-hydrogen production from starch wastewater via edible fungi from potato protein liquor (PPL) in airlift bioreactor. up-flow anaerobic staged reactor. Desalin. Water Treat. 54(1), Fermentation 3, 12 (2017) 50–58 (2015) 20. Jin, B., van Leeuwen, H.J., Patel, B., Doelle, H.W., Yu, Q.: Pro- 11. Bachmann, T.M., van der Kamp, J.: Environmental cost-benefit duction of fungal protein and glucoamylase by Rhizopus oligospo- analysis and the EU (European Union) industrial emissions direc- rus from starch processing wastewater. Process Biochem. 34(1), tive: exploring the societal efficiency of a DeNOx retrofit at a 59–65 (1999) coal-fired power plant. Energy 68, 125–139 (2014) 21. Annachhatre, A.P., Amatya, P.L.: UASB treatment of tapioca 12. Ferreira, J.A., Mahboubi, A., Lennartsson, P.R., Taherzadeh, M.J.: starch wastewater. J. Environ. Eng. 126(12), 1149–1152 (2000) Waste biorefineries using filamentous Ascomycetes fungi: present 22. Francis, F., Sabu, A., Nampoothiri, K.M., Ramachandran, S., status and future prospects. Bioresour. Technol. 215, 334–345 Ghosh, S., Szakacs, G., Pandey, A.: Use of response surface (2016) methodology for optimizing process parameters for the produc- 13. Nair, R.B., Taherzadeh, M.J.: Valorization of sugar-to-ethanol tion of α-amylase by Aspergillus oryzae. Biochem. Eng. J. 15(2), process waste vinasse: a novel biorefinery approach using edible 107–115 (2003) Ascomycetes filamentous fungi. Bioresour. Technol. 221, 469– 23. Murugesan, K., Selvam, A., Wong, J.W.C.: Flocculation and 476 (2016) dewaterability of chemically enhanced primary treatment sludge 14. Moustogianni, A., Bellou, S., Triantaphyllidou, I.E., Aggelis, by bioaugmentation with filamentous fungi. Bioresour. Technol. G.: Feasibility of raw glycerol conversion into single cell oil by 168, 198–203 (2014) zygomycetes under non-aseptic conditions. Biotechnol. Bioeng. 24. Jin, B., van Leeuwen, H.J., Patel, B., Yu, Q.: Utilisation of starch 112(4), 827–831 (2015) processing wastewater for production of microbial biomass protein 15. Ferreira, J.A., Lennartsson, P.R., Edebo, L., Taherzadeh, M.J.: and fungal α-amylase by Aspergillus oryzae. Bioresour. Technol. Zygomycetes-based biorefinery: present status and future pros- 66(3), 201–206 (1998) pects. Bioresour. Technol. 135, 523–532 (2013) 25. Huang, L.P., Dong, T., Chen, J.W., Li, N.: Biotechnological 16. Jin, B., van Leeuwen, J.H., Yu, Q., Patel, B.: Screening and selec- production of lactic acid integrated with fishmeal wastewater tion of microfungi for microbial biomass protein production and treatment by Rhizopus oryzae. Bioprocess Biosyst. Eng. 30(2), water reclamation from starch processing wastewater. J. Chem. 135–140 (2007) Technol. Biotechnol. 74(2), 106–110 (1999) 26. Mahboubi, A., Ferreira, J.A., Taherzadeh, M.J., Lennartsson, P.R.: 17. Sluiter, A., Hames, B., Ruiz, R., Scarlata, C., Sluiter, J., Temple- Value-added products from dairy waste using edible fungi. Waste ton, D., Crocker, D.: Determination of Structural Carbohydrates Manag. 59, 518–525 (2017) Paper 4

1 3 Souza Filho et al. Fungal Biol Biotechnol (2018) 5:5 https://doi.org/10.1186/s40694-018-0050-9 Fungal Biology and Biotechnology

RESEARCH Open Access Vegan-mycoprotein concentrate from pea-processing industry byproduct using edible filamentous fungi Pedro F. Souza Filho1*, Ramkumar B. Nair2, Dan Andersson3, Patrik R. Lennartsson1 and Mohammad J. Taherzadeh1

Abstract Background: Currently around one billion people in the world do not have access to a diet which provides enough protein and energy. However, the production of one of the main sources of protein, animal meat, causes severe impacts on the environment. The present study investigates the production of a vegan-mycoprotein concentrate from pea-industry byproduct (PpB), using edible filamentous fungi, with potential application in human nutrition. Edible fungal strains of Ascomycota (Aspergillus oryzae, Fusarium venenatum, Monascus purpureus, Neurospora interme- dia) and Zygomycota (Rhizopus oryzae) phyla were screened and selected for their protein production yield. Results: A. oryzae had the best performance among the tested fungi, with a protein yield of 0.26 g per g of pea- processing byproduct from the bench scale airlift bioreactor cultivation. It is estimated that by integrating the novel fungal process at an existing pea-processing industry, about 680 kg of fungal biomass attributing to about 38% of extra protein could be produced for each 1 metric ton of pea-processing byproduct. This study is the first of its kind to demonstrate the potential of the pea-processing byproduct to be used by filamentous fungi to produce vegan- mycoprotein for human food applications. Conclusion: The pea-processing byproduct (PpB) was proved to be an efficient medium for the growth of filamen- tous fungi to produce a vegan-protein concentrate. Moreover, an industrial scenario for the production of vegan- mycoprotein concentrate for human nutrition is proposed as an integrated process to the existing PPI production facilities. Keywords: Pea-processing byproduct, Edible filamentous fungi, Vegan-mycoprotein concentrate, Meat substitute

Background (PEM) can lead to conditions such as kwashiorkor and Attributed to the rise in population, urbanization, and marasmus. Additionally, the production of meat has a income, a steady growth in the consumption of protein heavy impact on the environment and makes a large con- from animal sources has been observed in developed tribution to the eutrophication process [3, 4]. In this con- countries over the last few decades [1]. Nevertheless, text, it is important to find an alternative, cheap, and less currently around one billion people in the world do not resource-consuming protein source to substitute meat have access to a diet which provides enough protein and or meat products. Fungal organisms such as mushrooms energy [2]. Lack of protein can result in severe health and truffles have traditionally been part of human nutri- problems such as growth failure, muscle weakness and an tion largely because of their flavor; however they can- impaired immune system. Protein-energy malnutrition not be considered as an important source of protein in comparison to meat-based sources [1, 5]. Considerable *Correspondence: pedro.ferreira_de_souza_fi[email protected] attention has been recently given to the use of filamen- 1 Swedish Centre for Resource Recovery, University of Borås, 50190 Borås, tous fungi as a commercial human food component; Sweden especially due to their high protein content with all the Full list of author information is available at the end of the article

© The Author(s) 2018. This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/ publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated. Souza Filho et al. Fungal Biol Biotechnol (2018) 5:5 Page 2 of 10 Souza Filho et al. Fungal Biol Biotechnol (2018) 5:5 Page 3 of 10

essential amino acids to human nutrition, easy digestibil- grains, like soy [19]. Pea proteins have also been dem- scaled-up in a bench airlift bioreactor, considering the changes in the gelatinized starch during cooling, when ity, low-fat content (cholesterol free), and the presence of onstrated as a useful ingredient in the formulation of industrial application potential of the process. the molecules recrystallize. This process leads to the pro- dietary fibers [6]. The fibre content (6% w/w) is also com- antihypertensive foods because of their antihypertensive duction of a starch with high resistance to the enzymatic parable with other vegetarian protein sources [7]. effects [16, 19, 20]. Methods attack, thus reducing the effectiveness of the microbial Several strains of edible filamentous fungi have been Pea proteins are commercialized in three forms: pea Substrate and enzymes metabolism [26–28]. The Erlenmeyer flasks were kept recognized as a traditional source of palatable food by flour, pea protein concentrate, and pea-protein isolate Pea-processing byproduct (PpB) used for this study was in a water bath shaker at 35 °C and 150 rpm (with a many societies around the globe, especially in Asia [8]. (PPI). The manufacturing of pea flour consists of the dry kindly provided by Protein Consulting AB (Sweden). The 9 mm orbital shaking radius). The pH of the sample was Rhizopus sp. has been used for centuries in the orien- milling of hulled peas, whereas pea protein concentrate powder was sieved through a pore size of 0.2–0.25 mm adjusted to 5.5 ± 0.1 prior to autoclaving by adding HCl tal cuisine in the preparation of fermented food such as is obtained by dry separation techniques. PPI production and was characterized using triplicate samples for its 1 M. The enzymes were added according to the substrate tempeh [9]. Aspergillus oryzae also has culinary applica- generally occurs by isoelectric precipitation at pH around carbohydrate, protein, ash, and moisture content. Cel- at a load of 150 U/g for α-amylase, 163 U/g for amylo- tions for the production of hamanatto, miso and shoyu. 4.5, followed by a membrane separation technique to lulase cocktail Cellic Ctec2 (Novozymes, Denmark) glucosidase, and 24 FPU/g for cellulase. At the end of Neurospora intermedia is used in the preparation of the increase the protein concentration, such as ultrafiltra- with 94 FPU/mL activity at 35 °C, amyloglucosidase the cultivations, the produced fungal mycelium was col- Indonesian staple food oncom [6]. Similarly, Monascus tion and diafiltration. PPI can be used in the preparation from Aspergillus niger (300 U/mL activity at 35 °C), and lected using a sieve, washed with ultra-pure water, dried purpureus has been used as coloring and flavoring agent of dairy-based beverages, sports and nutritional foods, α-amylase from Aspergillus oryzae (100 U/mg activity at at 70 °C and had the weight and protein content analysed. in food and beverages, as in the production of red yeast and other non-dairy sports products, such as vegan 35 °C) were supplied by Sigma-Aldrich Co. (Germany). Samples were taken during cultivation to follow the con- rice and rice wine [10–12]. Other applications of fila- style yogurts. Additionally, it can partially replace dairy sumption of sugars and the production of metabolites. mentous fungi include the production of several ingre- protein in therapeutic beverages and powders [19, 20]. Microorganisms All the cultivation experiments were conducted in dupli- dients for the food and beverage industries, especially Despite the high quality of the protein, the pea-process- Edible food-grade filamentous fungi were used in the cate and the mean values are presented with standard enzymes. In recent years, production of vitamins and ing byproduct (PpB) is considered to have poor func- present study. The Ascomycota strains were Neurospora deviations. polyunsaturated fatty acids by these microorganisms has tional properties. Therefore, its uses in food applications intermedia CBS 131.92 (Centraalbureau voor Schim- been receiving increased attention [13]. Single-cell pro- are limited and it is mainly produced as a byproduct of melcultures, Netherlands), Aspergillus oryzae var. ory- Scaling-up of the fungal cultivation in a bench scale airlift tein (SCP) can also be produced by filamentous fungi. the protein extraction process [21]. A novel and alterna- zae CBS 819.72, Monascus purpureus CBS 109.07, and reactor An example currently in the market is the filamentous tive approach to valorize this pea-processing byproduct Fusarium venenatum ATCC 20334 (American Type Cul- A 4.5-L airlift bioreactor (Belach Bioteknik, Sweden) fungus Fusarium venenatum, commercialized under the (PpB), as discussed in the present study is to convert it ture Collection, USA), and the Zygomycota strain was with a working volume of 3.5 L was used to scale-up the name Quorn™. The fungus is cultivated in a synthetic into a vegan-mycoprotein concentrate for human food Rhizopus oryzae CCUG 61147 (Culture Collection Uni- fungal cultivation process. The entire bioreactor and the medium with glucose, ammonium and supplemented applications, using edible strains of filamentous fungi. versity of Gothenburg, Sweden). All the fungal cultures draft tube were made of transparent borosilicate glass. with biotin. The costs associated with the substrate and Intake of mycoprotein may be beneficial to human health were maintained on potato dextrose agar (PDA) slants An internal loop with cylindrical geometry with 58 mm the lack of competition results in a market price for the [7, 22–24]. Several studies have investigated the choles- containing (in g/L) potato extract 4; glucose 20; agar 15 of diameter, 400 mm of height and 3.2 mm in thickness SCP that is higher than that of meat. Despite the high terol-lowering effects of mycoprotein; the results from and were renewed every 6 months. New PDA plates were was used to achieve the airlift-liquid circulation. Aeration price, the fungal SCP has found its place in the market these studies point to the same direction with reductions prepared via incubation for 3–5 days at 30 °C followed by at the rate of 0.42 v.v.m. (volumeair/volumemedia/min) was as a healthy substitute to meat, with its presence only in both total and LDL-cholesterol [7, 22]. The greatest storage at 4 °C. For preparing spore solution, PDA plates maintained throughout the cultivation, using a sintered in developed markets such as Europe and USA [6, 14]. benefits are seen in individuals with a higher cholesterol (72 h grown) were flooded with 20 mL sterile distilled stainless steel air-sparger with a pore size of 90 μm. Filtra- Nevertheless, the SCP from the mycelium of filamentous level at baseline and in hypercholesterolaemic subjects. water and the spores were released by gently agitating the tion of the inlet air was achieved by passing it through a fungi can be inexpensively produced when using cheap There is a difference between the macronutrients regard- mycelium with a disposable cell spreader. An inoculum polytetrafluoroethylene (PTFE) membrane filter (0.1 μm materials as substrates [15]. One such example is the pea ing satiety; protein is generally recognised as most sati- of 3 mL spore suspension (with a spore concentration of pore size, Whatman, Florham Park, NJ, USA). The cul- 5 6 processing industry byproduct that is being used in the ating, followed by carbohydrates and fat [25]. Compared 3.9 × 10 –3.8 × 10 spores/mL) per liter of the medium tivation was carried out at the natural pH of PpB media, present study. to other protein sources such as chicken, mycoprotein was used for the cultivations, unless otherwise specified. 6.1 (for 2% substrate) and 6.5 (for 3% substrate) with- Pea (Pisum sativum) is the second most important seems to be more satiating and hence have the possibil- For preparing fungal biomass inoculum, the spores were out any adjustments during the cultivation. An enzyme leguminous crop in the world with an annual produc- ity to decrease energy intake in subsequent meals [7, 23, inoculated into 100 mL YPD broth containing (in g/L) loading of 5 FPU of commercial cellulase enzyme com- tion above 17 million metric tons, finding its applications 24]. It is possible that fibres in mycoprotein, one-third glucose 20, peptone 20, and yeast extract 10. The culture plex (Cellic Ctec 2) per gram substrate was added (filter mainly as food and feed. Originally from western Asia chitin and two-thirds beta-glucan, might have a specific was incubated aerobically for 48 h at 35 °C and 125 rpm. sterilized) during the start of the cultivation (time 0). The and northern Africa, its production has spread to over effect on satiety [7]. Additionally mycoprotein appears to The fungal biomass was harvested at the end of the culti- fermentation was carried out at 35 ± 2 °C for 48 h, with 10 million hectares of farmlands, especially in Russia, affect the glycaemic response positively [7, 24]. The exact vation and used as the inoculum. Dry weight was deter- sample collection at every 12 h. China, Canada, Europe, Australia and the United States. mechanism that explains this is not known, but might be mined by drying at 105 °C overnight. Rich in protein, carbohydrate, dietary fiber, vitamins, and associated with its fibre content [7]. Analyses minerals, the peas are used to produce food ingredients The aim of the present study was hence to convert PpB, Experimental set-up for fungal cultivation The fungal spore concentration was measured using a such as proteins, starches, flours, and fibers [16–18]. Pea a cheap and low-nutritional value byproduct of the PPI The cultivations in 250 mL Erlenmeyer flasks were car- Bürker counting chamber. The total sugar, total solid, proteins have faced a growth in food applications due to production, into a vegan-mycoprotein concentrate for ried out using 100 mL of culture medium consisting only suspended solid and volatile solids of the samples were their nutritional and functional benefits, including their human food applications. Five strains of filamentous of the PpB substrate dissolved in distilled water. The opti- measured according to the National Renewable Energy balanced amino acid profile, positive fat- and water- fungi, namely A. oryzae, F. venenatum, M. purpureus, mum concentration of the medium was determined by Laboratory (NREL) methods [29, 30]. The total nitro- binding capabilities, emulsification and gelation proper- N. intermedia and R. oryzae, were screened for their testing the maximum load of PpB substrate which could gen content in the samples was determined by Kjeldahl ties, texture, and nutritional values. Moreover, allergies to growth to maximize the protein yield from the PpB. The go through the sterilization operation without causing method applying digestion, distillation, and acid–base pea are less frequent than allergies to other protein-rich best cases of the fungal growth were selected and further retrogradation. The term retrogradation refers to the titration using the InKjel P digestor and the Behrotest Souza Filho et al. Fungal Biol Biotechnol (2018) 5:5 Page 4 of 10 Souza Filho et al. Fungal Biol Biotechnol (2018) 5:5 Page 5 of 10

S1 distiller (Behr Labor-Technik, Germany). Protein was protein isolate process, was used as the substrate for the estimated by multiplying the nitrogen content by the cultivation of edible ascomycetes and zygomycetes fungi. 3 a b nitrogen-to-protein conversion factor of 6.25 [31]. HPLC These fungi are known for their palatability and high pro- (Waters 2695, Waters Corporation, Milford, U.S.A.) was tein content, with its potential application as human food used to analyze the components in all liquid fractions. components. The results on the characterization of the Acetic acid, ethanol, glucose, glycerol, lactic acid, and PpB and the fungal cultivation experiments together with 2 xylitol were analyzed using an analytical ion exchange the scale up studies in the airlift reactor are presented column based on hydrogen ions (Aminex HPX-87H, Bio- and discussed further. Rad, USA) operated at 60 °C with 0.6 mL/min of 5 mM Pea-processing byproduct (PpB) substrate characterization 1 H2SO4 as eluent. Arabinose, galactose, glucose, mannose, and xylose were analyzed using a lead (II)-based column The PpB was presented as a white odorless fine powder Concentration (g/L) (HPX-87P, Bio-Rad) with two Micro-Guard Deashing with the characterization as presented in Table 1. The (Bio-Rad) precolumns operated at 85 °C with 0.6 mL/ PpB substrate was characterized to have a total glucan 0 min ultrapure water as eluent. A UV absorbance detec- content of 62.38 ± 0.51% (w/w), more than 90% of this tor (Waters 2487), operating at 210 nm wavelength, being starch. Protein is the second most common com- 3 c d was used in series with a refractive index (RI) detector ponent in the material, amounting 18.19 ± 0.33% in dry (Waters 2414). Fructose and sucrose in the liquid samples weight basis. The C:N ratio of the PpB could be - deter and starch in the PpB media were measured using assay mined from the carbohydrate and protein contents as 2 kits of Megazyme (Ireland). The total chemical oxygen 10.28 ± 0.20. Retrogradation of the starch was observed demand (COD) was determined using NANOCOLOR® to determine the best concentration of the substrate PpB COD 15000 kit, with the photometric determination in the medium (distilled water) which would not cause of the samples using NANOCOLOR® photometers. the change in the quality of the liquid, while being auto- The viscosities of the samples were determined using a claved. The PpB concentrations (dry weight) of 1, 2, 3, 4, 1

Brookfield digital viscometer-model DV-E (Chemical and 5% (w/v) were tested. For the substrate loadings of 4 Concentration (g/L) Instruments AB, Sweden). and 5%, gelification of the medium was observed. At 3% substrate loading, the gelification was not as clear as in 0 Statistical analysis the previous cases although it was still noticed. For 1 and 0 6 12 18 24 30 36 Statistical analysis of the collected data was performed 2% substrate loadings, there was no observed retrograda- 3 e using the software MINITAB® (version 17.1.0). Analy- tion of starch. Fermentation time (h) ses of variance (ANOVA) of the data used general linear Additionally, the viscosity of the PpB suspension was models and a confidence level of 95% was used for all determined. The rheological properties of the cultiva- analysis. In the graphs, the average values are presented tion medium have effects on momentum, heat and mass 2 with an error bar representing one standard deviation. transfer, influencing the fermentation performance. In All the results presented in the tables are the average val- reactors, the flow properties of the media cause changes in the coalescence of the air bubbles, the bulk mixing, ues from duplicate experimental sets and are reported 1 with intervals representing the standard deviation. the process control, and the formation of stagnant zones [32]. The 2% PpB suspension was determined to have a Concentration (g/L) Results and discussion viscosity of 1.93 ± 0.15 cP before sterilization. After steri- With an abundance of food and a sedentary lifestyle in lization in the autoclave, however, the viscosity increased 0 many parts of the world there is a need of food that is 0 6 12 18 24 30 36 both nutritious and filling. Consumption of mycoprotein Table 1 Characterization of the pea-processing byproduct has been shown to improve health in relation to blood (PpB) Fermentation time (h) Fig. 1 Glucose concentration profile during the filamentous fungal cultivation in 2% (w/v) PpB substrate with no external enzyme supplementa- cholesterol concentrations, energy intake and glycemic Component Content (% w/w in dry basis) response. Furthermore, it can contribute to satiety and tion (filled triangles) and with the addition of 150 U/g of α-amylase (filled squares). The figure represents fungal strains M. purpureus (a); A. oryzae (b); Protein 18.19 0.33 F. venenatum (c); N. intermedia (d); and R. oryzae (e). Coordinates represent the mean values of duplicate tests; with error bars representing standard thus decreased energy intake in subsequent meals, some- ± deviations omitted due to negligible values Ash 2.98 0.03 thing that can increase weight-control [7, 23, 24]. There- ± Moisture 10.54 0.19 fore, mycoprotein can potentially contribute to both ± Arabinans 2.61 0.06 prevention and treatment of lifestyle dependent condi- ± tions such as obesity and type 2-diabetes; however more Xylans 0.00 to 10.47 0.12 cP, which means the viscosity increased coalescence and the decrease of the gas holdup. On the Galactans 2.30 0.04 ± research is needed to confirm this. The optimal dose of ± by 441%. The effect of the viscosity of the medium in the other hand, other researchers have found an increased Glucans 62.38 0.51 mycoprotein to boost health also remains unknown. In ± performance of bioreactors generally remains unclear. gas holdup in viscous media [33]. this study, pea-processing byproduct (PpB) with low Of which Some studies associate the increase in the viscosity with Starch 56.34 2.52 nutritional value, obtained as a byproduct of the pea ± the reduction of the liquid turbulence, promoting bubble Souza Filho et al. Fungal Biol Biotechnol (2018) 5:5 Page 6 of 10 Souza Filho et al. Fungal Biol Biotechnol (2018) 5:5 Page 7 of 10

PpB as substrate for efficient fungal growth have been cultivated in media with concentrations higher 16 2 The cultivation of four strains of ascomycetes (N. inter- than the ones obtained in this study [9, 34]. Biomass pro- a a media; A. oryzae; M. purpureus; and F. venenatum) and duction after 36 h is shown in Fig. 2. M. purpureus pro- one of the zygomycetes fungi (R. oryzae) was exam- duced significantly less biomass than the other strains (p 12 ined in a suspension containing 2% (w/v) PpB substrate value ≤ 0.031), except F. venenatum (p = 0.185). A. oryzae using 250 mL Erlenmeyer flasks. The tests were carried produced more biomass than R. oryzae (p = 0.074) and N. out with and without α-amylase addition and in dupli- intermedia (p = 0.038). The addition of α-amylase to the 8 1 cate samples. The effect of the cultivation temperature medium significantly affected only A. oryzae (p = 0.003) was not studied in this experimental set. Even without and M. purpureus (p = 0.038) growth. When comparing enzyme addition, the ascomycete strains showed a good the protein content of the harvested biomass, cultivation Concentration (g/L) capacity to hydrolyse starch to glucose (Fig. 1a). M. pur- of A. oryzae resulted in the highest yield of protein per Concentration (g/L) 4 pureus and R. oryzae consumption of glucose exceeded gram of PpB substrate, 0.14 g/g (Table 2). the sugar production rate after 24 h. On the other hand, From the previous results, three strains were being the sugar consumption rates of A. oryzae and N. inter- considered the most promising ones. A. oryzae, N. 0 0 media became higher than the sugar production rate intermedia, and R. oryzae were cultivated in the same 8 2 after 12 h. Consumption of glucose by F. venenatum medium as before (2% PpB substrate), with the addition b b was always lower than glucose production. On the other of the amyloglucosidase enzyme and the cellulase cock- hand, when α-amylase was added (Fig. 1b), M. purpureus tail to test if a higher biomass yield would be obtained. 6 consumed the glucose at a faster rate than its production As observed in Fig. 3, N. intermedia was most efficient after the first 12 h of cultivation. N. intermedia presented in consuming the glucose, with its concentration reach- the same behaviour. R. oryzae showed an even higher ing zero in about 18 h. Maximum ethanol concentration 4 1 glucose uptake, overtaking the glucose production rate reaching 5.98 g/L, was also observed with N. intermedia. after the first 6 h of cultivation. Glucose profile during A. The fungal protein yield obtained for A. oryzae and R. Concentration (g/L) oryzae and F. venenatum cultivation oscillated between oryzae was 0.09 g/g of PpB substrate while for N. inter- Concentration (g/L) 2 increasing and decreasing the glucose concentration in media it was 0.10 g/g. the medium during the cultivation, likely because of dif- ferent glucose production and glucose uptake rates. Cultivation of A. oryzae in the airlift bioreactor 0 0 N. intermedia produced the most ethanol among the A. oryzae growth was examined in a bench-scale air- 0 6 12 18 24 30 0 6 12 18 24 30 36 42 48 microorganisms (4.30 g/L with enzyme addition) while lift bioreactor. Since A. oryzae is known as an amylase Fermentation time (h) M. purpureus cultivation resulted in the lowest final etha- producer [35], for the further experiments the medium Time (h) Fig. 3 Glucose (a) and ethanol (b) concentration profiles during the Fig. 4 Sugar and ethanol concentration profiles during the cultiva- nol concentration (0.28 g/L without enzyme addition). was supplemented only with the cellulase cocktail. Four tion of A. oryzae in a 2% and b 3% (w/v) PpB substrate in an airlift bio- R. oryzae also converted glucose into lactic acid, yielding cultivations were run in the bioreactor to test two con- filamentous fungal cultivation in 2% (w/v) PpB substrate. Figure rep- resents fungal strains A. oryzae (filled squares); N. intermedia (asterisk); reactor. Figure represents glucose (filled squares); other sugars (filled 0.66 g/L. α-amylase addition to the media did not result in centrations of PpB substrate in duplicates: 2 and 3%, and R. oryzae (filled circles) triangles); and ethanol (filled circles). Other sugars are presented in a significant change in the ethanol and lactic acid produc- resulting in media with pH 6.1 and 6.5 respectively. The xylose equivalent tion for the ascomycetes strains. However, R. oryzae had results presented in Fig. 4 shows that for both tested its ethanol and lactic acid production increased by the scenarios glucose concentration increased up to 36 h. addition of the enzyme. Inhibitory effect of these com- Additionally, 44 h were not enough for the total con- known for their capacity to operate using higher aera- pounds were not considered since the microorganisms Table 2 Protein yield from the fungal biomass obtained after 36 h cultivation in 2% pea-processing byproduct sumption of the sugar. When the pea processing load tion rates when compared to traditional stirred fer- (PpB) substrate was 2%, the other sugars were consumed and at 36 h menters [36]. Moreover, efficient oxygen transfer and mixing are obtained when using airlifts [37]. As a result, Enzyme addi- Microorganism % Protein in dry Protein yield their concentration was below 0.10 g/L. On the other 300 tion fungal biomass (g/g PpB hand, when the initial load of PpB substrate was 3%, the cultivation in airlift bioreactor yielded low ethanol con- without α-amylase substrate) concentration of the sugars (except glucose) reached the centrations and high fungal protein. with α-amylase 250 maximum value at 36 h. The ethanol profiles were also Without M. purpureus 53.61 0.02 Process integration at the existing pea-processing 200 α-amylase A. oryzae 43.13 0.11 divergent, with maximum concentration being reached at 36 h (0.40 g/L) for 2% PpB substrate load and at 44 h industrial facilities 150 F. venenatum 55.28 0.09 The use of co-products from the industries as impor- N. intermedia 54.53 0.11 (1.01 g/L) for 3% PpB substrate load. The obtained pro- 100 tant sources of protein for human consumption has been R. oryzae 50.03 0.09 tein yields were 0.26 and 0.13 g/g of PpB substrate for Biomass yield (mg/g) receiving considerable attention. A method of protein 50 With α-amylase M. purpureus 58.66 0.03 2 and 3% of the substrate, respectively. The yield for production, applicable to a wide range of industrial pro- A. oryzae 46.36 0.14 2% PpB medium is almost twice the value obtained in 0 the shake flask experiments. Moreover, working with cesses, is to use edible filamentous fungi which could be M. purpureus A. oryzae F. venenatum N. intermedia R. oryzae F. venenatum 59.75 0.11 a low load of the substrate (and low viscosity) resulted used as a protein rich meat substitute. Filamentous fungi Fig. 2 Biomass yield; mg dry fungal biomass per gram of PpB sub- N. intermedia 54.11 0.11 in increased protein production. Airlift bioreactors are have been reported as important industrial microorgan- strate after 36 h of cultivation in 2% w/v PpB medium R. oryzae 54.79 0.11 isms for their capacity to produce numerous metabolites Souza Filho et al. Fungal Biol Biotechnol (2018) 5:5 Page 8 of 10 Souza Filho et al. Fungal Biol Biotechnol (2018) 5:5 Page 9 of 10

Author details 13. Archer DB. Filamentous fungi as microbial cell factories for food use. Curr 1 Swedish Centre for Resource Recovery, University of Borås, 50190 Borås, Opin Biotechnol. 2000;11(5):478–83. Sweden. 2 Mycorena AB, Stena Center 1A, 411 92 Gothenburg, Sweden. 14. Wiebe MG. Quorn™ myco-protein—overview of a successful fungal 3 Faculty of Caring Science, Work Life and Social Welfare, University of Borås, product. Mycologist. 2004;18(1):17–20. 50190 Borås, Sweden. 15. Anupama RP. Value-added food: single cell protein. Biotechnol Adv. 2000;18(6):459–79. Acknowledgements 16. Hoffmann B, Münch S, Schwägele F, Neusüß C, Jira W. A sensitive HPLC- Not applicable. MS/MS screening method for the simultaneous detection of lupine, pea, and soy proteins in meat products. Food Control. 2017;71:200–9. Competing interests 17. Stone AK, Avarmenko NA, Warkentin TD, Nickerson MT. Functional prop- The authors declare that they have no competing interests. erties of protein isolates from different pea cultivars. Food Sci Biotechnol. 2015;24(3):827–33. Availability of data and materials 18. Tulbek MC, Lam RSH, Wang Y, Asavajaru P, Lam A. Pea: a sustainable Not applicable. vegetable protein crop. In: Nadathur SR, Wanasundara JPD, Scalin L, editors. Sustainable protein sources. San Diego: Academic Press; 2017. Consent for publication p. 145–64. Not applicable. 19. Day L. Proteins from land plants—potential resources for human nutri- tion and food security. Trends Food Sci Technol. 2013;32(1):25–42. Ethics approval and consent to participate 20. McCarthy NA, Kennedy D, Hogan SA, Kelly PM, Thapa K, Murphy KM, Not applicable. Fenelon MA. Emulsification properties of pea protein isolate using homogenization, microfluidization and ultrasonication. Food Res Int. Funding 2016;89(1):415–21. This work was financially supported by the Swedish Research Council 21. Ratnayake WS, Hoover R, Warkentin T. Pea starch: composition, structure FORMAS, and the Coordination for the Improvement of Higher Education and properties—a review. Starch-Stärke. 2002;54(6):217–34. Personnel, CAPES (Brazil). 22. Ruxton CHS, McMillan B. The impact of mycoprotein on blood cholesterol levels: a pilot study. Brit Food J. 2010;112(10):1092–101. 23. Williamson DA, Geiselman PJ, Lovejoy J, Greenway F, Volaufova J, Martin Publisher’s Note CK, Arnett C, Ortego L. Effects of consuming mycoprotein, tofu or Springer Nature remains neutral with regard to jurisdictional claims in pub- chicken upon subsequent eating behaviour, hunger and safety. Appetite. lished maps and institutional affiliations. 2006;46(1):41–8. Fig. 5 Block flow diagram for the production of pea-protein isolate integrated with the production of vegan-mycoprotein concentrate by filamen- 24. Bottin JH, Swann JR, Cropp E, Chambers E, Ford HE, Ghatei MA, Frost GS. Received: 30 January 2018 Accepted: 6 March 2018 tous fungi. Adapted from [39] Mycoprotein reduces energy intake and postprandial insulin release without altering glucagon-like peptide-1 and peptide tyrosine-tyrosine concentrations in healthy overweight and obese adults: a randomised- controlled trial. Brit J Nutr. 2016;116(2):360–74. with high market potential [38]. Among these metabo- The scenario also includes a system in which the broth 25. Paddon-Jones D, Westman E, Mattes RD, Wolfe RR, Astrup A, Westerterp- Plantenga M. Protein, weight management, and satiety. Am J Clin Nutr. References lites, the vast assortment of enzymes produced by these containing ethanol could be either sent to a distillation 2008;87(5):1558S–61S. 1. Boland MJ, Rae AN, Vereijken JM, Meuwissen MPM, Fischer ARH, van microorganisms has special importance. These catalysts unit or to any adjacent ethanol facilities in the vicinity 26. Chung H-J, Lim HS, Lim S-T. Effect of partial gelatinization and retro- Boekel MAJS, Rutherfurd SM, Gruppen H, Moughan PJ, Hendriks WH. The gradation on the enzymatic digestion of waxy rice starch. J Cereal Sci. are responsible for making filamentous fungi a flex- which contain all the necessary installation for ethanol future supply of animal-derived protein for human consumption. Trends 2006;43(3):353–9. Food Sci Technol. 2013;29(1):62–73. ible group able to use a large range of biomaterials as purification [34]. However, the techno-economic feasibil- 27. Fredriksson H, Björck I, Andersson R, Liljeberg H, Silverio J, Eliasson AC, 2. Godfray HCJ, Beddington JR, Crute IR, Haddad L, Lawrence D, Muir JF, substrate. ity of using PpB as a potential substrate for edible fungal Åman P. Studies on α-amylase degradation of retrograded starch gels Pretty J, Robinson S, Thomas SM, Toulmin C. Food security: the challenge from waxy maize and high-amylopectin potato. Carbohydr Polym. The scheme of an industrial process for the produc- cultivation for feed or food component needs to be veri- of feeding 9 billion people. Science. 2010;327(5967):812–8. 2000;43(1):81–7. 3. Hartmann C, Siegrist M. Consumer perception and behaviour regarding tion of PPI is presented in Fig. 5 [39] with the proposal fied in detail and hence open for future studies. 28. Frei M, Siddhuraju P, Becker K. Studies on the in vitro starch digestibility sustainable protein consumption: a systematic review. Trends Food Sci of integrating the production of fungal biomass for and the glycemic index of six different indigenous rice cultivars from the Technol. 2017;61:11–25. Conclusion Philippines. Food Chem. 2003;83(3):395–402. the valorization of the byproduct. The starch and fib- 4. Tukker A, Jansen B. Environmental impacts of products: a detailed review 29. Sluiter A, Hames B, Ruiz R, Scarlata C, Sluiter J, Templeton D, Crocker D. of studies. J Ind Ecol. 2006;10(3):159–82. ers separated from the pea protein would be fed in a The pea-processing byproduct was proved to be an effi- Determination of structural carbohydrates and lignin in biomass. Golden, 5. Sadler MJ. Meat alternatives—market developments and health benefits. bioreactor. For dilution of the solid material, the low- cient medium for the growth of filamentous fungi to pro- CO: National Renewable Energy Laboratory; 2012. Trends Food Sci Technol. 2004;15(5):250–60. 30. Sluiter A, Ruiz R, Scarlata C, Sluiter J, Templeton D. Determination of protein content stream from the ultrafiltration step duce a vegan-protein concentrate. Fungal biomass with 6. Moore D, Chiu SW. Fungal products as food. In: Pointing SB, Hyde KD, edi- sugars, byproducts, and degradation products in liquid fraction process tors. Bio-exploitation of filamentous fungi. Hong Kong: Fungal Diversity could be used. This stream is sterile and rich in oli - about 46 and 54% protein content was obtained from PpB samples. Golden, CO: National Renewable Energy Laboratory; 2008. Press; 2001. p. 223–51. gosaccharides, which can also be used by the fungi as using edible strains of filamentous fungi, A. oryzae and 31. Ferreira JA, Lennartsson PR, Taherzadeh MJ. Production of ethanol and 7. Denny A, Aisbitt B, Lunn J. Mycoprotein and health. Nutr Bull. biomass from thin stillage by Neurospora intermedia: a pilot study for substrate. Following the same yields of the bench-scale N. intermedia respectively. Scaling-up of the process to 2008;33(4):298–310. process diversification. Eng Life Sci. 2015;15(8):751–9. 8. Smith H, Doyle S, Murphy R. Filamentous fungi as a source of natural airlift bioreactor, 1 ton of pea-processing byproduct a 4.5 L bench scale airlift bioreactor improved the A. ory- 32. Kemblowski Z, Kristiansen B. Rheometry of fermentation liquids. Biotech- 3 antioxidants. Food Chem. 2015;185:389–97. would consume 50 m of water, and would be predicted zae biomass yield, with a total of 0.26 g of fungal protein nol Biooeng. 1986;28(10):1474–83. 9. Ferreira JA, Lennartsson PR, Niklasson C, Lundin M, Edebo L, Taherzadeh 33. Wu Q, Wang X, Wang T, Han M, Sha Z, Wang J. Effect of liquid viscosity on to produce 680 kg of A. oryzae biomass with 260 kg of per g of PpB. Based on the results obtained, an industrial MJ. Spent sulphite liquor for cultivation of an edible Rhizopus sp. BioRe- hydrodynamics and bubble behaviour of an external-loop airlift reactor. sources. 2012;7(1):173–88. pure protein under ideal conditions. The biomass, how- scenario for the production of vegan-mycoprotein con- Can J Chem Eng. 2013;91(12):1957–63. 10. Chen M-H, Johns MR. Effect of pH and nitrogen source on pigment pro- ever, would need to be subjected to a heat treatment to centrate for human nutrition is proposed as an integrated 34. Mahboubi A, Ferreira JA, Taherzadeh MJ, Lennartsson PR. Value- duction by Monascus purpureus. Appl Microbiol Biot. 1993;40(1):132–8. added products from dairy waste using edible fungi. Waste Manag. reduce its RNA content, resulting in some loss of pro - process to the existing PPI production facilities. 11. Huang Z, Xu Y, Li L, Li Y. Two new Monascus metabolites with strong 2017;59:518–25. blue fluorescence isolated from red yeast rice. J Agr Food Chem. tein [14]. Moreover, tests for mycotoxin production Authors’ contributions 35. Nair RB, Taherzadeh MJ. Valorization of sugar-to-ethanol process waste 2007;56(1):112–8. should be carried out frequently and, in case of detec- RBN, PRL and MJT conceived the project. RBN performed the experimental vinasse: a novel biorefinery approach using edible ascomycetes filamen- 12. Tsukahara M, Shinzato N, Tamaki Y, Namihira T, Matsui T. Red yeast rice and laboratory work. PFSF, RBN, PRL and MJT analyzed and interpreted the tous fungi. Bioresour Technol. 2016;221:469–76. tion, application of corrective methods can affect the fermentation by selected Monascus sp. with deep-red color, lovastatin data regarding fermentation. DA contributed with the analysis of the data 36. Chisti Y, Jauregui-Haza UJ. Oxygen transfer and mixing in mechanically production but no citrinin, and effect of temperature-shift cultivation on productivity [14]. regarding the nutritional aspects. PFSF, RBN and DA wrote the paper. All agitated airlift bioreactors. Biochem Eng J. 2002;10(2):143–53. lovastatin production. Appl Biochem Biotechnol. 2009;158(2):476–82. authors read and approved the final manuscript. Souza Filho et al. Fungal Biol Biotechnol (2018) 5:5 Page 10 of 10

37. Moo-Young M, Halard B, Allen DG, Burrell R, Kawase Y. Oxygen transfer to 39. Fredrikson M, Biot P, Alminger ML, Carlsson N-G, Sandberg A-S. Produc- mycelial fermentation broths in an airlift fermentor. Biotechnol Bioeng. tion process for high-quality pea-protein isolate with low content of 1987;30(6):746–53. oligosaccharides and phytate. J Agric Food Chem. 2001;49(3):1208–12. 38. Ferreira JA, Lennartsson PR, Edebo L, Taherzadeh MJ. Zygomycetes-based biorefinery: present status and future prospects. Bioresource Technol. 2013;135:523–32.

Paper 5

Submit your next manuscript to BioMed Central and we will help you at every step: • We accept pre-submission inquiries • Our selector tool helps you to find the most relevant journal • We provide round the clock customer support • Convenient online submission • Thorough peer review • Inclusion in PubMed and all major indexing services • Maximum visibility for your research

Submit your manuscript at www.biomedcentral.com/submit

A solvent-free approach for production of films from pectin and

fungal biomass

Rajesh Gurram, Pedro F. Souza Filho*, Mohammad J. Taherzadeh, Akram Zamani

Swedish Centre for Resource Recovery, University of Borås, SE 50190 Borås, Sweden

*Corresponding author

Tel: +46 33 4354487

Fax: +46 33 4354003

E-mail: [email protected]

1

1 Abstract 21 1. Introduction

2 Self-binding ability of the pectin molecules was used to produce pectin films using the 22 Plastics, which are commonly derived from petroleum resources, harmfully affect the wild

3 compression molding technique, as an alternative method to the high energy-demanding and 23 and human lives [1]. Accordingly, bioplastics have been suggested as one of the best

4 solvent-using casting technique. Moreover, incorporation of fungal biomass and its effects on 24 alternatives for the conventional plastics. Along with plastics, residues of different industries

5 the properties of the films was studied. Pectin powder plasticized with 30% glycerol was 25 are sometimes challenging and harmful to the environment. Citrus waste (CW), originated

6 subjected to heat compression molding (120 °C, 1.33 MPa, 10 min) yielding pectin films with 26 from citrus fruit, is an example of such residues [2, 3]. The majority of the CW is produced by

7 tensile strength and elongation at break of 15.7 MPa and 5.5%, respectively. The filamentous 27 the citrus fruit processing industry where half of the weight of the citrus fruit is converted into

8 fungus Rhizopus oryzae was cultivated using the water-soluble nutrients obtained from citrus 28 waste [3]. Low pH and high moisture content of CW prevent its landfilling according to the

9 waste and yielded a biomass containing 31% proteins and 20% lipids. Comparatively, the 29 EU regulations [4], and the presence of essential oils is prejudicial to the composting and the

10 same strain was cultivated in a semi-synthetic medium resulting in a biomass with higher 30 biogas processes [2, 3]. Moreover, thermal treatments of CW yield low energy recovery due

11 protein (60%) and lower lipid content (10%). SEM images showed addition of biomass 31 to the high moisture content [2, 3]. To overcome these difficulties, biological treatments of

12 yielded films with less debris compared to the pectin films. Incorporation of the low protein 32 CW and production of value added products, e.g. by fermentation, has been proposed.

13 content biomass up to 15% did not significantly reduce the mechanical strength of the pectin 33 CW contains abundant amount of sugars, including poly-, di- and monosaccharides. Most

14 films. In contrast, addition of protein-rich biomass (up to 20%) enhanced the tensile strength 34 polysaccharides (cellulose, hemicellulose, and pectin) are components of the peel of the CW,

15 of the films (16.1-19.3 MPa). Lastly, the fungal biomass reduced the water vapor permeability 35 while the di- and monosaccharides (sucrose, glucose, and fructose) are present in the pulp.

16 of the pectin films. 36 Satari et al [5] reported fungal cultivation using a solution containing the free sugars extracted

17 37 from CW, without the addition of any nutrients, in a bench scale airlift bioreactor. Hydrolysis

18 38 of the structural polymers cellulose, hemicelluloses and pectin of CW, and ethanol production

19 Keywords: Citrus waste; pectin; compression molding; Rhizopus oryzae; bioplastics. 39 have also been reported [5-7].

20 40 Besides sugars, CW is also a major source for industrial production of pectin, the other

41 abundant compound in the citrus peel. Pectin is the term used to describe a group of

42 heteropolysaccharides naturally found in the cell wall of vascular plants which act as a

43 cementing matrix in the cellulosic fibers [8]. The natural role of pectin in providing

44 mechanical strength to the plant cell wall has encouraged the use of pectin as matrix in

45 biocomposite materials. Pectin has a good potential to be used in food packaging films and

3

46 biocomposite materials, which are renewable, biodegradable and biocompatible [9]. Pectin- 70 biomass is a rich source of different biopolymers. Improvement of the properties of the pectin

47 based biocomposite films are usually produced by solution casting [10] in which pectin is 71 films by blending the polymer with fungal biomass has not been investigated before.

48 dissolved in an aqueous acidic solution, and reinforcing materials are suspended in this 72 The purpose of the current work was to apply a solvent-free compression molding method for

49 solution. The solution is then casted over a smooth surface and dried to make a film [11]. 73 production of bioplastic films from citrus peel derived pectin. Moreover, as another solution

50 Production of bioplastic films via solution casting method however has a high demand of 74 to citrus waste (CW) challenges, free sugars and other water-soluble nutrients were extracted

51 energy which may avoid profitability of the process in large scales [12]. The molding method, 75 from CW and employed for cultivation of the filamentous fungus Rhizopus oryzae, in a pilot

52 on the other hand, usually has lower energy demand and processing time compared to the 76 scale bubble column reactor. Incorporation of the fungal biomass to the pectin films was also

53 casting method, thus being preferred for industrial applications [13]. For non-thermoplastic 77 investigated. Fungal biomass grown on a rich synthetic medium was used as a reference for

54 biopolymers, such as proteins, heat compression molding has been applied to produce 78 comparison of the properties of the obtained films. SEM analyses were used to study the

55 bioplastic items [14]. However, to the best of our knowledge, this technique has not been 79 molecular interactions between the pectin and the fungal biomass components.

56 employed for production of pectin based bioplastics. 80

57 Blending of pectin with clay nanoparticles such as halloysite nanotubes [15, 16] and layered 81 2. Material and Methods

58 double hydroxides materials [17] positively affect the mechanical performances, improved 82 2.1.Materials

59 thermal stability, and vapor barrier properties of the pectin films. Similarly, the blend of other 83 Citrus waste (CW) was kindly provided by Brämhults Juice AB (Borås, Sweden). For the

60 biopolymers such as polysaccharides (e.g. starch, cellulose, and chitosan) [11], proteins, and 84 production of pectin-based bioplastic films, citrus peel derived pectin (poly-D-galacturonic

61 lipids [18] were reported to improve the pectin films characteristics. For instance, in a review, 85 acid, ≥ 74%) from Sigma-Aldrich Inc. (St. Louis, MO, USA) was used. Anhydrous glycerol

62 Porta et. al [19] stated that addition of proteins, e.g. soy proteins, to pectin films not only 86 (analytical grade, 99%), from Scharlab S.L. (Barcelona, Spain), was used as plasticizer for the

63 brings a nutritional value to the films, but also increases the strength and improves the oxygen 87 production of bioplastic films. Other chemicals and reagents used in this study were

64 barrier characteristics. The improvements are caused by the strong interactions between the 88 purchased from Sigma-Aldrich Inc. (St. Louis, MO, USA) unless stated otherwise.

65 OH- and COO- groups of pectin and positively charged –NH groups of proteins. Lipids can 89

66 form dipole-charge and dipole-dipole interactions with polar functional groups of pectin 90 2.2. Microorganism

67 matrix and therefore, improve the characteristics of the pectin-based films [12, 18]. 91 Rhizopus oryzae (CCUG28958) used for biomass cultivation was acquired from the Culture

68 Furthermore, materials containing a mixture of several organic compounds, e.g. coffee 92 Collection of the University of Gothenburg, Sweden. The strain was grown in agar plates

69 ground, have been tested as filler for a pectin matrix with interesting results [20]. Fungal 93 (potato extract 4 g/L, dextrose 20 g/L, agar 15 g/L) for 5 days at 30 °C. The obtained plates

94 were then kept at 5 °C until use for a maximum of 30 days. Spore suspension for inoculation

4 5

95 was prepared by adding 20 mL of sterile water in a plate and gently stirring the liquid using a 120 cultivation, which was carried out at 35 °C and at an aeration rate of 1.5 vvm (volume of air

96 sterile L-shaped spreader. 121 per volume of medium per minute). The pH was adjusted to 5.4-5.5 at the beginning of the

97 122 cultivation. The initial pH of CWFS was 3.6 and was increased to 5.5 using 5 M NaOH

98 2.3. Extraction of citrus waste free sugars (CWFS) and nutrients 123 solution [5]. At the end of the cultivations, the broth was filtered through a kitchen sieve and

99 The soluble sugars and other soluble nutrients were extracted from CW according to Satari et 124 the harvested solid was washed with tap water and stored at -20 °C for further usage. As the

100 al. [5] with minor modification. Briefly, CW in wet form (60 kg) was mixed with tap water 125 media at the beginning of the cultivations contained no suspended solids, the collected solids

101 (90 L) and left overnight. Thereafter, the mixture was subjected to liquid extraction in a fruit 126 were considered to be only fungal biomass.

102 press machine (LANCMANTM VSPIX120) equipped with a textile filter (pore size 3 mm). 127 The broth was analyzed in terms of lactic acid, glycerol, ethanol, glucose, and other sugars

103 The citrus waste free sugars solution (CWFS) was collected and stored at 5 °C until use. 128 using a hydrogen-ion based ion-exchange column (Aminex HPX-87H, Bio-Rad, Hercules,

104 129 CA, USA) installed in a high-performance liquid chromatography (HPLC) system at 60 °C,

105 2.4. Cultivation in bubble column reactor 130 using 0.6 mL/min of 0.5 mM H2SO4 solution as the eluent. A refractive index detector

106 Cultivation of R. oryzae was performed using different media, namely CWFS and a nutrient 131 (Waters 2414, USA) was used to identify and quantify the components.

107 rich semi-synthetic medium [glucose (10 g/L), yeast extract (5.0 g/L), K2HPO4 (3.5 g/L), 132

108 CaCl2.2H2O (1.0 g/L), MgSO4.7H2O (0.75 g/L), and (NH4)2SO4 (7.5 g/L)]. Sterilization of the 133 2.5. Characterization of the fungal biomass

109 liquid media was performed in an autoclave (Systec, Germany) at 121 °C for 20 min. R. 134 The standard Kjeldhal method (InKjel P digestor + Behrotest S1 distiller, Behr Labor-

110 oryzae inoculum was prepared in 1-L baffled Erlenmeyer flasks containing 300 mL of sterile 135 Technik, Germany) was adopted for analysis of protein content in the fungal biomass. For this

111 CWFS medium or semi-synthetic medium at pH 5.5. These solutions were inoculated with 15 136 analysis, the biomass was pre-dried at 70 °C overnight. Briefly, the protein content in the

112 ml of spore suspension of R. oryzae and incubated in shaking water bath at 35 °C (for 24 h for 137 biomass was quantified in two steps, namely digestion (where the organic nitrogen content of

113 semi-synthetic medium or 48-55 h for CWFS medium) to obtain enough fungal biomass for 138 biomass was converted into ammonium sulphate and water) and distillation (where

114 inoculation of 26 L bubble column bioreactor (Bioengineering, Switzerland). 139 ammonium hydroxide was collected as distillate through the reaction of (NH4)2SO4 with 10.7

115 Inoculum (1.5% v/v) was added to a 20-L plastic container containing the sterile medium 140 M NaOH). The collected distillate was titrated with 0.1 M HCl to determine the protein

116 (CWFS or semi-synthetic) to complete 20 L of working volume. In addition, 5 ml of antifoam 141 concentration in fungal biomass (defined as 6.25 times the net nitrogen content of biomass)

117 204 (Sigma-Aldrich Inc, MO, USA) was also added to the medium to prevent foam 142 [21]. Sulfuric acid (Sigma-Aldrich Inc., MO, USA), Kjeldhal tablets (0.5 g CuSO4 + 5.0 g

118 generation during the cultivation. This inoculated solution was manually transferred to the 143 K2SO4; Thompsons and Capper LTD, UK), and antifoam tablets (0.97 g Na2SO4 + 0.03 g

119 sterile bioreactor (sterilized by direct steam injection at 130 °C for 30 min) to start the

6 7

144 silicone antifoam; Thompsons and Capper LTD, UK) were used as reagents for biomass 169 To prepare the films, the obtained pectin-glycerol blend was placed between two square-

145 digestion. 170 shaped high-density polyethylene sheets (12 cm × 12 cm) and placed in a Rondol 20 Ton

146 Organic solvent extraction was used for lipid determination according to Majdejabbari et. al 171 molding press (Rondol Technologies Ltd, UK). The compression molding process was

147 [22] with modification. The harvested biomass was lyophilized (Labconco: Kansas City, MO, 172 performed under operation conditions of 1.33 MPa and 120 ºC. The heating plates were

148 USA) and milled to 0.2-mm size powder using a Rotor mill (Fritsch Pulverisette14, Fritsch 173 previously set at the working temperature and the sample was kept between them for 10 min.

149 Industries, Germany). Extraction was carried out by adding 25 ml of 0.4% (w/v) biomass 174 The sheets were then removed from the press and left at ambient temperature to cool for 5

150 suspension in water to 50 ml of organic solvent mix (containing 40% petroleum ether, 40% 175 min. The obtained pectin biofilms were stored in PE bags at room temperature for further

151 diethyl ether, and 20% absolute ethanol) in a 250 ml separating funnel. The content in the 176 characterization.

152 funnel was thoroughly mixed manually for 10 min and left still for phase separation. The 177 Pectin-based biomass films were also produced by incorporation of fungal biomass. The

153 organic phase was collected in pre-weighted glass beakers to determine the lipid 178 fungal biomass was lyophilized and milled (particle size < 0.2-mm) before mixing with pectin

154 concentration. The aqueous phase was subjected to two more extractions similar to the first 179 and glycerol. Biomass concentrations varied in the range of 0-35% of the total mixture.

155 one to guarantee the complete extraction of lipids. The beaker containing the organic phase of 180 Glycerol content was kept at 30%. Pectin, fungal biomass, and glycerol were vigorously

156 the three extractions was left overnight in a fume hood for evaporation of the liquid and 181 blended manually. A longer mixing step (5 min compare to 2 min for pectin films) was

157 further dried at 105 °C until constant weight. By weight difference, the amount of lipids was 182 necessary to get a uniform dough material that was suitable for compression molding.

158 determined. 183 Afterwards, the matrix was shaped into a ball (total weight of 2.5 g) and conditioned

159 184 overnight at ambient temperature in PE bags to get better uniformity. Then, blends were

160 2.6. Solvent free method for production of pectin films 185 subjected to compression molding to produce films. Figure 1 illustrates different steps in

161 The pectin films were produced using a solvent-free compression molding approach, which 186 preparation of the pectin based films by compression molding approach.

162 involved thermo-mechanical treatment of pectin. Pectin (Sigma-Aldrich Inc., St. Louis, MO, 187

163 USA) was used with galacturonic acid ≥ 74.0% - dried basis, methoxy groups ≥ 6.7% - dried 188 2.7. Tensile analyses

164 basis, and molecular weight of 30-100 kg/mol. Glycerol was used as plasticizer. Preliminary 189 Tensile analyses, namely the yield point, the tensile strength (TS), the elongation at break (E

165 experiments (data not shown) indicated glycerol content (GC) of 30% (w/w) as the best 190 %), and the Young’s modulus were performed to determine the mechanical properties of the

166 concentration. Mixing of pectin powder with glycerol was carried out manually for 2 min 191 films. These properties were measured using an Elastocon H10KT tensile testing machine

167 with the help of a glass stirrer to get an uniform dough which was then formed into ball shape 192 (Elastocon AB, Sweden). The films were cut in dog-bone shaped testing specimens using a

168 (total weight of 2.5 g) and stored in polyethylene (PE) bags overnight at room temperature. 193 manual cutting press (EP 08, Elastocon AB, Sweden) equipped with a dog bone shape cutting

8 9

194 die EP 04 ISO 32-7. The pectin films were subjected to the tensile examination with gauge 219 samples were attached to a carbon tape and covered with gold. For cross-section imaging,

195 length of 22 mm, load force of 100 N, gap between grips of 50 mm, and width of dog bone 220 samples were immersed in liquid nitrogen for 1 min, broken, and attached to a carbon tape

196 shape film specimen of 4 mm. QMat 5.41a-Dongle 4631 software processor was used to 221 placed on a stub. Then, the gold coating was performed before imaging. The images were

197 process the data. All film specimens were tested in triplicates. 222 collected at magnification of 1000×, energy of the beam 10 kV, and working distance of 6

198 223 mm for surface images, and at magnification of 1500×, energy of the beam 25 kV, and

199 2.8. Water vapor permeability coefficient (WVPC) 224 working distance of 10 mm for cross-section images.

200 Water vapor permeability coefficient (WVPC) was determined according to the ASTM E96 225

201 [23]. The pectin and pectin-biomass films were placed on the top of a pre-dried (at 70 ºC, 226 2.10. Statistical analyses

202 overnight) glass container (diameter of 32-50 mm depending on the film diameter) and sealed 227 All experiments were carried out in duplicate. Data are reported as average ± standard

203 with a paraffin film. Prior to the film insertion, glass containers were filled with 50-100 g of 228 deviation. Error bars in graphs indicate one standard deviation. Results were analysed using

204 distilled water, providing a distance between the tested film and the water surface of 2-3 cm. 229 the software MINITAB® 17 Statistical Software (Minitab Inc., State College, PA, USA).

205 The weight of the entire setup, including the film, was measured (at time 0 h) and this was 230 Tukey tests were performed to determine statistical differences between results. A confidence

206 kept in a desiccator. The weight loss of the containers was measured every 24 h up to 5 days, 231 interval of 95% was considered in all analyses.

207 and finally a graph of weight loss versus time was plotted which showed a linear behavior. 232

208 The slope of the straight line was used for the WVPC calculation (in kg.s-1.m-1.Pa-1) according 233 3. Results and discussions

209 to Equation 1. 234 Pectin-based bioplastic films have the potential to reduce the environmental footprint of

210 (1) 235 human activity twofold: decreasing the negative effects of synthetic plastics and addressing ௌή௑ ஺ή୼௉ 236 the citrus waste management issue [26]. However, the solution casting method, the most- 211 Where S is the slope of the plot ܹܸܲܥ(kg/s), A ൌ is the beaker mouth area (m2), ΔP is the vapor 237 commonly used method for pectin film production, has a high demand of energy. In this work 212 pressure difference (Pa) between the two sides (inside of glass container and desiccator) of the 238 heat compression molding technique, with lower energy demand compared to casting [27, 213 film and X is the film thickness (m) [24, 25]. The test was done in duplicates for each film. 239 28], was employed for production of pectin films. Furthermore, the potential of the fungal 214 240 biomass for enhancement of the pectin films was investigated. The biomass was obtained by 215 2.9. Scanning Electron Microscopy (SEM) 241 cultivation in water soluble nutrients of citrus waste [5]. Fungal biomass cultivated on a rich 216 Scanning electron microscopy (SIGMA VP FE-SEM, Carl Zeiss AG, Germany) analysis was 242 semi-synthetic medium was also used as a reference. 217 conducted to study the morphology of the pectin based films. The micrographs were taken 243 218 from surface and cross-section of the films. Before taking images from the surface, the

10 11

244 3.1.Self-binding ability of pectin and production of pectin films 269 3.2. Production of fungal biomass for incorporation into pectin films

245 Thermo-triggered self-binding ability of proteins [28] and natural fibers [27] to produce 270 Citrus waste free sugars (CWFS) solution and a semi-synthetic medium were used for fungal

246 binder-less objects has been reported. Under the thermo-mechanical treatments, proteins 271 cultivation in a 26-L bubble column bioreactor. The fungal biomass was harvested when the

247 undergo disaggregation, denaturation, and dissociation reactions which can lead to the 272 sugar concentration in the medium was close to zero. In CWFS medium, R. oryzae first

248 formation of new links and the aggregation of proteins to new forms [28]. Melting/glass 273 consumed glucose. When the glucose level in the medium was low, consumption of other

249 transition as well as degradation reactions play roles in formation of new objects from natural 274 sugars, e.g. fructose and sucrose, was started. The profile of different components during the

250 fibers under thermo-mechanical treatments [27]. Plasticizers are often required to improve the 275 fungal fermentation is presented in Figure 3. In average, 65-72 h was necessary for R. oryzae

251 processability and the mechanical properties by reducing the interactions between the 276 to completely assimilate the glucose (≈ 20 g/L) and ca. 10 g/L of other sugars present in the

252 polymers chains. Glycerol has been used to improve the properties of the protein bioplastics 277 CWFS medium. Starting with 10 g/L glucose in semi-synthetic medium, the fungus needed

253 [28, 29]. The presence of water, moisture, has been reported to be necessary for production of 278 20 h for complete assimilation of the sugar. Therefore, the fungal exhibited comparable rates

254 cellulose bioplastics with good mechanical properties [27]. 279 of glucose assimilation in these media. However, different profiles of fermentation

255 In this work the self-binding ability of pectin was investigated. Glycerol was used as a 280 byproducts were observed in these two media (Figure 3).

256 plasticizer. Pectin powder was mixed with glycerol to form a dough-like material. This 281 The concentration of lactic acid and ethanol remained low during fungal cultivation in the 257 material was then subjected to a heat compression molding process in similar conditions used 282 CWFS medium. In the semi-synthetic medium, the obtained values for lactic acid and ethanol 258 for production of proteins bioplastics, namely, 120 °C and 1.33 MPa for 10 min. Higher tested 283 were 0.068 and 0.254 g/g of consumed sugar at the end of cultivation, respectively (Table I). 259 temperature (150 °C) resulted in films with burnt edges and was not further used. The results 284 Moreover, while in the semi-synthetic medium the production of glycerol was lower than the 260 confirmed the self-binding ability of pectin as shown in Figure 2. 285 production of ethanol (0.12 g glycerol/g ethanol), the glycerol production was much higher 261 Using the solution casting method, Cavallaro et al. [15] prepared pectin films using 286 than the ethanol production in CWFS medium (34.5 g glycerol/g ethanol). The result suggests 262 polyethylene glycol (PEG) 20000 as plasticizer at a weight ratio pectin/plasticizer of 4. Below 287 different metabolic pathways were used in these media for glycerol production. This 263 this value, the obtained films were fragile and with several voids. In this study, glycerol 288 hypothesis is corroborated when taking into account the composition of the biomasses grown 264 concentrations higher than 30% resulted in very thin and mechanically weak films that were 289 in the different media. 265 not suitable for further mechanical analyses (data not shown). On the other hand, pectin 290 The protein concentration was only 0.31 g/g biomass for the R. oryzae cultivated in CWFS 266 powder could not be properly homogenized using less than 30% glycerol. Thus, glycerol 291 medium while it was 0.60 g/g biomass for the fungus cultivated in semi-synthetic medium. In 267 content of 30% was found to be the optimum value to prepare the pectin films. 292 contrast, the total lipid concentration (g/g) in biomass was 0.2 g/g for the fungus cultivated in 268 293 CWFS medium, and only 0.1 g/g for the fungus cultivated in semi-synthetic medium. The

12 13

294 difference can be related to the different nitrogen composition of the media. The semi- 319 based films. The thickness of pectin films was 1.03 mm which was reduced to 0.2-0.4 mm

295 synthetic medium was prepared with a total nitrogen content of 2.0 g/L, while nitrogen 320 (Table II) in pectin-fungal biomass films. The biomass-incorporated pectin films were more

296 concentration was 1.3 g/L in CWFS medium, as determined by the Kjeldahl method, and no 321 flexible and softer than pure pectin films. Visually, incorporation of the fungal biomass

297 supplementation was used. 322 resulted in darker and more opaque films (Figures 2D, E and F).

298 Protein synthesis in microorganisms is mostly influenced by the nitogen source in the 323

299 cultivation medium. Microbial growth under nitrogen starvation, on the other hand, leads to 324 3.4.Scanning Electron Microscopy (SEM) of pectin based films 300 accumulation of lipids, an adaptation necessary to survive under stress conditions. In such 325 The surface SEM images of the pectin based films are shown in Figure 4. A bumpy, dense, 301 scenario, the tricarboxylic acid cycle is interrupted, causing accumulation of citrate in the cell. 326 and cracked surface with randomly distributed small particles (in white color) was observed 302 Citrate is then converted into oxaloacetate, which in turn is converted into malate. The last is 327 in pure pectin film (Figure 4A). The small particles on the surface of the pure pectin film were 303 further converted into pyruvate. All this reactions cause a net production of NADPH and 328 identified as non-homogenized pectin debris in the pectin-glycerol matrix. With addition of 304 acetyl-coenzyme A, which are used to produce fatty acids and triacylglycerides (TAGs) [30]. 329 10% of low-protein biomass (i.e. cultivated in CWFS medium), the small white particles 305 In fact, fatty acid metabolism is much more complex and involves several other reactions 330 disappeared and the film showed a coarse surface with few agglomerates (Figure 4B). The 306 which are not part of this study [31]. The accumulation of lipid in the cells occurs via a 331 obtained result indicates that biomass inclusion improves the uniformity of the pectin-glycerol 307 dynamic equilibrium of fatty acid synthesis and degradation [31]. When degraded by lipases, 332 matrix. Moreover, the SEM images indicate the fungal cells (that commonly present a 308 TAGs are hydrolyzed into free fatty acids and glycerol [32]. Production of glycerol during the 333 cylindrical shape) were disrupted since no structure similar to the cells has been observed. 309 course of fungal cultivation in CWFS has probably been performed though a similar 334 Disruption of the cells means the intracellular components (e.g., proteins) have been released 310 mechanism. Low ethanol yield and high lipid content in the biomass confirm this hypothesis. 335 in the biofilm matrix and could interact with the pectin molecules. Pectin films with addition 311 In semi-synthetic medium however, glycerol has been produced as a result of the alcoholic 336 of 20% of protein-rich and low-protein biomasses presented more agglomerates (Figure 4C 312 fermentation in order to maintain the cytosolic redox balance [33]. 337 and 4D, respectively) than the 10% low protein biomass film (Figure 4B). These 313 338 agglomerates, which were not present in the pure pectin film, indicate the incompatibility of 314 3.3. Incorporation of fungal biomass into pectin films 339 the biomass components and pectin, leading to the formation of heterogeneous structures [34]. 315 Pectin-fungal biomass films were prepared using 30% glycerol. The concentration of fungal 340 For a deeper analysis of the morphology, the cross-sectional images of pectin films were 316 biomass in the pectin films varied between 0 and 35%. Blends containing more than 35% of 341 obtained and are presented in Figure 5. Pure pectin film had a rugged structure with gaps, 317 biomass were not suitable for compression molding (data not shown). Thickness of the 342 which irregularly appeared in the film. The film was organized like a group of closely packed 318 obtained films was generally decreased by increasing the biomass concentration in the pectin-

14 15

343 surfaces (Figure 5A). The lack of a solvent in the blend matrix might be the reason for the 367 from literature. All the films exhibited the same behavior, with the TS and yield strength at

344 resulted structure and the poor homogenization of the pectin and glycerol. Interestingly, a 368 the same value; i.e., the rupture of the material happened during the elastic behavior. This

345 more uniform texture was observed in the pectin films with addition of fungal biomass, with 369 means the stress vs. strain graphs were straight lines whose linear coefficients represent the

346 fewer gaps in the cross-sectional area compared to pectin film (Figures 5B and 5C). 370 Young’s modulus. Pectin-glycerol blend yielded 1.03-mm thick films with TS of 15.7 MPa

371 and E% of 5.5 (Table II). The thickness of the film in a compression-molding process may be 347 The pectin film containing low-protein biomass (Figure 5C) showed a clumpy structure with 372 controlled by the concentration of plasticizer and the applied pressure. Kang et al. [36] 348 few holes on the plane. The same structure was observed for the protein-rich biomass 373 produced pectin films using the solution casting method with lower thickness (0.17 mm), E% 349 containing film with more structural discontinuities. This observation could be a result of the 374 (3.45), and TS (0.153 MPa) compared to the film prepared in this study. Pectin-glycerol films 350 biomass composition (protein and lipid contents) and pectin interactions with biomass. 375 prepared by Liu et al. (2007) [11] using the solution casting method yielded higher TS (17.0 351 Moreover, a different internal organization was observed according to the film composition. 376 MPa), and lower E% (2.5%) and thickness (0.15 mm). Similarly, Cavallaro et al. [15], 352 The less-oriented network of the pectin film compared to the composite pectin films yielded 377 produced much thinner films (0.06 mm) with higher TS (26 MPa) and lower E% (1.6%) using 353 greater thickness (Table II), denser texture as well as cracks and gaps in the cross-section. 378 PEG 20 000 as plasticizer. Fishman et al. (2004) [37] produced pectin-glycerol films at the 354 Solution-casted pectin films showed a smoother morphology, according to Liu et al. (2007) 379 same composition of this study by extrusion method and got films with lower TS (9.9 MPa) 355 [11] and Galus & Lenart (2013) [35], compared to the compression-molded pectin film in this 380 and higher E% (10.9). When comparing the Young’s modulus, the film obtained in this study 356 study. This might be due to better homogenization of the materials in casting method 381 (298 MPa) had a value similar to the one obtained when preparing the pectin film by 357 compared to the compression molding technique. Solution casted soy-protein isolate films 382 extrusion (201 MPa) [37]. On the other hand, the pectin films reported in the literature 358 unmodified and modified with chitosan, however, showed rough and bubbly structure [34] 383 prepared by solution casting had much higher values for the Young’s modulus (1082 [11] and 359 similar to the biomass-containing pectin films of this study. Macroscopically, a rough, dense 384 2650 MPa [15]). 360 and brittle texture has been observed for a solution-casted pectin-soy flour protein (90-10%) 385 Promising results was obtained when the fungal biomass was incorporated in the pectin films 361 film [11]. This is probably a consequence of the incompatibility of the pectin with proteins, 386 up to 20%. In this range, the tensile yields of the low-protein biomass films were 15.9, 12.7, 362 which are one of the major ingredients of the fungal biomass. 387 and 7.6 MPa for 10, 15, and 20% biomass ratio, respectively. The respective E% was 3.5, 2.5, 363 3.5.Mechanical Properties of films 388 and 3.8%. The films made with the protein-rich biomass exhibited tensile strength of 19.3,

364 The tensile strength (TS), elongation at break (E%), and Young’s modulus of the films are 389 19.2, and 16.1 MPa, and E% of 1.6, 1.9, and 4.3%, respectively. Comparing the results, the

365 measured. Table II shows tensile properties and thickness values of pectin-based films at 390 pectin films with protein-rich biomass demonstrated better mechanical properties than the

366 various blend ratios with fungal biomass obtained in this study as well as reference values 391 pectin films with low protein content biomass. Therefore, the composition of the fungal

16 17

392 biomass had a specific importance on the mechanical properties of the pectin films. Liu et al. 417 that is used before solution casting of whey protein isolate films. The type of protein used and

393 (2007) [11] argued that, at the conditions they tested, the active aldehyde of the pectin and the 418 the presence of other substances in the biomass are possible explanations of the obtained

394 primary amines of the tested proteins (soybean flour and fish skin gelatin) form cross-linking 419 results. The inclusion of the low-protein content fungal biomass in the films had little effect in

395 which can enhance the mechanical strength of the pectin-protein films. Young’s modulus of 420 the Young’s modulus as a result of the little changes produced in the TS and E% of the films.

396 the pectin films with the protein-rich biomass increased more than fourfold. Values above 421 A maximum increase of approximately 80% in the Young’s modulus was obtained when

397 1000 MPa were obtained when using 10 or 15% biomass in the blend. This is the result of the 422 using 15% of low-protein biomass in the film (533 MPa) compared to the pectin films without

398 increase in the TS and the decrease in the E%; i.e., the inclusion of the fungal biomass yielded 423 any fungal biomass (298 MPa).

399 more resistant films possibly because of the effect of the fungal biomass acting as fibers 424 The results of this study indicate that the removal of the solvent for production of pectin films 400 reinforcing the matrix. Nevertheless, loss of mechanical properties of the films was obtained 425 can have a positive effect on the mechanical characteristics of the films and incorporation of 401 when further increasing the biomass concentration in the blend matrix. This may be the result 426 fungal biomass can further improve these characteristics. 402 of the increase of the heterogeneous structures formed in the film when adding biomass to the 427 3.6.WVPC analysis of films 403 pectin matrix, as observed by SEM images as well (Figure 4). The values obtained for E% for

404 the films containing or not fungal biomass were not statistically different (95% confidence 428 WVPC analysis was conducted in this study to determine the water vapor permeability of the

405 interval). 429 films made with the protein-rich fungal biomass, which presented the most promising

430 mechanical characteristics. High WVPC (kg.s-1.m-1.Pa-1) means the film has a poor resistance 406 For the pectin films with low-protein content biomass, up to 15% of biomass yielded films 431 to the passage of water vapor. According to the obtained results (Figure 6), the pure pectin 407 with higher TS and, above 15%, they showed almost similar values compared to the pectin 432 film (0% biomass) showed the greatest values of WVPC (6.92 × 10-13 kg.s-1.m-1.Pa-1). 408 films reported in the literature [37, 38]. Pectin films made in this study with the protein-rich 433 However, it also presented the largest standard deviation. No statistical differences were 409 biomass showed higher TS than the films made by compression molding with 30% GC and 434 observed in the WVPC values of pectin films containing protein-rich biomass (p > 0.05); the 410 70% whey protein [39], and solution casting pectin-starch film with 30% GC [38]. The films 435 films showed values ranging from 2.35 ×10-13 (35% biomass content) to 3.58 × 10-13 kg.s-1.m- 411 of this study also showed lower TS and higher E% than the films obtained by the solution 436 1.Pa-1 (30% biomass content). 412 casting of pectin-soy flour protein (10% protein) film [11]. This is in contrast to the results

413 obtained by Sothornvit et al. (2007) [39] who obtained higher TS in the compression-molded 437 Overall, addition of biomass to the pectin films decreased the WVPC, hence fungal biomass

414 whey protein isolate films compared to those produced using the solution casting method. 438 may have increased the impermeabilization of the films to the passage of water vapor.

415 They argued that the high heat and pressure applied during the molding process may have 439 Moreover, addition of biomass also reduced the thickness of the films and this may have

416 induced a higher rate of cross-linking of the protein chains compared to the heat denaturing 440 contributed to the WVPC results. As observed in the SEM images of the cross-sectional area,

18 19

441 addition of biomass in the matrix reduced the gaps in the structure of the films. Besides 464 This work was financed by the Coordination for the Improvement of Higher Educational

442 reducing the thickness, the removal of these gaps may have also contributed to the 465 Personnel (Capes-Brazil).

443 impermeabilization of the films, owing to the more compact structure. Nevertheless, further 466

444 investigation is needed to determine the effectiveness of the biomass addition to the pectin 467 Conflict of Interest

445 films in reducing the WVPC. 468 The authors declare that they have no conflict of interest.

469 446 All the films prepared in the current study using pectin demonstrated higher WVPC than the 470 References 447 solution casting chitosan-carboxymethyl chitosan films (0.59 ×10-14 – 1.125 ×10-14 kg.s-1.m- 471 1. R. Ackah, D. Carboo and E. T. Gyamfi, Challenges of plastic waste disposal in Ghana : a 1 -1 448 .Pa ) from literature [24]. The pectin-fungal biomass films also presented higher WVPC than 472 case study of solid waste disposal sites in Accra. Elixir Mgmt. Arts 49 (2012) 9879-

449 the solution casting films prepared from locust bean gum and plasticizer (polyethylene glycol 473 9885. 474 2. B. Ruiz and X. Flotats, Citrus essential oils and their influence on the anaerobic digestion 450 200, glycerol, propylene glycol, or sorbitol) which ranged between 1 × 10-14 and 6 × 10-14 475 process: an overview. Waste Manage. 34 (2014) 2063-2079. -1 -1 -1 451 kg.s .m .Pa [40]. 476 3. M. Lohrasbi, M. Pourbafrani, C. Niklasson and M. J. Taherzadeh, Process design and 477 economic analysis of a citrus waste biorefinery with biofuels and limonene as products. 452 478 Bioresource Technol. 101 (2010) 7382-7388. 479 4. European Commission, Directive 2008/98/EC of the European Parliament and of the 453 4. Conclusion 480 Council of 19 November 2008 on waste. Official J of the European Union L 312 (2008) 454 Getting benefit of the self-binding capacity of pectin, compression molding has been proven 481 22.11.

455 to be an efficient method to produce pectin-based bioplastics, yielding films with interesting 482 5. B. Satari, K. Karimi, M. Taherzadeh and A. Zamani, Co-Production of Fungal Biomass 483 Derived Constituents and Ethanol from Citrus Wastes Free Sugars without Auxiliary 456 mechanical characteristics. Moreover, incorporation of protein-rich fungal biomass up to 15% 484 Nutrients in Airlift Bioreactor. Int J Mol Sci 17 (2016) 302. 457 decreased the presence of debris, increased the Young’s modulus, and decreased the water 485 6. P. R. Lennartsson, P. Ylitervo, C. Larsson, L. Edebo and M. J. Taherzadeh, Growth

458 vapor permeability of the pectin films. R. oryzae is a very versatile microorganism and was 486 tolerance of Zygomycetes Mucor indicus in orange peel hydrolysate without 487 detoxification. Process Biochem 47 (2012) 836-842. 459 able to grow in a medium with nutrients extracted from citrus waste without nutritional 488 7. F. Talebnia, M.P. Bafrani, M. Lundin and M. Taherzadeh, Optimization study of citrus 460 supplementation. The protein content of the fungal biomass positively favored the mechanical 489 wastes saccharification by dilute acid hydrolysis. BioResources 3 (2007) 108-122.

461 properties of the bioplastic. 490 8. M. Ochoa-Villarreal, E. Aispuro-Hernández, I. Vargas-Arispuro and M. Á. Martínez- 491 Téllez, Plant cell wall polymers: function, structure and biological activity of their 462 492 derivatives, in: Polymerization, ed. A. S. Gomes (IntechOpen, London, 2012), pp. 63-86. 463 Acknowledgements 493 9. S. Pilla, Handbook of bioplastics and biocomposites engineering applications (John 494 Wiley & Sons, New Jersey, 2011).

20 21

495 10. U. Siemann, Solvent cast technology – a versatile tool for thin film production, in: 528 22. S. Majdejabbari, H. Barghi and M.J. Taherzadeh, Synthesis and properties of a novel 496 Scattering Methods and the Properties of Polymer Materials, eds. N. Stribeck and B. 529 biosuperabsorbent from alkali soluble Rhizomucor pusillus proteins. Appl Microbiol 497 Smarsly (Springer, Berlin, 2005), pp. 307-316. 530 Biot, 92 (2011) 1171-1177. 498 11. L. Liu, L. Liu, C.-K. Liu, M. L. Fishman and K. B. Hicks, Composite films from pectin 531 23. V. D. Alves, S. Mali, A. Beléia and M. V. E. Grossmann, Effect of glycerol and amylose 499 and fish skin gelatin or soybean flour protein. J Agr Food Chem 55 (2007) 2349-2355. 532 enrichment on cassava starch film properties. J Food Eng, 78 (2007) 941-946. 500 12. V. Bátori, D. Åkeson, A. Zamani and M. J. Taherzadeh, Pectin Based Composites, in: 533 24. S. Dayarian, A. Zamani, A. Moheb and M. Masoomi, Physico-mechanical properties of

501 Handbook of Composites from Renewable Materials, eds. V. K.Ǧ Thakur, M. K. Thakur 534 films of chitosan, carboxymethyl chitosan, and their blends. J Polym Environ, 22 (2014) 502 and M. R. Kessler (Scrivener Publishing, Beverly, 2017), pp. 487-517. 535 409-416. 503 13. F. Zubeldía, M. R. Ansorena and N. E. Marcovich, Wheat gluten films obtained by 536 25. C. Caner, P. J. Vergano and J. L. Wiles, Chitosan Film Mechanical and Permeation 504 compression molding. Polym Test, 43 (2015) 68-77. 537 Properties as Affected by Acid, Plasticizer, and Storage. J Food Sci, 63 (1998) 1049- 505 14. A. Jerez, P. Partal, I. Martínez, C. Gallegos and A. Guerrero, Protein-based bioplastics: 538 1053. 506 effect of thermo-mechanical processing. Rheol Acta, 46 (2007) 711-720. 539 26. V. Bátori, M. Jabbari, D. Åkesson, P. R. Lennartsson, M. J. Taherzadeh and A. Zamani, 507 15. G. Cavallaro, G. Lazzara and S. Mioloto, Sustainable nanocomposites based on halloysite 540 Production of Pectin-Cellulose Biofilms: A New Approach for Citrus Waste Recycling. 508 nanotubes and pectin/polyethylene glycol blend. Polym Degrad Stabil, 98 (2013) 2529- 541 Int J Polym Sci, 2017 (2017). 509 2536. 542 27. T. Pintiaux, D. Viet, V. Vandenbossche, L. Rigal and A. Rouilly, High Pressure 510 16. M. Makaremi, P. Pasbakhsh, G. Cavallaro, G. Lazzara, Y. K. Aw, S. M. Lee and S. 543 Compression-Molding of alpha-Cellulose and Effects of Operating Conditions. Materials 511 Milioto, Effect of Morphology and Size of Halloysite Nanotubes on Functional Pectin 544 (Basel), 6 (2013) 2240-2261. 512 Bionanocomposites for Food Packaging Applications. ACS Appl Mater Inter, 9 (2017) 545 28. E. M. Ciannamea, P. M. Stefani and R. A. Ruseckaite, Physical and mechanical 513 17476-17488. 546 properties of compression molded and solution casting soybean protein concentrate based 514 17. G. Gorrasi, V. Bugatti and V. Vittoria, Pectins filled with LDH-antimicrobial molecules: 547 films. Food Hydrocolloids, 38 (2014) 193-204. 515 Preparation, characterization and physical properties. Carbohyd Polym, 89 (2012) 132- 548 29. C. Gao, M. Stading, N. Wellner, M. L. Parker, T. R. Noel, E. N. C. Mills and P. S. 516 137. 549 Belton, Plasticization of a Protein-Based Film by Glycerol: A Spectroscopic, 517 18. L. Bonnaillie, H. Zhang, S. Akkurt, K. Yam and P. Tomasula, Casein Films: The Effects 550 Mechanical, and Thermal Study. J Agr Food Chem, 54 (2006) 4611-4616. 518 of Formulation, Environmental Conditions and the Addition of Citric Pectin on the 551 30. M. Jin, P. J. Slininger, B. S. Dien, S. Waghmode, B. R. Moser, A. Orjuela, L. da Costa 519 Structure and Mechanical Properties. Polymers, 6 (2014) 2018. 552 Sousa and V. Balan, Microbial lipid-based lignocellulosic biorefinery: feasibility and 520 19. R. Porta, L. Mariniello, P. Di Pierro, A. Sorrentino and C. V. L. Giosafatto, 553 challenges. Trends Biotechnol, 33 (2015) 43-54. 521 Transglutaminase Crosslinked Pectin- and Chitosan-based Edible Films: A Review. Crit 554 31. H. Chen, G. Hao, L. Wang, H. Wang, Z. Gu, L. Liu, H. Zhang, W. Chen and Y. Q. Chen, 522 Rev Food Sci, 51 (2011) 223-238. 555 Identification of a critical determinant that enables efficient fatty acid synthesis in 523 20. V. A. Cataldo, G. Cavallaro, G. Lazzara, S. Milioto and F. Parisi, Coffee grounds as filler 556 oleaginous fungi. Sci Rep-UK, 5 (2015). 524 for pectin: Green composites with competitive performances dependent on the UV 557 32. J. Wongwatanapaiboon, W. Malilas, C. Ruangchainikom, G. Thummadetsak, S. 525 irradiation. Carbohyd Polym, 170 (2017) 198-205. 558 Chulalaksananukul, A. Marty and W. Chulalaksananukul, Overexpression of Fusarium 526 21. A. Mahboubi, J. A. Ferreira, M. J. Taherzadeh and P. R. Lennartsson, Value-added 559 solani lipase in Pichia pastoris and its application in lipid degradation. Biotechnol Biotec 527 products from dairy waste using edible fungi. Waste Manage, 59 (2017) 518-525. 560 Eq, 30 (2016) 885-893.

22 23

561 33. K. T. Scanes, S. Hohrnann and B. A. Prior, Glycerol production by the yeast 583 FIGURE CAPTIONS 562 Saccharomyces cerevisiae and its relevance to wine: a review. S Afr J Enol Vitic, 19 563 (1998) 17-24. 564 34. K. Li, S. Jin, X. Liu, H. Chen, J. He and J. Li, Preparation and characterization of 565 chitosan/soy protein isolate nanocomposite film reinforced by Cu nanoclusters. Polymers, 566 9 (2017) 247. 567 35. S. Galus, and A. Lenart, Development and characterization of composite edible films 568 based on sodium alginate and pectin. J Food Eng, 115 (2013) 459-465. 569 36. H. J. Kang, , C. Jo, N. Y. Lee, J. H. Kwon and M. W. Byun, A combination of gamma 570 irradiation and CaCl2 immersion for a pectin-based biodegradable film. Carbohyd Polym, 571 60 (2005) 547-551. 572 37. M. L. Fishman, D. R. Coffin, C. I. Onwulata and R. P. Konstance, Extrusion of pectin 573 and glycerol with various combinations of orange albedo and starch. Carbohyd Polym, 57 574 (2004) 401-413. 575 38. M. L.Fishman, D. R. Coffin, J. J. Unruh and T. Ly, Pectin/starch/glycerol films: Blends 576 or composites? J Macromol Sci A, 33 (1996) 639-654. 577 39. R. Sothornvit, , C.W. Olsen, T.H. McHugh, and J.M. Krochta, Tensile properties of 578 compression-molded whey protein sheets: Determination of molding condition and 579 glycerol-content effects and comparison with solution-cast films. J Food Eng, 78 (2007) 580 855-860. 581 40. Ö. A. Bozdemir and M. Tutaş, Plasticiser Effect on Water Vapour Permeability 582 Properties of Locust bean gum-Based Edible Films. Turk J Chem, 27 (2003) 773-782.

584

585 Figure 1: Steps involved in preparation and production of pectin fungal biomass films.

24 25

) 20 (a) 15 10 5 0 Con centration (g/L centration Con 0 20 40 60 Time (h) 10 (b) 8 6 4 2 586 0 Concentration (g/L) Concentration 0 10 20 587 Figure 2: Optical photos of the pectin-based biofilms. A: pectin powder; B: pectin (70%) - Time (h) 588 glycerol (30%) dough-like matrix; C: pectin (70%) - glycerol (30%) bioplastic; D: CWFS 593

589 medium cultivated biomass (20%) – pectin (50%); E: semi-synthetic medium cultivated 594 Figure 3: Concentration profile of glucose (▲), other sugars (♦) (in xylose equivalent), glycerol 590 biomass (15%) – pectin (55%); F: semi-synthetic medium cultivated biomass (25%) – pectin 595 (■) and ethanol (*) during cultivation of R. oryzae in a) CWFS medium and b) semi-synthetic 591 (45%). 596 medium in a 26 L bubble column bioreactor. 592

597

26 27

598

599 Figure 4: SEM micrographs of surface of compression molded 30% GC pectin based bio

600 plastic films. A) Pectin film without biomass (70:30); B) Pectin- Low protein biomass film

601 (60:10); C) Pectin-Protein rich biomass film (50:20); D) Pectin-Low protein biomass film

602 (50:20), (Magnification = 1000X). 603

604 Figure 5: SEM images of the cross-section of compression molded 30% GC pectin based bio

605 plastic films. A) Pectin film without biomass (70:30); B) Pectin-Protein rich biomass film

606 (50:20); C) Pectin-Low protein biomass film (50:20), (Magnification = 1500X).

28 29

611 TABLES 13 8 612 Table I: The yields of fungal biomass and primary metabolites (in g/g sugar consumed), as ) x 10 -1 6 613 well as protein and lipid contents (in g/g biomas) of fungal biomass obtained in CWFS and .Pa -1

.m 614 semisynthetic media after 72 and 24 h cultivation, respectively. Data presented as average ±

-1 4 615 standard deviation and n=3. 2

WVPC WVPC (kg.s 0 CWFS Semi-synthetic 0 10 20 25 30 35 Component medium medium % of fungal biomass in the matrix 607 Ethanol 0.004 ± 0.000 0.254 ± 0.017 608 Figure 6: WVPC of pectin films with incorporation of protein-rich fungal biomass at different Lactic acid 0.013 ± 0.004 0.068 ± 0.002 609 compositions (%) of fungal biomass. Glycerol 0.138 ± 0.030 0.029 ± 0.001 Fungal biomass 0.26 ± 0.03 0.35 ± 0.01 610 Protein 0.31 ± 0.02 0.60 ± 0.12 Lipid 0.20 ± 0.04 0.10 ± 0.01 616

30 31

617 Table II: Mechanical properties and thickness of the pectin films obtained by heat

618 compression molding (glycerol content of 30%) with and without incorporation of fungal

619 biomass, in this study, as well as solution casting, according to [11, 15, 36-39].

Young’s Pectin Biomass Cultivation Tensile strengh Elongation at Thickness modulus Reference (%) (%) medium (MPa) break (%) (mm) (MPa) 70 0 - 15.7 ± 0.53 5.5 ± 1.7 298 1.03 a 15.9 ± 0.84 3.5 ± 1.6 499 0.42 60 10 b 19.3 ± 3.5 1.6 ± 0.9 1350 0.41 a 12.7 ± 1.29 2.5 ± 0.9 533 0.4 55 15 b 19.2 ± 2.9 1.9 ± 1.3 1230 0.35 a 7.6 ± 0.64 3.8 ± 1.1 206 0.35 50 20 b 16.1 ± 2.3 4.3 ± 2.7 446 0.38 a 5.2 ± 1.1 1.6 ± 1.1 393 0.27 45 25 b 12.5 ± 1.27 3.1 ± 0.6 406 0.34 a 6.5 ± 2.01 1.4 ± 0.4 462 0.25 40 30 b 11.2 ± 2.12 4.5 ± 1.0 250 0.32 a 7.1 ± 1.46 3.8 ± 0.8 187 0.28 35 35 b 9.2 ± 0.81 4.1 ± 1.5 237 0.34 Solution casting Pectin-glycerol 17.0 ± 3.4 2.5 ± 0.6 1082 0.15 [11] film Solution casting (90 -10 %) Pectin- 24.0 ± 3.0 2.9 ± 0.9 1213 - [11] soy flour protein film Solution casting (80 -20 %) Pectin- 26 1.52 2650 0.06 [15] PEG 20 000 film Solution casting Pectin-polyvinyl 0.153 3.45 - 0.17 [36] alcohol-glycerol (40-20-40%) film Extruded pectin-glyerol (70-30%) 9.9 ± 2.4 12.8 ± 6.8 201 - [37] film Compression molded Whey 10 ± 0.3 43 ± 15 251 1.32 [39] protein-glycerol (70-30%) film Solution casting Pectin-Starch (40- 21 12 500 - [38] 30%) film with 30% GC 620 a CWFS; b Semi-synthetic

32