Investigating the methane producing pathway in lab-scale biogas reactors subjected to sequential increase of ammonium and daily acetate-pulsing

(1)

Linköping University | Department of Physics, Chemistry and Biology Master thesis, 30 hp | M. Sc. Chemical Biology, Protein Science and Technology Spring term 2020 | LITH-IFM-A-EX—20/3775--SE

Investigating the methane

producing pathway in lab-scale

biogas reactors subjected to

sequential increase of ammonium

and daily acetate-pulsing

Swedish University of Agricultural sciences

Sofia Moberg

Examinator: Martin Karlsson Supervisor, LiU: Sandra Waern Supervisor, SLU: Bettina Müller

(2)

Datum

Date 2020-06-18 Avdelning, institution

Division, Department

Department of Physics, Chemistry and Biology Linköping University

URL för elektronisk version

ISBN

ISRN: LITH-IFM-A-EX--20/3775--SE

_________________________________________________________________

Serietitel och serienummer ISSN

Title of series, numbering ______________________________

Språk Language Svenska/Swedish Engelska/English ________________ Rapporttyp Report category Licentiatavhandling Examensarbete C-uppsats D-uppsats Övrig rapport _____________ Titel Title

Investigating the methane producing pathway in lab-scale biogas reactors subjected to sequential increase of ammonium and daily acetate-pulsing

Författare

Author Sofia Moberg

Nyckelord

Keyword

Syntrophic acetate oxidation, syntrophic acetate oxidizing bacteria, Wood-Ljungdahl pathway,

Sammanfattning

Abstract

Syntrophic acetate oxidizing bacteria converts acetate into hydrogen and carbon dioxide and through the mutualistic syntrophic partnership with methanogens the products are further converted to methane in biogas processes operating at high ammonia concentrations. There is very little known about SAOBs, only five have been characterized and had their genome analyzed. The aim of this project is to gain further knowledge about the methane producing pathway of SAOBs with a proteomic approach. Proteins were extracted from biogas sludge with a phenol-based approach and trypsin digestion, peptide recovery were performed using the Suspension Trapping method. Measurement of the peptide content was made with LC-MS/MS. The peptide profiles obtained were screened for the proteins expressed of the mesophilic SAOB Syntrophaceticus schinkii. The data supports earlier suggestion that it utilizes the Wood-Ljungdahl pathway for hydrogen production. Furthermore, the peptide profile revealed that enzymes for the glycine reductase complex and the glycine cleavage system were expressed during high ammonia concentration, indicating a potential role of these enzymes in the methane producing pathway. However, due to partial failure of the sample preparation for mass spectrometry measurements no quantification conclusions could be made. A discussion on how to further improve sample preparation methods as well as how to access the proteome to a large extent is presented.

(3)

Abstract

Syntrophic acetate oxidizing bacteria convert acetate into hydrogen and carbon dioxide and through the mutualistic syntrophic partnership with methanogens the products are further converted to methane in biogas processes operating at high ammonia concentrations. There is very little known about SAOBs, only five have been characterized and had their genome analyzed. The aim of this project was to gain further knowledge about the methane producing pathway of SAOBs with a proteomic approach. Proteins were extracted from biogas sludge with a phenol-based approach and trypsin digestion and peptide recovery were performed using the Suspension Trapping method. Measurement of the peptide content was made with LC-MS/MS. The peptide profiles obtained were screened for the proteins expressed of the mesophilic SAOB Syntrophaceticus schinkii. The data supports earlier suggestions that it utilizes the Wood-Ljungdahl pathway for hydrogen production. Furthermore, the peptide profile revealed that enzymes for the glycine reductase complex and the glycine cleavage system were expressed during high ammonia concentration, indicating a potential role of these enzymes in the methane producing pathway. However, due to partial failure of the sample preparation for mass spectrometry measurements no quantification conclusions could be made. A discussion on how to further improve sample preparation methods as well as how to access the proteome to a large extent is presented.

(4)

Abbreviations

BSA Bovine serum albumin

DTT Dithiothreitol

Ech Energy-converting hydrogenase

FTHFS Formyl-H4F synthase

IAA Iodoacetamide

LC-MS/MS Liquid chromatography - tandem mass spectrometry

MS Mass spectrometry

Rnf Ferredoxin-NAD:oxidoreductase

RT Room temperature

SAO Syntrophic acetate oxidation

SAOB Syntrophic acetate oxidizing bacteria

SDS-PAGE Sodium dodecyl sulfate polyacrylamide gel electrophoresis

STrap Suspension Trapping

TFA Trifluoroacetic acid

VFA Volatile fatty acids

(5)

1 Introduction

1.1 Background

Biogas production has become more widespread during the last couple of years because of its environmental benefits such as reducing the use of fossil resources and waste managing. Organic waste from households, industries and agriculture are used for the anaerobic

digestion, which results in methane rich biogas (Merlin Christy, et al., 2014). However, there are some challenges with using high energy biomass as substrates. For example, degradation of waste with high levels of protein results, in addition to biogas, in high levels of ammonium and ammonia in the process. Ammonia is released during protein degradation and is known to inhibit microbial processes, causing process instability, and affecting the biogas production (Hagos, et al., 2017).

Ammonia inhibits the acetoclastic methanogenesis which under low ammonia conditions converts acetate that has been formed during the anaerobic digestion to methane (Yenigün & Demirel, 2013). Under high ammonia conditions the acetate is instead oxidized to hydrogen and carbon dioxide by syntrophic acetate oxidizing bacteria (SAOB) which cooperate with hydrogenotrophic methanogens that converts the hydrogenand carbon dioxide to methane (Rajagopal, et al., 2013).

There are large quantities of protein rich waste available from households and the food industry, the substrate has also a relatively high energy potential compared to other biogas substrates. Furthermore, the resulting ammonium rich biogas residue is a good fertilizer which lowers the demand for mineral produced fertilizer and increases the profit of the biogas since the residue will be more valuable (Hagos, et al., 2017), (Westerholm, et al., 2016). These are the main reasons for performing anaerobic digestion at high protein and ammonium

concentrations.

To be able to increase efficiency in processes operating at high ammonia concentration, knowledge about the syntrophic acetate oxidation (SAO) and the SAOBs are required. At present time, very little is known about SAOBs and only a few SAOBs have been

characterized. However, for a couple of SAOBs the genome has been analyzed which to some extent have given a better understanding of the process (Mosbæk, et al., 2016) (Manzoor, et al., 2018). This knowledge is gained from SAOBs in pure culture, co-culture and enrichment culture. However, there are no current knowledge of the change of the methanogenic

pathways during step wise increasing of ammonium. There are also several community studies but only few functional studies which leads to a knowledge gap.

1.2 Aim of the project

The aim of this project was to investigate the methane producing pathway during increasing ammonia levels over time by analyzing the functional activities using a meta-proteomics approach. Increasing knowledge about the methane producing pathways will make it possible to impact on the efficiency of biogas production by supporting and stabilizing SAOBs.

The methodological approach included at first extraction and purification of proteins from the microorganisms in biogas sludge. Thereafter trypsin digestion with the suspension trapping

(9)

2 (STrap) sample preparation method was performed, the yielded peptides were analyzed with liquid chromatography tandem mass spectrometry (LC-MS/MS). The obtained metaproteome was screened for peptides, which mapped to the amino acid sequences deduced from all open reading frames of the genome of the SAOB S. schinkii.

The proteome will be a good complement to the genome that earlier has been used to increase knowledge about the SAOBs and will certainly contribute to further knowledge of the SAOBs and the pathways used.

1.3 Limitations

Due to the time limit of 20 weeks of this project the execution was restricted to the

intracellular proteome of the microorganisms as well as using only one method for extraction, purification, and digestion instead of comparing results from several methods. The number of samples that were analyzed was reduced due to the cost for the mass spectrometry analysis also meaning that each sample was measured once instead of, for example, as duplicates.

1.4 Ethical aspects and social relevance

There are no ethical conflicts regarding this project. On the contrary, increased knowledge and efficiency of the biogas process will most likely lead to a decreased use in fossil fuels and other non-reusable energy resources meaning it will contribute to more sustainable energy usage and waste management.

2 Process

In the beginning of the project a time plan in form of a Gantt scheme, (see Appendix D) was established to give an overview of the project and planned activities. The activities regarded testing of various protocols, performance of extraction, purification and digestion for the samples and analysis of LC-MS/MS data. Activities regarding report writing, literature studies and other administrative parts of the project was also considered. Milestones was added as a tool to easier keep track of the plan.

At the end of every week the plan was compared to what had been done during the week, if needed, the plan was revised and adjustments for the upcoming weeks were made.

(10)

3

3 Theory and Methodology

3.1 Scientific background

3.1.1 Anaerobic digestion

Biogas is produced from the anaerobic digestion of organic matter and mainly consist of methane and carbon dioxide. However, in raw biogas also small amounts of hydrogen sulfide are found. Under the oxygen-free conditions of anaerobic digestion a complex community of microorganisms, such as bacteria and archaea, degrades organic compounds to biogas, nutrients and other elements in a synchronized manner (Merlin Christy, et al., 2014). The community structure of microorganisms has a big role in the stability of the anaerobic

digestion and the different subgroups needs to be coordinated and in balance, otherwise it can lead to metabolic imbalances, inhibition and failure of the reactor (Mosbæk, et al., 2016). Anaerobic digestion proceeds in a series of stages. The four main stages include, hydrolysis, acidogenesis, acetogenesis and methanogenesis (Figure 1). The first three stages are

performed by a large community of bacteria whereas the methanogenesis is performed by a specialized group of archaea known as methanogens (Hagen, et al., 2016). Both the hydrolytic and acidogenic microorganisms are fast growing whereas the acetogenic bacteria are slow growing and sensitive to both oxygen and fluctuations in environmental conditions. Similarly, methanogens are also slow growing and sensitive to environmentally changes and are

considered to perform the overall rate limiting step of the whole anaerobic digestion (Merlin Christy, et al., 2014).

Acidogenesis

Mono and oligomers

(Monosaccharides, amino acids, fatty acids)

Complex Polymers

(Proteins, lipids, polysaccharides)

Organic acids

(Butyric acid, Acetic acid, VFAs)

H

2

+ CO

2

Acetate

CH

4

+ CO

2 Hydrolysis Acetogenesis Hydrogenotrophic Methanogenesis Acetoclastic Methanogenesis

(11)

4

Figure 1 The process of anaerobic digestion as a flowchart. The process has four main stages, these are hydrolysis, acidogenesis, acetogenesis and methanogenesis (which can be divided into hydrogenotrophic- or acetoclastic methanogenesis). The flowchart is inspired by figure 1 in (Parawira, 2012) with modifications.

3.1.2 Hydrolysis

During hydrolysis, complex substrates such as proteins, lipids, polysaccharides, and other organic polymers are hydrolyzed into oligomers and monomers such as monosaccharides, amino acids, and fatty acids. This is done by hydrolytic enzymes secreted by a large number of hydrolytic bacteria. The yielded monomers can thereafter be utilized by the hydrolytic bacteria as well as other bacteria in the community (Merlin Christy, et al., 2014).

3.1.3 Acidogenesis

During acidogenesis, also known as primary fermentation, degradation of the smaller organic molecules resulting from the hydrolysis occur. This results in organic acids such as acetic acid, butyric acid, and other shorter volatile fatty acids (VFA). It also yields alcohols, lactate, hydrogen, and carbon dioxide. The amount of VFAs produced during this stage is important for the total biogas production of the process since the optimal precursors for the methane formation are acetic and butyric acids (Merlin Christy, et al., 2014). This is due to that methane can only be produced by methanogens with a limited number of substrates during anaerobic digestion. The main precursor is acetic acid. However, one mol of butyric acid can rapidly be converted to two mol acetic acid during anaerobic digestion, making it a major intermediate (Hwang, et al., 2001).

3.1.4 Acetogenesis

During acetogenesis, or anaerobic oxidation, the VFAs and alcohols formed during

acidogenesis are converted to acetate, hydrogen, and carbon dioxide. This however requires a partnership between different microorganisms since the degradation of these compounds depends on a very low hydrogen partial pressure. This is known as syntrophy (literally meaning “eating together”). The hydrogen formed by the acetogenic bacteria is consumed by hydrogen-consuming hydrogenotrophic methanogens which thereby keeps the hydrogen partial pressure low, meaning that the degradation of VFAs and alcohols to acetic acid can proceed (Merlin Christy, et al., 2014).

The degradation of, for example, butyrate and propionate also occur in syntrophy during acetogenesis. For example, the degradation of butyrate to methane and carbon dioxide requires three different bacteria that specializes in one of following reactions: conversion of butyrate to acetate and hydrogen, conversion of hydrogen and carbon dioxide to methane, and conversion of acetate to carbon dioxide and methane (Schink, 1997).

(12)

5 During methanogenesis methane is produced from hydrogen and carbon dioxide, and from acetate. This is achieved by two different pathways. During hydrogenotrophic

methanogenesis methane and water are formed from hydrogen and carbon dioxide, it is also common for these methanogens to be able to use formate as the electron donor instead of hydrogen. The second pathway is the acetoclastic methanogenesis which converts acetate to methane and carbon dioxide (Merlin Christy, et al., 2014).

3.1.6 Ammonia as an inhibitor in anaerobic digestion

One of the major toxic inhibitors of anaerobic digestion is ammonia which causes proton imbalance, potassium deficiency, changes of the intracellular pH and suppression of enzymatic reactions in the cells of the microorganisms which further also lead to

accumulation of VFAs. This makes the process unstable and reduces the production of biogas and methane (Wang, et al., 2020), (Westerholm, et al., 2016). The acetoclastic methanogens are the most ammonia sensitive microorganisms in the anaerobic digestion meaning that the growth of these methanogens is easily affected by high ammonia concentrations (Chen, et al., 2008). Ammonia (NH3) is in equilibrium with ammonium (NH4+) and the equilibrium can

shift depending on the process conditions.

Thus, inhibition of anaerobic digestion with ammonia is therefore affected by several factors such as temperature, pH, acclimation time and what kind of substrate is used. Temperature and pH will affect ammonia in such a way that when it increases, there will be an increase in ammonia inhibition since the equilibrium between ammonia and ammonium will shift towards the more toxic neutral ammonia yielding even higher ammonia concentrations (Westerholm, et al., 2016), (Chen, et al., 2008). Slowly increasing ammonia levels and letting the microorganisms acclimatize will make them more tolerant to high ammonia levels. This could be due to internal changes in the methanogens, changes in the population of

methanogens or changes in the microbial community structure (Chen, et al., 2008). The inhibiting concentration of total ammonia nitrogen can range from 1.7 g/L to 14 g/L where 50 % of the methane production is reduced (Chen, et al., 2008). However, several studies have suggested that the process is completely inhibited at total ammonia nitrogen concentrations around 3 g/L (Rajagopal, et al., 2013).

Nitrogen-rich substrate such as substrates with high levels of protein is the main source that increase the ammonia concentration.

3.1.7 Syntrophic acetate oxidation

With increasing ammonia levels, acetoclastic methanogenesis becomes inhibited and syntrophic acetate oxidation (SAO) starts to establish. During SAO, acetate is converted by syntrophic acetate oxidizing bacteria (SAOB) to hydrogen and carbon dioxide which then is converted via the hydrogenotrophic pathway to methane, as seen in Figure 2 (Manzoor, et al., 2018).

(13)

6

Figure 2 The pathway of SAO. Conversion of acetate to methane during high ammonia concentrations.

The immediate conversion and consumption of hydrogen to methane enables the otherwise thermodynamic unfavorable acetate oxidation. Combining the two reactions (equations (1) and (2)) yields a small combined free energy change that is negative, meaning that the overall reaction becomes thermodynamically favorable, see equation (3) (Manzoor, et al., 2018). CH3COO− +H+ _{+ 2H2O → 2CO2 + 4H2} _∆G0_′_{= +95 kJ per mol rct. (1)}

4H2 + CO2 → CH4 + 2H2O ∆G0_′_{= −131 kJ per mol rct. (2)}

CH3COO− + H+ → CH4 + CO2 ∆G0′ = −35 kJ per mol rct. (3)

The SAOBs and the hydrogenotrophic methanogens are said to be in syntrophy with each other meaning that the microorganisms are forced to cooperate as none of the microorganisms can degrade the substrate without the other. This results in a combined metabolic activity between the microorganisms that is impossible to accomplish with only one of them (Schink, 1997).

Currently, only five SAOBs have been characterized, all of which were originally isolated from anaerobic biogas reactors. Two of these are thermophilic (Thermacetogenium

phaeum and Pseudothermotoga lettingae) (Hattori, et al., 2000), (Balk, et al., 2002), and three

are mesophilic (Clostridium ultunense, Syntrophaceticus schinkii and Tepidanaerobacter

acetatoxydans) (Schnürer, et al., 1996), (Westerholm, et al., 2010), (Westerholm, et al., 2011).

3.1.8 Wood-Ljungdahl pathway

It is suggested that the mechanism used by the SAOBs to convert acetate is done with the Wood-Ljungdahl (WL) pathway which consists of a methyl and carbonyl branch (Figure 3) (Müller, et al., 2013). The majority of the characterized SAOBs are acetogens which are known to utilize the WL pathway (Manzoor, et al., 2016).

For majority of the SAOBs it is assumed that the oxidation of acetate is done with the reversed WL pathway. The gene for formyltetrahydrofolate synthase (FTHFS) have been discovered in all currently characterized SAOBs which suggests that the pathway, or fractions of the pathway is used (Müller, et al., 2013).

H

₂

+ CO

₂

Acetate

CH

₄

+ CO

₂ Hydrogenotrophic Methanogenesis Acetoclastic Methanogenesis Syntrophic acetate oxidation

X

(14)

7

Figure 3 Picture of the Wood-Ljungdahl pathway by Westerholm, et al. (2016). The pathway consists of two branches and several reactions which are needed for the reduction of carbon dioxide and production of acetate. One ATP is consumed, and one is produced, meaning that there is no formation of net ATP (Westerholm, et al., 2016). SAOBs utilize the reversed WL pathway.

3.1.9 Wood-Ljungdahl pathway in SAOBs

The three different mesophilic SAOBs have been characterized with a genome centric

approach which has yielded some knowledge about similarities and differences between them. For example, it is shown for S. schinkii that there are key genes that are encoded and

transcribed for the whole WL pathway in the genome (Manzoor, et al., 2016). However, in the genome of T. acetatoxydans there are no genes for formate dehydrogenase needed for the last step of the methane branch resulting in formate, carbon dioxide and hydrogen meaning that the formate and hydrogen has to be consumed by the methanogenic partner and that there will be a further decrease in energy available for T .acetatoxydans compared to before (Müller, et

(15)

8 al., 2015). In the genome of C. ultunese several key enzymes for the WL pathway are missing. In fact, there are only genes for the methyl branch found. This suggests that there must be another pathway. For T. actatoxydans another route could be oxidative tricarboxylic acid cycle, however this has not been proved and C. ultunese lacks key genes for this route (Manzoor, et al., 2018).

S. schinkii also lacks key enzymes for the oxidative tricarboxylic acid cycle and therefore do

not utilize it. However, other than being able to use the WL pathway it is suggested that S.

schinkii can use an alternative pathway which is a combination of WL pathway and the

glycine cleavage system, this way the pathway only uses the methyl pathway of WL pathway.

S. schinkii has genes for this alternative pathway although they are not expressed. However,

further evidence for this route is needed to support it (Manzoor, et al., 2016).

3.1.10 Energy conserving systems in SAOBs

There is no coherent shared energy conserving system for the SAOBs, they differ from each organism. The only factor they have in common are the [Fe-Fe] hydrogenases which are encoded in the genomes of all the SAOBs (Manzoor, et al., 2018).

In S. schinkii both the energy-converting hydrogenase (Ech) and

ferredoxin-NAD:oxidoreductase (Rnf) are encoded in the genome. However, there are low transcription levels of the Rnf complex which agrees to the close relative T. phaeum which has no genes for Rnf at all. Hence it is suggested that Ech is important when it comes to energy

conservation in S. schinkii which is further proved by the fact that all subunits of Ech is expressed during high levels of ammonia (Manzoor, et al., 2016).

The genome of S. schninkii also codes for a large number of other hydrogenases. For example, there are several gene clusters for [Fe-Fe] hydrogenase and one for [Ni-Fe] hydrogenase which have all subunits expressed during high ammonia concentrations and is therefore suggested to be of importance for energy conservation. The genome also codes for a NAD(P)-binding oxidoreductase/heterodisulfide reductase complex which is believed to take part in reverse electron transport (Manzoor, et al., 2016).

3.2 Methodology

3.2.1 Protein extraction and purification

In the project presented here, it was of importance to extract proteins at high quality and with sufficient quantity. To extract intracellular proteins from the cell it needs to be lysed. There are several extractions methods that can be used for this purpose, for example utilizing different types of mechanical forces (bead beating, grinding), temperature differences (freezing and thawing) and sound (ultrasonication) (Saraswathy & Ramalingam, 2011). Chemical extraction such as the use of phenol could also be done either by itself or in addition to the methods mentioned above. To hinder degradation of proteins during the sample

preparation proteases needs to be inhibited, also this can be done with phenol. Furthermore, the extraction procedure using phenol allows for a high clean up capacity (Faurobert, et al., 2007). The extraction strategy is depending on type of protein and the source of the protein, however the use of phenol for protein extraction from biogas sludge is suggested to work well

(16)

9 (Heyer, et al., 2013). Another well-established way of extraction is in addition to phenol also use chloroform. Chloroform is compatible with phenol and is commonly used to separate proteins from DNA and RNA. Furthermore, membrane proteins are effectively extracted by chloroform (Xiong, et al., 2014), making it suitable for proteomic applications.

After extraction, purification from contaminating molecules need to be made. The fewer the purifications steps are the better, since it keeps the yield of the protein high (Saraswathy & Ramalingam, 2011).

3.2.2 SDS-PAGE

In order to verify successful protein purification and degradation, SDS PAGE was applied. SDS-PAGE, short for Sodium dodecyl sulfate polyacrylamide gel electrophoresis is a protein separation method were the proteins are separated according to the molecular weight. Proteins with smaller molecular weights will migrate faster through the polyacrylamide gel network (Fritsch & Krause, 2003). For the proteins to move in the same direction in the gel they need to obtain a uniform charge, this is achieved by adding SDS to the sample which gives the proteins a net negative charge. Furthermore, the native proteins are denatured by the SDS which disturbs the non-covalent forces that creates the three-dimensional protein structure, such as hydrogen bonding, hydrophobic interactions, and ionic interactions (Saraswathy & Ramalingam, 2011). This leads to a uniform free solution mobility for all proteins (Fritsch & Krause, 2003).

The procedure takes place in an electric field by which the SDS-treated proteins are first concentrated in the first part of the gel, the stacking gel. Thereafter the proteins are separated in the separating gel (Saraswathy & Ramalingam, 2011). The bands are visualized either by staining or with an unstained variant where the proteins are made fluorescent directly in the gel with a short photoactivation and is detected by a stain-fee enabled gel imager.

3.2.3 Protein digestion

To be able to analyze the metaproteome, the proteins must be digested to peptides under controlled circumstances by a protease such as trypsin. Thereafter the peptides must be washed and recovered. Some common methods for protein digestion are SDS-PAGE in gel digestion, FASP and STrap.

SDS can be used to enhance enzyme digestions. However, high levels of SDS reduces trypsin activity as well as causing suppression in MS. Other alternatives to SDS, such as organic solvents, chaotropic agents, etc. often introduces other problems such as reduced

solubilization potential and incompatibility with LC-MS/MS or the enzyme (Kachuk, et al., 2015). During in gel digestions the proteins are separated on the gel, thereafter the bands are excised and digested with, for example, trypsin. The method also removes compounds that could be harmful for a mass spectrometer, for example salts and detergents. However, there is a loss of peptides due to that some peptides is entrapped in the gel matrix during the

(17)

10 Filter aided sample preparation (FASP) removes SDS from protein samples applied to a filter by disrupting the SDS micelles with urea. Further treatment and digestion of the sample is done directly in the filter. The peptides can then be eluted and used for further measurements. However, the protocol is time consuming and there is a big variation between filter batches meaning it is not suitable for high-throughput proteomics studies (Ludwig, et al., 2018). Suspension trapping (STrap) is a sample preparation method used for bottom-up proteomics experiments which commonly utilizes tryptic proteolysis and peptide characterization with mass spectrometry. The STrap tip consists of a quartz fiber trap and reversed phase membrane C18 plugs. In the quartz trap a protein suspension is captured. This Protein suspension is formed when the acidified protein sample containing a SDS mixture is added together with a methanolic solution to the STrap-tip. The methanolic solution solubilizes the SDS which, together with other contaminants, is filtered out. Thereafter the proteins entrapped in the quartz are digested by a protease yielding peptides that are then transferred to the C18 part of the tip. The peptides are captured by the C18 membrane where it is desalted and lastly eluted (Zougman, et al., 2014). This method, compared to the previous mentioned methods, is considered to be a fast method and is also suitable for several types of proteins (Alvarado, et al., 2010), and has therefore been applied in this project.

3.2.4 LC-MS/MS

Mass spectrometry (MS) is an analytical method, often used for biochemical oriented issues such as proteome and metabolome applications. In this project, LC-MS/MS has been used to analyze the peptides representing the metaproteome of the samples analyzed.

During a MS measurement, the peptide sample is ionized and fragmented. The ions are then separated according to their mass-to-charge ratio. In the case of tandem mass spectrometry (MS/MS) the ion is fragmented one more time after the first mass detector, detected, and analyzed. The abundance of the ion is measured and converted into electrical signals. The signals are processed giving rise to a MS-spectrum which is plotted as abundance versus mass-to-charge ratio. Higher abundance yield a higher peak (de Hoffmann & Stroobant, 2007).

Analysis of more complex samples, as in this study, require that a separation method, for example liquid chromatography (LC) is coupled with the mass spectrometer. This is because the various compounds in the sample has to be introduced one at a time into the spectrometer (de Hoffmann & Stroobant, 2007). LC separates a sample with a liquid mobile phase that flows through a stationary phase. The sample molecules interact both with the mobile phase and the surface of the stationary phase. The molecules that bind the strongest to the stationary phase will take longer time to be eluted, meaning that a separation of the compounds in the sample will occur. The separation can be made with different factors, some common types of LC are separated with respect to polarity, hydrophobic interactions, charges, affinity for certain ligands and size (Desidero & Nibbering, 2008).

The mass of the compounds analyzed with MS/MS should not exceed a molecular weight of 4 kD because the fragmentation and data interpretation becomes more difficult at higher

molecular weights. However, peptides are an exception despite having relatively large molecules because they contain several bonds with chemical stability similar to each other.

(18)

11 Furthermore, the fragmentation patterns of peptides in the second mass detector gives

information about the amino acid sequence which leads to their identification. Ideally a peptide spectrum from MS/MS measurements will illustrate peaks in a ladder formation (Figure 4). The distances between the peaks will correspond to the amino acid sequence (Desidero & Nibbering, 2008).

Figure 4 A typical layout of a spectrum of a peptide measured with LC-MS/MS.

3.2.5 MaxQuant and Perseus

MaxQuant is a software that is used for MS-based shotgun proteomics data analysis which uses several advanced algorithms to identify peptides, proteins, and post translational

modification sites. The software utilizes Andromeda which is its own peptide database search engine. With MaxQuant it is possible to analyze data from various quantification techniques, for example label-free quantification and isobaric labeling (Tyanova, et al., 2016).

During the workflow two peptide searches are performed which allows MaxQuant to identify more than one peptide from on MS/MS spectrum as the second peptide search searches for signals from co-fragmentation of additional precursors. To estimate and control false-positive identifications, which is needed to be controlled in several levels of complexity, MaxQuant utilizes the target-decoy strategy. Posterior error probability is used within this strategy. It merges peptide properties, such as length, charge, and number of modifications together with the Andromeda score to yield the quality of a peptide spectrum match. Moreover, global false discovery rates are controlled during further workflow. (Tyanova, et al., 2016).

The Perseus software can be used to interpret data such as protein quantification, interaction and post-translational data from MaxQuant. The output tables from MaxQuant is loaded into

(19)

12 Perseus and based on the column name Perseus can identify type of data. Perseus can be used for several applications such as, expression proteomics, interaction proteomics, cross-omics data analysis etcetera (Tyanova, et al., 2016).

(20)

13

4 Materials and Methods

All instruments used are described in Appendix AAppendi.

4.1 Biogas reactor set-up

There were five lab-scale biogas reactors with a working volume of 5 liters and a total volume of 10 liters that the samples were taken from (Figure 5). They were fed household food waste and slaughterhouse waste from the municipality of Uppsala once a day. In addition, four of the reactors were fed in the same manner with increasing amounts of albumin to increase the ammonium and ammonia levels. Two of the reactors that were fed with albumin were further pulsed with acetate two times a day and should be regarded as duplicates. This was done to provoke acetate oxidation, making the expressed proteins of SAOBs more prominent than the background. The last biogas reactor was used as a control meaning it did not get any

additional albumin addition or acetate pulsing (Table 1).

The reactors were mesophilic with a temperature of 37 ℃ and was stirred continuously at a speed of 90 rpm. Before the project, the reactors had been handled in an identical way with stable conditions for a period of time (139 days). Thereafter, the albumin addition and pulsing of acetate was initiated.

Table 1 The conditions for the different biogas reactors used in the project.

Reactor C (control) A1 A2 B1 B2

Additional protein

No Albumin Albumin Albumin Albumin Albumin

Acetate pulsing No acetate pulsing No acetate pulsing No acetate pulsing Acetate pulsing Acetate pulsing

(21)

14

4.2 Sample collection and preparation

Samples were taken from the five reactors 2 hours after the first acetate pulsing of the day and right before feeding except for the samples taken at day 274 which were taken 40 minutes after pulsing. A majority of the samples, with an exception for the last two time points, were taken during a previous project and was kept frozen at – 20 ℃ until this project had begun. The first set of samples was taken at the beginning of the trial (day 141), at an ammonium concentration of 1.7 - 2.0 g/L depending on the reactor. Thereafter samples were taken each time the concentration had increased by approximately 1 g/L from the last time the samples were taken up to 6 g/L. For the samples taken at day 274 and 322 the ammonium

concentration is uncertain since no concentration measurements were done at these occasions. However, the last measured ammonium concentration is presented to give an idea of what it might be. In the table below (Table 2) the ammonium concentrations at which times the selected sludge samples were taken are presented.

Table 2 Ammonium concentration in the different reactors at different time points

Day of ammonium concentration measurements Day of sampling C (g/L) A1 (g/L) A2 (g/L) B1 (g/L) Pulsed B2 (g/L) Pulsed 141 141 1.9 1.8 1.7 1.8 2.0 148 148 1.8 2.7 2.7 2.7 2.8 158 158 1.9 3.6 3.6 3.9 3.5 162 162 1.9 4.1 4.1 3.7 4.2 247 274 2.0 3.5 3.5 4.7 4.5 305 322 3.0 6.0 6.0 6.1 5.9

4.3 Crude protein extraction and purification

4.3.1 Testing of protocols

Initially one protocol (Heyer, et al., 2016) for the extraction and purification of the proteins was tested several times. The samples that were used during the testing were taken from the biogas reactors the same day as the extraction started except for the first testing were the samples were taken from another set of reactors a couple of days beforehand.

200 µL of sludge in triplicates was transferred to a bead beating tube together with 400 µL 2 M sucrose and 700 µL phenol. The samples were cooled on ice, bead beat at 6.0 meter/second for 40 seconds and cooled again. Thereafter the tubes were centrifuged for 10 minutes at room temperature (RT) and 10 000 ×g. The upper phenol phase was transferred to a new Eppendorf tube with the same volume of 1 M sucrose added. The samples were shaken for 10 minutes at RT and 300 rpm and centrifuged as previous. The upper phenol phase of each of the

triplicates was collected and pooled in a 15 mL falcon tube. To precipitate the protein, four times the volume of the sample of 0.1 M ammonium acetate in methanol was added and the sample was incubated for 1 hour at -20 ℃. Thereafter it was centrifuged for 10 minutes at 4 ℃ and 10 000 × g and the supernatant was discarded.

(22)

15 Washing of the pellet was done by adding, to the pellet, three times the sample volume used for protein precipitation of ice- cold acetone. Thereafter the sample was incubated at -20 ℃ for 15 minutes followed by centrifugation for 10 minutes at 4 ℃ and 10 000 × g. Lastly the supernatant was discarded. This washing step was performed in a total of four times

alternating 80 % acetone and 70 % ethanol. After the final removal of the supernatant the pellet was air-dried and 2 mL urea/thiourea buffer (42.05 g Urea, 15.2 g thiourea, 1 g DTT and 100 mL water) was added. The sample was shaken until the pellet was completely solved. Protein concentration determination was made with an amido black assay except for the last attempt when it was made with a Lowry assay. Both assays were performed using bovine serum albumin (BSA) for generating a standard curve (as described in Appendix B).

Several variants of the protocol were tested due to a very low protein yield and no indication of proteins in the extracted and purified sample. This included centrifugation of the sample to make the sludge more concentrated (using 200 mg of the pellet), using a fresh phenol, using one sample instead of triplicates and switching out the urea buffer to 1 % SDS. The protein concentration determination was also altered, BSA solution used as a standard curve was remade and finally a Lowry assay was used instead of an amido black assay (Table 4). However, the results were still not satisfying (see 5.1.1) and after several trials another protocol (Thermofischer, 2016) was tested (as described in section 4.3.2) and the

concentration determination was instead made with the Lowry assay. The second protocol yielded satisfactory results and it was decided to move forward with this protocol instead, see 4.3.2 to 4.3.4.

When testing the second protocol the Lowry assay was used as concentration determination thereafter measurements with a nanodrop at 280 nm was made and a comparison between the two methods were made. Furthermore, it was decided to use the nanodrop for further protein concentration determination.

4.3.2 Final protein extraction protocol

Homogenization of the sludge sample was done by bead beating (ZR BashingBead lysis tubes, 0.1 and 0.5 mm ultra-high density beads, Zymo research) 200 µL sample and 1 mL of TRIzol (phenol containing reagent) together for 40 seconds at the speed 6 meter/second, the sample was cooled on ice before and after the procedure. After five minutes of incubation in RT, 200 µL of chloroform was added to the tube which was shaken by hand for

approximately 15 seconds to further extract the protein. The sample was yet again incubated in RT, this time for approximately 2 minutes. To separate the phases the sample was

centrifuged for 15 minutes at 4 ℃ and speed 12 000 × g. The top aqueous phase was

removed, and the remaining interphase and lower phenol-chloroform-phase was transferred to a new tube and mixed by inversion with 300 µL 100 % ethanol. After incubation at RT for two to three minutes the sample was centrifuged for 5 minutes at 4 ℃ and at a speed of 2000 × g to pellet the DNA. 500 µL of the supernatant was transferred to a new 2 mL Eppendorf tube containing 1.5 mL of isopropanol and the tube was inverted a few times to precipitate the protein. After incubation at RT for 10 minutes the protein was pelleted by centrifugation for 10 minutes at 4 ℃ and 12 000 × g, finally the supernatant was removed.

(23)

16

4.3.3 Washing of protein pellet

Washing of the pellet was done as similar to the original protocol and was done in four steps alternating 80 % acetone and 70 % ethanol. 1.5 mL ice-cold acetone or ethanol was added to the pellet and incubated at -20 ℃ for 15 minutes followed by centrifugation for 10 minutes at 4 ℃ and 10 000 × g. Thereafter the supernatant was discarded. After the final removal of the supernatant the pellet was left to air-dry for ten minutes without it not drying out completely. The pellet was resuspended in 200 µL of 1 % SDS by pipetting up and down in combination with vortexing. When the pellet was completely dissolved, remaining insoluble components left were pelleted by centrifugation according to the previous step. The supernatant was transferred to a new tube by pipette and stored at -20 ℃ until further use.

4.3.4 Protein concentration determination and verification

Determination of the protein concentration for the crude protein samples were done with nanodrop measuring at an absorbance at 280 nm.

SDS-PAGE was used to verify that the protein purification had been successful. 21 µL of SDS loading buffer and 20 µg of the sample was added to a tube and put on a thermo shaker at 1 400 rpm, 60 ℃ for 5 minutes. Thereafter a centrifugation for 10 min at RT and 16 400 was executed. The samples and a protein standard (BIO-RAD precision plus protein unstained standard

)

were applied to a stain free gel (BIO-RAD Mini-PROTEAN TGX stain free-gel) which were set to run at a current of 0.01 A for 30 minutes and then increased to a current of 0.02 A for 45 minutes. The image was developed with the stain free tray and the Gel Doc EZ Imager from Bio-Rad.

4.4 Protein digestion and peptide purification

The digestion and peptide purification was performed with the STrap protocol (Zougman, et al., 2014).

4.4.1 Testing of protocol

Before performing the STrap protocol, the trypsin activity was tested by mixing 50 µg of protein sample and 15 µL 33ng/µL trypsin in 50 mM ammonium bicarbonate. Thereafter followed an incubation step, initially for 1 hour at 47 ℃. Verification of the trypsin digestion were made with SDS-PAGE. Several tests were made; one test compared two different trypsin solutions and one compared with the intact protein sample. Finally, due to

unsatisfactory results, incubation factors such as time and temperature were compared, were a longer incubation time at lower temperature (20 hours, 37 ℃) and a shorter incubation time at higher temperature (2 hours, 47 ℃) were compared. The whole STrap protocol was also tested once (as described in section 4.4.2 to 4.4.4) to ensure the procedure would be practicable. Also, this time an SDS-PAGE was made to verify the results.

(24)

17 A final concentration for all of the solutions used in this step refer to the final concentration reached after the complete sample preparation. The final volume of the complete sample preparation is 100 µL.

For the sample preparation an amount of crude protein sample resulting in 1.67 µg/µL of protein at the final volume of 100 µL was transferred to a new tube (see Appendix C). Thereafter the sample was adjusted by adding 5 µL of Tris-HCl with a pH of 7.9 to reach a final concentration of 50 mM, 20 µL of SDS to reach a final concentration of 4% and 1 µL of dithiothreitol (DTT) to reach a final concentration of 10 mM. Water was added to a reach a volume of 81 µL.

The sample was incubated at 95 ℃ for 10 minutes and then cooled down to RT. 10 µL

Iodoacetamide (IAA) was added to a final concentration of 50 mM, the sample was incubated in dark at RT for 20 minutes. Acidification of the sample was made by adding phosphoric acid corresponding to 1/10th of the sample volume (9 µL). Finally, a confirmation of the acidity was made using pH paper.

4.4.3 STrap-tip preparation

The STrap-tip was made by stacking two layers of C18 membrane in the bottom and nine layers of MK360 filter paper (Munktell) on top of each other in a 200 µL pipette tip with the help of a blunt end needle, creating a column. Eppendorf tubes were prepared by puncturing of the lids with a clean scissor, creating a hole that fit the STrap-tip. The STrap-tip and Eppendorf tube were then assembled (Figure 6).

(25)

18

Figure 6 The assembly consisting of the complete STrap tip containing the MK360 filter on top and C18 membrane in the bottom, placed in an Eppendorf tube with punctured lid.

4.4.4 Trypsin digestion

170 µL of strapping solution (18 mL Methanol and 2 mL Tris (pH 7.5)) was added to each STrap-tip and 30 µL (i.e. 50 µg) of the prepared sample was gently added to the top of the strapping solution to form the protein suspension. Thereafter the assembly was centrifuged for 10 minutes at RT and 4000 × g to trap the protein suspension in the column.

To wash the column 100 µL of strapping solution was added to the tip and centrifuged as the previous step. Thereafter, 100 µL of 50 mM ammonium bicarbonate was added and

centrifuged through. The tip was then transferred to a clean tube and 15 µL of 33 ng/µL trypsin dissolved in 50 mM ammonium bicarbonate was added to the tip and gently spun into the column with a table centrifuge, leaving a couple of millimeters of solution left on the top of the membrane. The assembly was then incubated for two hours at 47 ℃ in an incubator to perform the digest.

After digestion, the solution left in the tip was fully spun through and 50 µL of 0.5 % trifluoroacetic acid (TFA) was added to the flow-through, the acidity was confirmed with a

MK360 C18

(26)

19 pH paper. The TFA containing flow-through was then transferred back to the tip and

centrifuged through to wash the column.

Thereafter, the column was further washed with 100 µL of 0.1 % TFA and the tip was transferred to a new tube. Finally, 100 µL of elution solution (1600 µL acetonitrile and 400 µL 0.5 % TFA (diluted with water)) was added to the tip and centrifuged for 30 seconds at 1000 × g, the samples rested for approximately 2 minutes and was then completely eluted by centrifugation for 5 minutes at 2500 × g.

The peptide samples of digested proteins were stored in the elution solution at -20 ℃.

4.4.5 Peptide concentration determination and verification

Peptide concentration determination was made with the Eon microplate spectrophotometer and the Take3 attachment and measured with UV at 205 nm.

For verification of the trypsin digestion 20 µL of the sample was transferred to a new tube and dried in the concentrator set to mode V-AL at 30 ℃ for 20 min. Thereafter 1 x loading buffer was added to the samples which were incubated at 95 ℃ for 5 minutes and thereafter

centrifuged for 10 minutes at RT and 16 400 × g. The samples, a protein sample control and the protein standard were loaded to a stain free gel which were set to run at a current of 0.01 A for 30 minutes and then increased to 0.02 A for 45 minutes. The image was developed with the stain free tray and the Gel Doc EZ Imager from Bio-Rad.

4.5 Mass spectrometry

To yield information about what peptides are present in the samples and which proteins they represent, the samples was analyzed with LC-MS/MS.

Because of time restraints and economical restraints, it was decided to not analyze all of the samples. The samples were selected according to the ammonium concentration (Table 3).

Table 3 Samples sent for mass spectrometry analysis. Reactor B1 and B2 are acetate pulsed.

Day of sampling Reactor (ammonium conc.) Reactor (ammonium conc.) Reactor (ammonium conc.) Reactor (ammonium conc.) 141 A1 (1.8) A2 (1.7) B1 (1.8) B2 (2.0) 158 A1 (3.6) A2 (3.6) B1 (3.9) B2 (3.5) 274 A1 (3.5) A2 (3.5) B1 (4.7) B2 (4.5) 322 A1 (6.0) A2 (6.0) B1 (6.1) B2 (5.9) C (3.0)

The LC-MS/MS analyses were performed with a LTQ Orbitrap XL (Thermofisher) using a four hour gradient per sample, executed by the University of Tartu, Estonia.

4.6 Mass spectrometry analysis of results

The mass data obtained from LC-MS/MS were analyzed with the MaxQuant software in a manner similar to an earlier published protocol (Tyanova, et al., 2016). The analysis was

(27)

20 made using the label free peptide quantification option. Due to time constraints it was decided to only analyze data for S. schinkii.

Most of the settings of MaxQuant were used in default mode, below is a list of what was added.

1. Configuration of the peptide database search engine (Andromeda) was made with the databases generated from the genome of the SAOB S. schinkii (Manzoor, et al., 2016) and the last option for “indicator parse role” was selected.

2. The raw data obtained from the LC-MS/MS measurements were loaded into MaxQuant and the “no fraction” option was selected.

3. Under “group specific parameters” and “modifications” “Deamidation (NQ)” was added and under “label free quantification” “LFQ” was selected.

4. Under “Global parameters” and “sequence” the FASTA file for S. Schinkii was loaded and under “advanced identification” “second peptide” was selected. Under “label free quantification” the options of “stabilize LFQ ratios”, “require MS/MS for LFQ comparison” “advanced site intensities” were selected.

Before the analysis started the number of processors were set to 39.

The results were then gathered and compiled with the Perseus software. The categorical columns “only identified by site”, “reverse” and “potential contaminants” were filtered out. Thereafter the values were transformed with the log2(x) function and the rows were filtered by setting the minimum valid values to 1. Thereafter a multi scatter plot was made. The values of the label free quantification intensity for each sample were used in excel to create a heat map which illustrate the protein expression level of selected proteins in the reactors during different ammonia concentrations.

(28)

21

5 Results

5.1 Protein extraction and purification

5.1.1 Initial testing and setting up of the protocol

The protein concentration from the testing of the two protocols ranged from 0.04 µg/µL to 14.09 µg/µl, depending on what method was used (Table 4). The samples that were used during the testing were taken from the biogas reactors the same day as the extraction started except for the first testing were the samples were taken from another set of reactors a couple of days beforehand. Hence, the samples were named based on at what time and reactor it was taken from.

For the first samples tested the protein concentration was very low and the SDS PAGE did not show any bands indicating that the extraction and purification was unsuccessful (Figure 7). Therefore, the same protocol was tested again, however the protein concentration was yet again low and no bands visible on the gel.

To investigate if the problem was that the sludge sample from the reactors contained very low protein concentrations the sludge was concentrated by centrifugation and the pellet was used for cell lysis and protein extraction. However, the protein concentration was still very low. Consequently, another phenol, which was fresh, was used in case the original phenol did not perform correctly. At the same time a longer bead beating was tested in case the protein was not properly extracted from the cells. Yet again the protein concentration was very low, and no bands were visible on the gel.

Lastly it was decided to use test 1 % SDS instead of urea buffer. Furthermore, a completely different protocol using the phenol-based solution TRIzol was tested at the same time. The protein concentration determination method was changed to Lowry assay instead of amido black assay in case this was a part of the problem with low protein concentrations. Both the first protocol and the second protocol yielded a decent protein concentration as well as visible protein bands on the SDS PAGE indicating that a successful recovery of the proteins had been made. Since the second protocol had a higher yield in ratio to used amount of sludge sample (first protocol used three times as much sample) it was decided to use the second protocol.

Table 4 Testing of two different protocols for protein extraction and purification. Illustrating sample, method used, protein concentration determination method used and protein

concentration.

Sample Protocol Method Protein concentration

determination method

Protein concentration (µg/µL)

X 1 Phenol, sludge, triplicates,

solved in urea/thiourea buffer

Amido black, standard curve BSA

1.55

Y 1 Phenol, sludge, triplicates,

1.40

1 1 Phenol, sludge, triplicates,

0.43

2 1 Phenol, sludge pellet,

triplicates, solved in urea/thiourea buffer

Amido black, standard curve BSA (new solution)

(29)

22

3 1 Phenol (fresh), sludge, one

sample, solved in urea/thiourea buffer

0.04

sample, solved in urea/thiourea buffer

0.04

sample, beadbeat x3, solved in urea/thiourea buffer

0.05

sample, beadbeat x3, solved in urea/thiourea buffer

0.06

5 1 Phenol (fresh), sludge,

triplicates, solved in 1% SDS

Lowry, standard curve BSA (new solution)

7.69

6 1 Phenol (fresh), sludge,

9.08

5 1 Phenol (fresh), pellet,

10.58

6 1 Phenol (fresh), pellet,

14.09

6 2 TRIzol, sludge, one sample,

solved in 1% SDS

6.62

6 2 TRIzol, sludge, one sample,

solved in 1% SDS

1.71

Samples were named based on at what time and which reactor it was taken from. Some samples were used for several trials. Protocol 1 used phenol and bead beating as

homogenization and the final protein pellet were resolubilized in urea/thiourea buffer except in the last trial were the pellet was resolubilized in 1% SDS. Two different phenol solutions were tested. The sample was done in triplicates or just one sample and the sludge was either used directly (sludge) or centrifuged (in that case the sludge pellet is used). Protocol 2 used TRIzol and beat beating as homogenization, it only used one sample (instead of triplicates) and the final protein pellet was resolubilized with 1% SDS. Two protein concentration determination methods were used; one amido black assay with a standard curve made with bovine serum albumin (BSA) for which two different solutions of BSA were used. The Lowry assay were also used with the new BSA used for standard curve.

The gels from the SDS-PAGE used for verification of the protein extraction and purification showed various results (Figure 7). The two first gels, representing extraction with the original protocol, did not show any protein bands on the gel. The third test which used sludge pellet for extraction showed a blur, but no bands were visible. Neither were no bands visible on the fourth gel where the extraction was performed with fresh phenol and longer bead beating. On the final gel, where the protein pellet was solved in 1% SDS and a new protocol was tested, there were individual protein bands discernable for all the samples regardless of how the sample was prepared.

(30)

23

Figure 7 The images of SDS-PAGE gels from testing of the protein extraction and purification methods. Sample number are the same as in Table 4. A) Samples tested with the original protocol without any modifications. B) Sample tested with original protocol without any modification. C) Sample where the sludge sample was concentrated by centrifugation and the pellet was used for cell lysis and protein extraction, also another stock solution for the BSA was used. D) Original protocol where completely fresh phenol was used, and no triplicates were used (only one sample). Two samples followed the protocol and two of the samples was bead beat in a total of three times. E) Original protocol (Ph) and the second protocol (Tr) was tested. The final samples from both the original and new protocol were resolubilized in 1% SDS. Two of the samples made with the original protocol were samples directly from the reactor and two were centrifuged (C) and the protein extraction was made from the pellet.

Comparison of protein concentration determination with Lowry assay and Nanodrop, respectively, were made on the first set of samples that were prepared (samples from day 274). The different methods agreed with each other for the majority of samples (Table 5). There were two Nanodrop measurements that deviated. However, the corresponding dilution of the deviated sample agreed somewhat to the Lowry measurements and therefore it was decided to further continue the protein concentration measurements with the Nanodrop.

A B C D E 31x 33x 41x 43x 5Ph 5Ph C 6Ph 6Ph C 6Tr 6Tr X Y 1 2 250 150 100 75 50 37 25 20 15 10 kD

(31)

24

Table 5 The protein concentrations for a set of samples yielded with two different methods: Lowry assay and Nanodrop.

Method used C (µg/µL) A1 (µg/µL) A2 (µg/µL) B1 (µg/µL) B2 (µg/µL) Lowry 6.3 5.6 6.3 6.6 8.7 Nanodrop (1:5) 1.8 5.8 5.8 5.8 9.3 Nanodrop (1:10) 6.6 6.8 5.4 11.4 7.4

Samples used for nanodrop were diluted 5 respectively 10 times. The concentration of the undiluted samples are shown.

5.1.2 Crude Protein purification from reactor samples

By using the second protocol set up in 4.3.2 crude protein could be successfully purified from all samples. The obtained protein concentration of the reactor samples ranged from 3.2 µg/µL to 8.7 µg/µL (Table 6Fel! Hittar inte referenskälla.).

The SDS gels for validation of the crude protein extraction and purification revealed that there were several bands present in the gel lanes loaded with sample (Figure 8) indicating good quality of the crude protein extract. Five of the gels had a too high concentration of loading buffer which made the bands appear blurry.

Table 6 Protein concentration of the crude protein samples from the different biogas reactors and time points, determined with nanodrop at an absorbance of 280 nm.

Day C (µg/µL) A1 (µg/µL) A2 (µg/µL) B1 (µg/µL) Pulsed B2 (µg/µL) Pulsed 141 4.1 4.3 3.8 3.5 4.1 148 7.9 7.4 5.2 5.0 4.2 158 3.5 3.8 3.2 4.0 4.0 162 6.2 7.0 5.0 5.1 5.7 274 6.3 5.6 6.3 6.6 8.6 322 5.2 5.4 5.0 5.1 5.2

(32)

25

Figure 8 Gels from SDS-PAGE after protein extraction and purification. C – control, A1, A2 – additional albumin, B1, B2 – additional albumin and acetate pulsing. The date on each

Day 141 Day 148 Day 158 Day 162 Day 274 Day 322 A1 C A2 B1 B2 A1 C A2 B1 B2 A1 C A2 B1 B2 A1 C A2 B1 B2 A1 C A2 B1 B2 _C _A2 _{B2 B1}Day 141 A1 B1 kD 250 150 100 75 50 37 25 20 15 10

(33)

26

picture is the date when the samples were taken and corresponds to the ammonium levels (see Table 2). On the gel for samples on day 322 sample B1 from day 141 is also present since more sample was needed for further applications.

5.2 Protein digestion and peptide purification

5.2.1 Initial trypsin and protocol testing

In order to optimize the protocol and the incubation factors for trypsin there were a couple of tests performed with the trypsin and the STrap protocol.

First, two different trypsin solutions were tested, one old and one new, it was discovered that the older solution had lost its cleavage capabilities (Figure 9A). Consequently, the solution that did work better was compared to undigested samples (Figure 9B). Since individual bands were still present in the digested samples, indicating that the sample was not completely digested, longer incubation time at a bit lower temperature was tested to see if that would impact the digestion. These tests also resulted in visible bands on the gel (Figure 9C). However, there was a visible difference between the digested samples and the undigested samples suggesting that the bands could belong to trypsin. It should also be noted that one of the samples got a considerably less amount of trypsin than the other samples. Since there were no big difference in the gel between the incubations factor it was decided to use 2 hours at 47 ℃ since it was more time efficient.

The gel from the execution of the complete STrap protocol (Figure 10), which removes trypsin from the sample. Showed no visible protein bands and a big collection of peptides at the bottom of the gel in the lanes with digested peptides, indicating complete and successful digestion.

(34)

27

Figure 9 Images of gel from testing of trypsin activity and protein digestion. A) comparison of two trypsin solutions, one prepared recently (n) and one prepared two years ago (o). A

sample with only water and the freshly prepared trypsin is also present. All these samples were incubated at 47 ℃ for one hour. B) Two different samples with the new trypsin solution (t) and the same two samples without trypsin. Also here is a sample with only water and trypsin present. Incubation is at 47 ℃ and one hour C) Two samples with trypsin incubated at 37 ℃ for 20 hours (tA120, tB120) and two samples with no trypsin at the same incubation

conditions (A120, B120).The same two samples were also incubated with trypsin at 47 ℃ for

two hours (tA12, TB12).

C B tC tB2 C B2 tH₂O tA1₂₀ tB1₂₀ A1₂₀ B1₂₀ tA1₂ tB1₂ oB2 oA2 oA1 nH₂O nB2 nA2 nA1 A kD 250 150 100 75 50 37 25 20 15 10

(35)

28

Figure 10 Image of an SDS-gel with samples that have been digested and purified with the complete STrap protocol (t) with an incubation for Trypsin at 47 ℃ for 2 hours. The same sample but undigested is also shown.

Since the STrap protocol gave good results in the gel no adjustment other than prolonging incubation time to 2 hours at 47 ℃ were made.

5.2.2 Peptide purification from the crude protein extracts

The obtained peptide concentrations varied a lot depending on the sample and ranged from 0.04 µg/µL to 0.49 µg/µL (Table 7). The samples with lower concentrations (approximately 0.2 µg/µL and below, samples A1 and B1 taken at day 141) was decided to be redone for LC-MS/MS measurements to obtain a higher concentration to make sure there was a margin for enough sample to use for the LC-MS/MS measurements. However, for the latest prepared set of samples (day 322) there were time constraints meaning that they had to be used anyway.

Table 7 The peptide concentration for the digested samples.

Day C (µg/µL) A1 (µg/µL) A2 (µg/µL) B1 (µg/µL) B2 (µg/µL) 141 0.18 0.06/0.11 0.20 0.13/0.09 0.24 148 0.03 0.15 0.09 0.08 0.20 158 0.49 0.44 0.42 0.40 0.38 162 0.35 0.34 0.34 0.29 0.24 274 0.3 0.44 0.36 0.36 0.26 322 0.07 0.13 0.11 0.10 0.04

Highlighted samples are the samples that were sent for mass spectrometry measurements. Samples that display two values were digested two different times and the version with the highest concentration was used for the mass spectrometry measurements.

The SDS-PAGE made after trypsin digestion showed that there were clear collections of peptides from digestion of the protein in the bottom of most of the lanes and no protein bands were visible (Figure 11). This indicates that digestion was succesfully performed and peptides were recovered. tA1 A1 tC C tA2 A2 tB2 B2 kD 250 150 100 75 50 37 25 20 15 10

(36)

29

Figure 11 Images of SDS gels with trypsin digested samples. The sample to the left (cA1) is a control with the crude protein sample for sample A1 which was left out for the samples of day

Day 141 Day148 Day 158 Day 162 Day 274 Day 322 cA1 cA1 cA1 cA1 cA1 A1 C A2 B1 B2 A1 C A2 B1 B2 A1 C A2 B1 B2 A1 C A2 B1 B2 A1 _C B1 _B2 A1Day141 B1Day141 A2 B2 B1 A2 C A1 kD 250 150 100 75 50 37 25 20 15 10

(37)

30

322 by mistake. In the gel for samples from day 322 two samples from day 141 was present to verify the peptide sample since more sample was needed for further applications.

5.3 Mass spectrometry data

The LC-MS/MS spectra had a similar layout for all samples where the majority of the larger peaks were eluted and mass determined late in the experiment (Figure 12), indicating

contamination with nucleic acids and polysaccharides. During the analysis it was also noted that the peptide concentration for the samples were overestimated with a factor of 1.5 and that two of the samples did not contain any peptides.

Figure 12 LC-MS/MS spectrum for sample A1, day 158.

The peptide profile in the samples were compared to each other with the help of a multi-scatter plot (Figure 13) with the data yielded from LC-MS/MS measurements and further processing of them in MaxQuant. The multi-scatter plot showed that there were some samples that were empty or containing few peptides. However, there were multiple samples showing multiple peptide peaks as well as samples that was indicated to have similar protein content to each other due to the linear formation of the dots.

(38)

31

Figure 13 Multi-scatter plot of the protein content in the samples compared to each other. The order of the samples is sorted to date and starts with were the lowest ammonium concentration were detected. Reactor order is A1, A2 (additional albumin) B1, B2 (acetate pulsing). The last sample is the control (C) for the last set of samples. The samples are ordered in the same way along both axes.

A heat map was made with the data acquired from the multi-scatter plot, illustrating expressed proteins recovered for each sample (Table 8, Table 9 and Appendix E). It was clear that some proteins were only expressed at lower or higher ammonium concentrations but there were also proteins that were seen expressed at several ammonium concentrations.

A1 A2 B1 B2 A1 A2 B1 B2 A1 A2 B1 B2 A1 A2 B1 B2 C Day 141 Day 158 Day 274 Day 322

A1 A1 A1 A1 C A2 A2 A2 A2 B1 B1 B1 B1 B2 B2 B2 B2 Da y 1 5 8 Da y 1 4 1 Da y 2 7 4 Da y 3 2 2

Investigating the methane producing pathway in lab-scale biogas reactors subjected to sequential increase of ammonium and daily acetate-pulsing