Protein and peptide databases enabling rapid monitoring and quantification of microbes and conversions from enrichment, or mixed culture, production systems, and other microbial consortia

FIELD OF THE INVENTION

The present invention relates to a method of monitoring a microbiome, a method of controlling a reactor comprising a microbiome, or a method of determining an effect of a medicament or drug in an environment comprising a microbiome, wherein in both cases said microbiome is monitored according to said method, and to a microbiome monitoring com puter program comprising instructions for monitoring a microbiome, which methods are ef ficient, relatively quick, and relatively cheap.

BACKGROUND OF THE INVENTION

The present invention is in the field of a method of monitoring a microbiome. A mi crobiome may be considered to relate to a characteristic microbial community occupying a reasonably well-defined habitat which has distinct physio-chemical properties. The term thus not only refers to the microorganisms involved but typically also encompasses their theatre of activity. The microbiome may be defined as a characteristic microbial community occu pying a reasonably well-defined habitat which has distinct physio-chemical properties. The microbiome not only refers to the microorganisms involved but also encompass their theatre of activity, which results in the formation of specific ecological niches. The microbiome, which forms a dynamic and interactive micro-ecosystem prone to change in time and scale, is integrated in macro-ecosystems including eukaryotic hosts, and here crucial for their func tioning and health. It is noted that the term “microbiota” is separated from the term microbi ome in that the microbiota is considered to consists of the assembly of microorganisms be longing to different kingdoms (Prokaryotes [Bacteria, Archaea], Eukaryotes [e.g., Protozoa, Fungi, and Algae]), while their theatre of activity includes microbial structures, metabolites, mobile genetic elements (such as transposons, phages, and viruses), and relic DNA embed ded in the environmental conditions of the habitat (see https://en.wikipedia.org/wiki/Microbiome). The present invention is however more related to the microbial community than to the habitat. The habitat of the microbial community is not ed and taken into account, but the present invention relates more to the characterization of the microbial community.

A microbiome is a complex community of microbial species. Not only a large num ber of different species may occur within one microbiome, but also quantities of the species may vary among species. Often a limited number of microbiome species are dominant in mass, but still such may relate to a significant number of species. In particular in microbial production methods, when for instance food is produced, a more limited number of species is typical present. In that respect attention may be paid to sterilize the habitat of the microbi ome before production, therewith inherently limiting a variation within the microbiome. Al- 2 so the species in the microbiome may vary over time, not only in abundance as in biomass thereof, but also some may become negligible whereas others may become abundant. A sim ilar variation in terms of microbiome strains may occur, adding to the complexity. Even ge netic changes may occur. So the characteristics of the microbiome in this respect may vary over time.

It is noted that mixed or enriched microbial cultures are considered promising pro duction systems for biotechnology and pharmaceutical industry. Microbial communities ex hibit metabolic capabilities which may be unique or superior compared to pure cultures. Fur thermore, many microbial communities live in close relation to humans, or other hosts, and therefore may directly impact on human health and well-being. State of the art measurements of the composition or metabolic potential of microbial communities typically relay on stain ing, or genetic tools, which are unspecific, or generate very large data, and are very time consuming. Furthermore, the relevant information of the actually expressed bio mass/metabolic composition is not obtained thereby, but only information relating to species being potentially present per se.

For obtaining biomass mass protein information spectrometry-based community pro- teomics (metaproteomics) may be used, which technique measures the proteins directly. Un fortunately, common metaproteomic approaches are also very time consuming, and require very high resolution, expensive instrumentation and advanced bioinformatics tools and knowledge for interpretation, hence are highly complex So currently there are no suitable high throughput approaches which can monitor, and control mixed microbial communities (production systems, or natural consortia) at the biomass/metabolic level (proteins, enzymes) on a routine basis. As a consequence, mixed cultures are commonly used and considered only as a black box, which inherently is difficult to control and to operate. Furthermore, some applications, such as medical applications in relation to the gut microbiome, are hampered since monitoring techniques lack fast and spe cific methods.

Some scientific articles, rather incidentally, relate to metaproteomics mainly. Mikan et al. in “Metaproteomics reveal that https:/doi.org/10.1038/s41396-019-0503-z, reveal that rapid perturbations in organic matter prioritize functional restructuring over taxonomy in western Arctic Ocean microbiomes during 10 days. They used a novel peptide-based en richment analysis and observed significant changes (p-value < 0.01) in biological and molec ular functions associated with carbon and nitrogen recycling. It is noted that (me- ta)proteomics always uses peptides. The novel enrichment analysis mentioned indicates that they looked for peptide signals that are more abundant in one measurement compared to an other one. The method used may be considered as a discovery-based peptide-centric ap proach. The us of a DNA library is a common procedure in metaproteomics. Here it is used as a database for metaproteomics (in order to obtain amino acid sequences, taxonomy and function). The article is mainly concerned with a data analysis advancement to better get 3 community functions from metaproteomics data. Franzosa et al, in “Sequencing and beyond: integrating molecular 'omics' for microbial community profiling”, is a review article mainly concerned with advances in DNA sequencing have enabled culture-independent profiling of microbial community membership and function — the field of metagenomics. These ap proaches have rapidly expanded our knowledge of human-associated and environmental mi- crobiomes (https:/doi:10.1038/nrmicro3451). Li et al. in “RapidAIM: a culture- and metap- roteomics-based Rapid Assay of Individual Microbiome responses to drugs” developed an approach to screen compounds against individual microbiomes in vitro, using metaprote omics to both measure absolute bacterial abundances and to functionally profile the microbi ome (doi:10.1186/S40168-020-00806-Z). Calusinska et al. in “A year of monitoring 20 mes- ophilic full-scale bioreactors reveals the existence of stable but different core microbiomes in bio-waste and wastewater anaerobic digestion systems”, used high-throughput sequencing of small rRNA gene, and a monthly monitoring of the physicochemical parameters for 20 dif ferent mesophilic full-scale bioreactors over 1 year, to generate a detailed view of AD mi crobial ecology towards a better understanding of factors that influence and shape these communities (DOI: 10.1186/s 13068-018- 1195-8).

The present invention therefore relates to a method of monitoring a microbiome and further aspects thereof, which overcomes one or more of the above disadvantages, without compromising functionality and advantages.

SUMMARY OF THE INVENTION

It is an object of the invention to overcome one or more limitations of the methods of monitoring a microbiome of the prior art and at the very least to provide an alternative there to. The present invention relates to a new method of monitoring a microbiome, over time, comprising providing the microbiome, the microbiome comprising a population having a population biomass, the population comprising a variety of microbial species and/or a varie ty of microbial strains (a genetic variant, a subtype or a culture within a biological species) wherein each microbial species and/or microbial strain individually provides a species bio mass or strain biomass to the population, wherein microbial species and microbial strains are in particular selected from Archaea, Bacteria, Eukaryote, Algae, Fungi and small protists,

(1) within the microbiome characterizing at least 50 wt.% of population biomass in terms of the biological taxonomy, in particular at least 5 levels of taxonomy, and typically 7-9 levels of taxonomy, more in particular the taxonomic levels being selected from Kingdom, Phylum, Class, Order, Family, Genus, Species, and if possible from strain, and sub-strain, microbial species and/or strains being present in the population biomass, in particular characterizing at least 70 wt.% of said population biomass, more in particular characterizing at least 85 wt.% of said population biomass, even more in particular characterizing at least 95 wt.% of said population biomass, (2a) at least two times extracting protein sequences from said popula tion biomass, the two extractions forming a time sequence, which extractions are typically directly made from the biomass, (2d) directly or indirectly determining an amount of at least 4 one extracted protein sequence, wherein the amount may be relative to a total amount or ab solute in terms of (bio)mass, and typically said amount is determined using chemical analyt ics, and (2f comparing in the time sequence a later amount of extracted protein sequence with an earlier amount of protein sequence, such as comparing a last amount of extracted protein sequence with a first amount of extracted protein sequence. In particular the present method relates to a method of monitoring a microbiome which comprises (2c 1) selecting a sub-population of the microbial species, wherein said selection is typically done with a com puter/dedicated software, (2c2) determining a sub-set of protein sequences and/or peptides representing the sub-population, or metabolic functions of the microbiome, and after step (2d) (2e) directly or indirectly analysing the amounts of extracted protein sequences and/or peptides of the sub-set by comparing the sub-set against a database comprising protein se quences and/or peptides of the sub-set, and determining biomass per microbial species of the sub-population, in particular biomass proteinous contribution per microbial species. In step (2cl) Currently selection is done empirically, e.g. looking for sequences at a selected taxo nomic ranking (at least genus), and/or peptides that are good to measure/quantify (physico chemical properties), and/or have uniform proportions, representing the community, or gene ra of interest. Of course, machine learning may be used in an alternative, in particular once inventors have collected data over many plants, over a longer period of time. Typically a cut off is used at a lower end of the percentage/amount obtained, such as at 1% or 0.1%. The invention therewith also relates to i) a specifically designed algorithm which extracts system relevant protein/peptide information from metagenomics and proteomics data of the particu lar enrichment culture or community. The relevant information thereto are proteins/peptides unique for taxonomic rankings, and/or responsible for certain (metabolic) conversions. The invention therewith also relates to ii) thereby generated protein/peptide database containing the system relevant information for a particular stage of the system (e.g. for the core micro- biome/function of activated granular sludge water treatment system at different operational stages, gut microbiome at different health stages, or a mixed microbial production system at different productivities). These protein/peptide databases are found to enable a rapid and quantitative monitoring of mixed microbial cultures, by means of highly simplified instru mentation for measurement (such as low resolution mass spectrometers) and highly simpli fied software for data analysis. Furthermore, this is found to enable multiplexed sample pro cessing, measurements and high-throughput data processing. Thereby, complete analysis times can easily be reduced to within one working day. With the specifically designed soft ware which extracts information/data from metagenomics and metaproteomics experiments a database with reduced complexity is provided. It is found that the microbial (enrichment) culture peptides/protein database, represents a specific condition of the communi ty/enrichment. The present method provides rapid and focused quantitative monitoring of mixed cultures using e.g. low resolution mass spectrometric approaches. So, by approaching the enormous complexity of microbial consortia and enrichment cultures in a stepwise ap- 5 proach, previously very difficult to monitor and to control systems can now be monitored adequately. Therewith relevant information of the actually expressed biomass/metabolic composition is obtained. It is found that e.g. mass spectrometry based community prote- omics (metaproteomics), which measures the proteins directly, overcomes the limitations. Therewith one can monitor, and control/interpret mixed microbial communities (production systems, or natural consortia) at the biomass/metabolic level (proteins, enzymes). The mixed cultures are no longer a black box. Also for other applications, such as medical applications, e.g. in relation to the gut microbiome, are now provided with fast and specific methods. The invention overcomes the above problems, by reducing the complexity to a level, which ena bles rapid, quantitative and low spec instrumentation/data interpretation monitoring. In case a genome or protein/peptide could not be mapped, it was attributed to a less specific, that is higher taxonomic, level. Typically a genome c.q. peptide/protein sequence could be annotat ed to more than one close-by member, for a given taxonomic level; then the closest hit was typically chosen, or in an alternative a higher taxonomic level. In the end a majority of pos sible peptides/protein sequences, which could have been possible on genomic data, were not found in practice, and a more limited set could be used. Moreover, not all peptides are con sidered suitable for quantitative analysis, therefore peptides can be ranked to not only pro vide the representative subset of the community, but also the most the most suitable subset from an analytical viewpoint.

In a second aspect the present invention relates to a method of controlling a reactor comprising a microbiome, comprising monitoring a microbiome according to the invention, and based on said microbiome, or changes therein, adapting at least one parameter selected from temperature, flow, pH, static residence time, solid retention time, nitrogen content, phosphorous content, amount of biomass, amount of nutrients, oxygen content, flow, alkalin ity content, fatty acids content, redox values, feed flux, production installation of input sludge, method of production of input sludge, age of input sludge, organic carbon content COD of input sludge, method of production of input sludge, dosing of chemicals during pro duction of input sludge, remaining concentration of dosing chemicals left, process setting during production of input sludge, polyelectrolyte concentration, type of polyelectrolyte, bowl speed, pressure applied to the sludge, gas produced, stir rate, ammonium concentration in an effluent stream, concentration of protein sequences, concentration of sugars, concentra tion of cellulosic material, amount of degradable organic matter, cation concentration, differ ential speed, trace elements, in particular oxygen content, or not adapting a parameter, or stopping operation of the reactor. The adaptation may be provided as such, or may comprise a feedback loop, in which, based on said microbiome, adaptation is performed.

In a third aspect the present invention relates to a method of determining an effect of a medicament or drug, or of an purposive action, or change of habit, such as when changing a diet, in an environment comprising a microbiome, comprising monitoring a microbiome ac cording to the invention, and adapting an amount of medication or drug, and/or adapting an 6 administration regime of medication or drug, and/or changing purposive action or habit.

In a fourth aspect the present invention relates to a microbiome monitoring computer program comprising instructions for monitoring a microbiome according to the invention or for operating a reactor according to the invention or for determining an effect of a medica ment or drug in an environment comprising a microbiome according to the invention, the instructions causing the computer (1) to carry out the following steps: (1) characterizing at least 50% of the microbial species being present in the population, (2d) directly or indirectly determining an amount per extracted protein sequence, and (2f) comparing in the time se quence a later amount of extracted protein sequence with an earlier amount of protein se quence, such as comparing a last amount of extracted protein sequence with a first amount of extracted protein sequence.

In a firth aspect the present invention relates to a microbiome monitoring computer program comprising instructions for monitoring a microbiome according to the invention or for operating a reactor according to the invention or for determining an effect of a medica ment or drug in an environment comprising a microbiome according to the invention, where in the computer program comprises instructions for learning, such as machine learning, adaptive learning, and combinations thereof.

The present invention provides a solution to one or more of the above mentioned problems and overcomes drawbacks of the prior art.

Advantages of the present description are detailed throughout the description. DETAILED DESCRIPTION OF THE INVENTION

In an exemplary embodiment of the present method of monitoring a microbiome char acterizing the microbial species in step (1) is a qualitative characterization.

In an exemplary embodiment of the present method of monitoring a microbiome the characterization in step (1) is performed using a genetic sequence of a species and matching said genetic sequence with a genetic sequence -database comprising genetic sequence -data of possible microbial species, such as by genomics or metagenomics, in particular wherein said genetic sequence is a DNA sequence, a RNA sequence, a gene sequence, an enzyme sequence, or part thereof, or combination thereof.

In an exemplary embodiment of the present method of monitoring a microbiome be fore determining an amount per extracted protein sequence (2b) the extracted protein se quences are cleaved into peptide fragments.

In an exemplary embodiment of the present method of monitoring a microbiome the extracted protein sequences are cleaved into peptide fragments each individually comprising 6-100 amino acids, preferably 7-75 amino acids, in particular 8-55 amino acids, such as 10- 12 amino acids.

In an exemplary embodiment of the present method of monitoring a microbiome cleavage is obtained by at least one specific protease, preferably a protease of mixed nucleo philic superfamily A, preferably a serine protease, in particular a chymotrypsin-like protease 7 or a suhtili in-like protease, such as Trypsin /CAS 9002-07-7 ).

In an exemplary embodiment of the present method of monitoring a microbiome (2d) determining an amount of protein sequence directly is performed using high resolution mass spectrometry or wherein (2d) determining an amount of protein sequence indirectly is per formed using high resolution mass spectrometry on peptides and/or proteins.

In an exemplary embodiment of the present method of monitoring a microbiome the step of (1) characterizing at least 50 wt.% of the species being present in the population is performed only once, such as in a well-defined condition, e.g. a reactor performing as ex pected.

In an exemplary embodiment of the present method of monitoring a microbiome the step of (2c 1) selecting a sub-population of the microbial species is performed only once.

In an exemplary embodiment of the present method of monitoring a microbiome the step of (2c2) determining a sub- set of protein sequences representing the sub-population is performed only once.

In an exemplary embodiment of the present method of monitoring a microbiome in step (2c2) protein sequences and/or peptides are selected which represent at least one high taxonomic level, in particular an Order level, a Family level, or Genus level, or represent at least one metabolic pathway present in a variety of species and/or strains, and/or [(Every species can contribute with approximately 2,5K protein sequences and every protein se quence can give approx. 10-50 peptide fragments, so the theoretical numbers get very large)].

In an exemplary embodiment of the present method of monitoring a microbiome in step (2d) determining an amount of at least one extracted protein sequence relates to a rela tive amount or to an absolute amount. The amount may be determined in weight terms (e.g. mg), but it is found more practical to use a relative amount, or abundance, or likewise peak area in a chromatogram. In a further step these may be compared mutually, or compared to the total protein/peptide signal, or even being relative to an amount of injected peptides, which may be considered to function as a calibration.

In an exemplary embodiment of the present method of monitoring a microbiome a cal ibration is provided.

In an exemplary embodiment of the present method of monitoring, a protein and/or peptide database of the present microbiome is generated.

In an exemplary embodiment of the present method of monitoring a sub- set of the pro tein and/or peptide database of the present microbiome is created.

In an exemplary embodiment of the present method of monitoring a microbiome a relative amount is relative to an earlier determined amount or relative to an amount of at least one other protein sequence.

In an exemplary embodiment of the present method of monitoring a microbiome in step (2d) an amount of at least one most abundant extracted protein sequence is determined. 8

In an exemplary embodiment of the present method of monitoring a microbiome in step (2d) an amount of at least one most characterizing extracted protein sequence is deter mined, in particular an extracted protein sequence with the most linear quadratic estimate weight.

In an exemplary embodiment of the present method of monitoring a microbiome the steps of (2a) at least two times extracting protein sequences from said population are per formed as often as required for process control, such as based on statistical process control, or based on out of range control, or a combination thereof.

In an exemplary embodiment of the present method of monitoring a microbiome in step (1) at least once an amplification technique is used, such as PCR.

In an exemplary embodiment of the present method of monitoring a microbiome the population comprises 10 ¹ - 10 ⁷ different species, in particular 2*10 ¹ -10 ⁶ different species, more in particular 10 ² -10 ⁵ different species.

In an exemplary embodiment of the present method of monitoring a microbiome the sub-population comprises 2-10 ⁵ of the different species of the population (0.001-10%), in particular 4-10 ³ of the different species of the population (0.1-1%), more in particular 6-10 ² different species, such as 7-20 different species.

In an exemplary embodiment of the present method of monitoring a microbiome 10 ² - 10 ^s of different protein sequences are extracted, in particular 10 ³ -10 ⁴ of different protein sequences.

In an exemplary embodiment of the present method of monitoring a microbiome 10 ¹ - 10 ⁴ of different peptide fragments are formed, in particular 10 ² -10 ³ of different peptide fragments.

In an exemplary embodiment of the present method of controlling a reactor comprising a microbiome the reactor is selected from a digestion reactor, a continuous stirred tank reac tor, a batch reactor, a repeated batch reactor, a sequence batch reactor, a single reactor with segmented sub-reactors, a plug flow reactor, a post-digestion reactor, a dewatering device, and combinations thereof.

In an exemplary embodiment of the present method of controlling a reactor comprising a microbiome the reactor comprises wastewater, such as wastewater from housings, from industry, from hospitals, from facilities in general, or a food comprising a mixed microbial population, such as bear, wine, a dairy product, such as yoghurt, or cheese, or a fermentation product, or a digestion product, or an enrichment product, or a microbial consortium product.

In an exemplary embodiment of the present method of controlling a reactor comprising a microbiome monitoring is performed 1-168 times per week.

In an exemplary embodiment of the present method of determining an effect of a me dicament or drug in an environment comprising a microbiome the microbiome is selected from a mammal, such as a gastro-intestinal microbiome, a skin microbiome, an oral microbi ome, a rectal microbiome, a genital tract microbiome, and an urinary microbiome. 9

In an exemplary embodiment the present microbiome monitoring computer program may further comprise instructions for storing microbiome data, in particular for characteriz ing a microbial species, a genetic sequence of said microbial species, a protein sequence produced or present of said microbial species, or a peptide fragment produced or present of said microbial species,

The invention will hereafter be further elucidated through the following examples which are exemplary and explanatory of nature and are not intended to be considered limit ing of the invention. To the person skilled in the art it may be clear that many variants, being obvious or not, may be conceivable falling within the scope of protection, defined by the present claims.

FIGURES

Figure la,b, 2a, b and 3 show details of the present invention.

DETAILED DESCRIPTION OF FIGURES

In the figures:

Figure 1 shows top phylum levels for WWTP-I (fig. lb) and WWTP-II (fig. la) as established from the granules by metaproteomics. The microbiomes are comparable, which may be expected in view of both examples relating to a WWTP, and at the same time results show clear differences in abundances. These differences may be attributed to different per formances of the respective plants considering their different location, wastewater and possi bly different operation.

Fig. 2 shows top 75% genera compromising 75% of peptide peak areas for WWTP-I (fig. 2b) and WWTP-II (fig. 2a) as established from the granules by metaproteomics. The microbiomes are again comparable but again show clear differences in the abundance of individual members. These differences may be attributed to different performances of the respective plants considering their different location, wastewater and possibly different oper ation.

Fig. 3 shows the general process for the peptide subset selection procedure used for routine monitoring of the microbial community. Given numbers are based on the example from the WWTP-I. Initial metagenomics and metaproteomics experiments identify the (commonly) observed peptide sequences, which are further narrowed down by selecting for taxon informative peptides (e.g. genus or species level) or metabolic function informative peptides. The informative peptides are further filtered for the methodologically most suitable sequences, equally representing the microbiome, or specific community members. EXPERIMENTS

First of all a number of operating reactors were identified. On series of reactors relate to so-called Nereda reactors These reactors produce bacterial sludge. Aerobic granular sludge and anammox granular sludge, and the processes used for obtaining them are known to a person skilled in the art. For the uninitiated, reference is made to Water Research, 2007, 10 doi:10.1016/j.watres.2007.03.044 (anammox granular sludge) and Water Science and Tech nology, 2007, 55(8-9), 75-81 (aerobic granular sludge), as well as previous applications in this respect from the same applicant. From these reactors microbiomes were retrieved. These microbiomes were characterized and monitored according to the invention. Metadata or raw data derived from these Nereda plants relates to:

1. metagenomics data from 3 Nereda waste water treatment plants;

2. conventional metaproteomics data from the same 3 Nerada plants;

3. a peptide database which contains all possible peptides (exemplified from one Nereda plant);

4. a peptide database which contains a subset of observed peptides (exemplified for one Nereda plant);

5. a peptide database which contains a selected subset used for further monitor ing (from the database described directly above).

In order to monitor these plants over time, Nereda treatment plant samples, from dif ferent locations, were obtained. Including those which are found to operate sub-optimal. The 2 plants look comparable in regards to the microbiome. It is considered that these similarity is understood that the core functions operate in a similar way.

The following section relates to raw data from a to be submitted scientific publication of H.B.C. Kleikamp et al, entitled “A deep comparative metaproteomic investigation of the core microbiome of aerobic granular sludge”, which publication and its contents are incorpo rated by reference.

Here inventors compare metagenomics and metaproteomics based microbiome anal ysis for three full-scale aerobic granular sludge wastewater treatment plants (Nereda™ tech nology). To enable rapid metaproteomic sampling at reduced cost, further improvements are made to the existing proteomics pipeline by using habitat specific databases and the investi gation for taxon and metabolic function relevant peptides. Differences in observed taxonom ic and functional distributions for the successfully operating Dutch AGS plants using 16S, metagenomic and metaproteomic analysis is discussed. This gives a core microbiome of aer obic granular sludge systems. The evaluation shows that proteomics is more suitable than genomics for the biomass characterisation of the plants.

Experiments

SAMPLING, PROTEIN EXTRACTION AND PROTEOLYTIC DIGESTION.

Activated granular sludge was sampled from two of the above waste water treatment plants (WWTP). Granules (approx. 2.0 mm diameter) were freeze dried and grinded using a mortar and pestle and further subjected to beads beating using glass beads in a TEAB/B-PER buffer. Following an additional freeze/thaw step and incubation at elevated temperature (95° C) for a short time period, the tubes were cooled and centrifuged at full speed using a bench top centrifuge for 10 minutes. The supernatant was collected and protein was precipi tated using trichloroacetic acid (TCA). Following a short cooling the solution was centri- 11 fuged at full speed using a bench top centrifuge to collect the protein pellet. The protein pel let was washed once with ice cold acetone and reconstituted in 6M Urea (aiming for a pro tein concentration of lpg/pL), reduced using Dithiothreitol (DTT) and alkylated using Iodo- acetamide (IAA). The protein solution was finally diluted to below 1M urea using 200mM bicarbonate buffer, before addition of sequencing grade trypsin at a trypsi protein ratio of approx. 1:50. Samples were digested at 37°C over-night. Obtained peptides were desalted using an Oasis HLB SPE well plate (Waters), according to the manufacturers protocol. The eluate was speed vacuum dried and resolubilised in 0.1% TFA solution for further prefrac tionation using a high pH reverse phase peptide fractionation kit (Thermo) according to the protocol provided by the manufacturer. Fractions were speed-vacuum dried and resolubilised in H2O, containing 0.1% formic acid and 3% acetonitrile. Peptide/protein contents were es timated using a Nanodrop spectrophotometer.

SHOTGUN METAPROTEOMIC ANALYSIS.

Aliquots of the fractions, corresponding to approx. 250 ng protein, were analysed in duplicates using a shot-gun proteomics approach. Briefly, the samples were analysed using a nano-liquid-chromatography system consisting of an ESAY nano EC 1200, equipped with an Acclaim PepMap RSLC RP C18 separation column (50 pm x 150 mm, 2pm and 100A), and a QE plus Orbitrap mass spectrometer (Thermo). The flow rate was maintained at 300 nL/min over a linear gradient to 30% solvent B over 60 or 90 minutes, and finally to 75% B over additional 30 minutes. Solvent A consisted of H2O containing 0.1% formic acid, and solvent B consisted of 80% acetonitrile in H2O and 0.1% formic acid. The Orbitrap was op erated in data-dependent acquisition (DDA) mode where the top 10 mass peaks were isolated and fragmented using a NCE of 28. The AGC target was set to le5, at a max IT of 54ms and 17.5K resolution at MS2.

DNA EXTRACTION AND METAGENOMIC ANALYSIS.

DNA from granules was extracted using the DNeasy UltraClean Microbial Kit (Qi- agen, The Netherlands). Following extraction, DNA was checked for quality by gel electro- phorese and by using a Qubit 4 Fluorometer (Thermo Fisher Scientific, USA). Metagenomic sequencing was performed by Novogene Ftd. (Hongkong, China). Briefly, for library con struction, a total amount of lpg DNA per sample was used as input material. Sequencing libraries were generated using NEBNext® Ultra™ DNA Fibrary Prep Kit for Illumina (NEB, USA) following manufacturer’s recommendations. The DNA sample was fragmented by sonication to a size of 350bp, then DNA fragments were end-polished, A-tailed, and li gated with the full-length adaptor for Illumina sequencing with further PCR amplification. PCR products were purified and libraries were analysed for their size distribution using an Agilent 2100 Bioanalyzer, and quantified using real-time PCR. The clustering of the index- coded samples was performed on a cBot Cluster Generation System according to the manu- 12 facturer’s instructions. After cluster generation, the library preparations were sequenced on an Illumina HiSeq platform and paired-end reads were generated. Raw reads were quality checked and statistical low-quality reads were trimmed. Trimmed reads were assembled us ing metaSPAdes v3.13.0 with default settings. For scaffolds larger than 1500 base pairs, tax onomic affiliation was determined using RefineM. Taxonomic annotation was performed according to GTDB.

METAPROTEOMIC DATA ANALYSIS AND PEPTIDE SEQUENCE DATABASE PROCESSING.

The mass spectrometric raw data were analysed using the bioinformatics software so lution PEAKS Studio X using the metagenomics constructed protein assembly database ob tained from the (AGS) granule material. Data were analysed allowing for 20 ppm parent ion and 0.02 m/z fragment ion mass error, 2 missed cleavages, carbamido methylation as fixed and methionine oxidation and N/Q deamidation as variable modifications. Peptide spectrum matches were filtered against 1% false discovery rate (FDR) and protein identifications with > 2 unique peptides were accepted as significant. The identified peptide sequences were fur ther matched against a peptide database unique for the present activated granular sludge mi- crobiome - constructed from the metagenomics database - using Matlab R2020b, to assign a taxonomic lineage to the obtained peptide sequences using the lowest common ancestor ap proach (LCA). Moreover, sequences were accessed for common occurrence between treat ment plants and suitability for quantification, e.g. by considering such as the presence of chemical modifications, abundance or additional scoring parameters. Finally, a subset of the genus level peptides was selected to uniformly represent the community or to represent spec ified taxa, or functions of interest. Taxon proportions are represented here by summing up (unique) peptide sequence frequencies, or their intensities respectively.

Results

Two waste water treatment plants (WWTP-I and WWTP-II, located in the Nether lands, were analysed. Table 1 shows metagenomics results annotations of top taxonomic levels of granules of WWTP-I.

Table 1: shows the top identified taxonomies by metagenomics (based on % mapped reads) from granules of the WWTP-I, detailed from the Phylum to Genus level.. Data are cut-off at a certain level, so not all data is shown.

Figs la-b show a graphical representation of metaproteomics data obtained from WWTP-II (fig. la) and WWTP-I (fig. lb). From a somewhat distinct perspective the two data-sets are rather comparable, and clearly have some differences. 13

Phylum % of mapped % of mapped

Order

_ _ _ reads reads

Proteobacteria 19,61% Burkholderiales 12,01%

Bacteroidota 14,06% Chitinophagales 9,59%

Acidobacteriota 4,31% Actinomycetales 2,67%

Actinobacteriota 3,56% Rhodobacterales 2,05 %

Chloroflexota 1,85% Vicinamibacterales 1,85%

Gemmatimonadota 0,95% unclassified 1,78%

Myxococcota 0,91% Sphingomonadales 1,51%

Vermcomicrobiota 0,44% Flavobacteriales 1,50% unclassified 0,39% Anaerolineales 1,35%

Nitrospira 0,32% Rhizobiales 1,07%

Gemmatimonadales 0,93%

% of mapped reads UBA5704 0,85%

Gammaproteobacteria 14,36% Competibacterales 0,84%

Bacteroidia 12,73% Bacteroidales 0,78 %

Alphaproteobacteria 5,22 % Haliangiales 0,70 %

Actinobacteria 3,22% Bryobacterales 0,66%

Vicinamibacteria 1,89% SJA-28 0,62%

Anaerolineae 1,64% Ignavibacteriales 0,60%

Ignavibacteria 1,23% Steroidobacterales 0,44%

Thermoanaerobaculia 1,08% Pedosphaerales 0,40% unclassified 0,97 % Propionibacteriales 0,36%

Gemmatimonadetes 0,93% Cytophagales 0,34%

Polyangia 0,83% Caulobacterales 0,33%

Acidobacteriae 0,71% Holophagales 0,32%

Verrucomicrobiae 0,44% Nitrospirales 0,32%

Holophagae 0,32%

Nitrospiria 0,32%

Acidimicrobiia 0,31 %

% of ma ed % of mapped

Genus reads

Burkholderiaceae 6,14% unclassified 8,81%

Chitinophagaceae 4,33% UBA7236 2,04%

Rhodocyclaceae 4,32% Ferruginibacter 1,73% unclassified 3,51% JOSHI-001 1,68%

Saprospiraceae 2,42% JJ008 1,53%

Dermatophilaceae 2,40% PHOS-HE28 1,15%

BACL12 2,37% Propionivibrio 1,05%

Rhodobacteraceae 2,04% Tabrizicola 1,04%

UBA2999 1,66% GCA-2748155 1,00%

Sphingomonadaceae 1,50% OLB14 0,94% envOPS12 1,27% Rubrivivax 0,91%

PHOS-HE28 1,16% Dechloromonas 0,79%

Gemmatimonadaceae 0,88% Competibacter 0,78 %

UBA5704 0,84% UBA5704 0,77 %

Competibacteraceae 0,84% Accumulibacter 0,74 %

Haliangiaceae 0,70% SCN-70-22 0,73 %

Bryobacteraceae 0,65% UBA2376 0,64%

OLB5 0,59% Rhodoferax 0,63%

Ignavibacteriaceae 0,56% OLB5 0,59%

Gallionellaceae 0,54% Sulfuritalea 0,55%

Steroidobacteraceae 0,43% Fen-999 0,55%

Pedosphaeraceae 0,35% UBA690 0,54%

Holophagaceae 0,32% Ignavibacterium 0,53%

Nitrospiraceae 0,32% Ga0077559 0,50 %

Anderseniellaceae 0,30% Rhizobacter 0,44% 14

Table 1.

2a.Taxon 2b.Taxon

Total intensi- ^ Total intensi- ^ (Phylum level) (Phylum level) WWTP-II ty WWTP-I ty

Proteobacteria 52754122670 6%2 Proteobacteria 25408237260 68,6 Actinobacteriota 6816455200 8,9 Nitrospirota 4913487100 13,27 Bacteroidota 5715209590 7,5 Actinobacteriota 2172902210 5,87 Acidobacteriota 3454495810 4,5 Bacteroidota 1937695600 5,23 Chloroflexota 3116359250 4,1 Chloroflexota 1177024000 3,18 UBA10199 2566575090 3,4 Acidobacteriota 1109375100 3 Nitrospirota 541768900 0,7 Cyanobacteria 68673000 0,19 Myxococcota 381402200 0,5 UBA10199 63995000 0,17 V ermcomicrobiota 261479400 0,3 Gemmatimonadota 55265900 0,15 Gemmatimonadota 209259400 0,3 Myxococcota 30523600 0,08 Cyanobacteria 164637200 0,2 V ermcomicrobiota 21798100 0,06 Elusimicrobiota 97227400 0,1 MBNT15 17971800 0,05 MBNT15 39840000 0,1 Planctomycetota 15434900 0,04

Desulfobacterota_A 38363100 0,1 Elusimicrobiota 14464000 0,04

Planctomycetota 10307700 <0,1 Desulfuromonadota 9993000 0,03

Firmicutes_G 6577000 <0,1 Firmicutes_G 8436500 0,02

Poribacteria 4628000 <0,1 Marinisomatota 7692000 0,02

Desulfnromonadota 1917800 <0,1 SAR324 5080200 0,01

Table 2a shows top phylum levels for WWTP-I, Table 2b shows top phylum levels for WWTP-II

Tables 2a, b show the underlying data for the metaproteomics established microbiome composition bar graphs shown in fig. la,b.

Again results are quite comparable, which may be expected in view of both examples relating to a WWTP, and at the same time results show clear difference. These differences may be attributed to different performances of the respective plants.

Figs. 2a and 2b represent a graphical display relating to genera compromising 75% of pep tide peak areas.

15

Tables 3 a (WWTP-II) and 3b (WWTP-I) show the underlying data.

Tables 3a,b show the underlying data for the metaproteomics established microbiome composition bar graphs shown in figs. 2a, b.

3a. Taxon 3b. Taxon (Genus level) Total (Genus level) Total WWTP-I intensity % WWTP-II intensity _ %

1504419491

Competibacter 10200821350 33.3 Competibacter 0 26,2

Nitrospira_A 4913487100 16 Accumulibacter 3944569200 6,9

Accumulibacter 3732808230 12,2 UBA7399 2450375500 4,3

Rhizobacter 1133889810 3.7 Propionivibrio 1682610290 2,9

Nitrosomonas 1010608500 3.3 Nitrosomonas 1622785500 2,8

Ignavibacterium 564430700 1.8 JOSHI-OOl 1531125500 2,7

Fen-999 541494000 1,8 Dechloromonas 1467696200 2,6

Dechloromonas 516698100 1,7 UBA7236 1447233250 2,5

Microbacterium 478092200 1,6 Tabrizicola 1356120100 2,4

JOSHI-OOl 476970200 1,6 Rhizobacter 1312651900 2,3

OLB14 375720300 1,2 S PLOW 02-01-44-7 1237316400 2,2

GCA-2699125 349522000 1,1 OLB14 1207645300 2,1

Propionivibrio 346273900 1,1 GCA-2748155 946845400 1,6

Ga0077559 317533100 1 OLB5 928050900 1,6

UBA690 308181500 1 QKVK01 859242600 1,5 Bowdeniella 856436900 1,5 Methylomicrobium 818980500 1,4 Ga0077526 775912200 1,4 Rubrivivax 750710610 1,3 Acidovorax_B 679174800 1,2 Pedococcus 624438000 1,1 UBA5704 595463600 1 ZC4RG36 580422500 1 39-52-133 547569800 1

Fig. 3 shows schematics of reducing data for monitoring. Starting with some 850000 protein sequences from metagenomics, which would result in some 7.7 million peptide se quences (considering the selected constraints) a reduced set of to be monitored peptides of some 650 peptides is selected (see also table 4).

Table 4: . WWTP-I # peptides

Theoretically possible peptides by metagenomics 7738222

Actually observed peptides 23529

Total specific to genus level (and lower) 8436

Selected subset genus (and lower) 655 16

Table 5 shows examples of peptide sequences for WWTP-I to be monitored. These sequences are for clarification and need not be searched.

The sequences are examples of the selected peptide sequences which would be used to specifically monitor Competibacter, as well as others, for the other genera found present in the treatment plants.