*4.2. Bead Bashing Was Determined to Be the Optimal Lysis Method*

Bead bashing was determined to be a more effective lysis technique when compared against sonication (Figure 2a). These are non-aggressive lysis techniques employed for *Blastocystis* as it does not possess a cell wall and is a single-celled organism, so cells are not connected by an extracellular matrix. One study by Geier et al. on *Caenorhabditis elegans* investigated different bead beating techniques, including some at cryogenic temperatures which produced successful results. A tissue homogenizer proved to be the most effective method here, yet it should be considered that *C. elegans* is a multicellular organism, meaning a more aggressive lysis technique is required [10]. Other research has demonstrated that cryopulveristation and tissue homogenisers were successful techniques for the lysis of mammalian cells [12,13]. However, sonication had proved successful in *Arabidopsis thaliana* [9], which has a cell wall and is tougher to break than *Blastocystis*. As sonication and bead bashing had both proved successful in tougher cells than *Blastocystis,* these two methods were selected. Bead bashing produced reproducible results (Figure 2a) against sonication, with only two selected peaks determined as outliers amongst all the samples. Nevertheless, the results of the extracted metabolite concentration ratios were not significant. The differences in concentrations of metabolite extracted ranged between 0.48 and 1.31 (Figure 2a) for most of the selected extracted metabolites, with the exception of formate and alanine in the 3C vs. 4C sample set, whose differences in concentration ranged between 0.14 and 2.58. The number of metabolites extracted produced three reproducible triplicates all suggesting that bead bashing was a better lysis technique than sonication and thus, bead bashing was consistently more successful than sonication.

#### *4.3. Temperature Was Not an Important Factor in Metabolite Extraction*

Incubation temperature was determined to not be a significant factor in successful metabolite extraction from *Blastocystis.* Additionally, as higher temperatures are more likely to facilitate chemical reactions, performing the experiment at room temperature may be essential for maintaining metabolite integrity. This is consistent with a past study by Beltran et al. [11]. However, it could also be the case that a 3-min incubation at the relevant temperature may not be long enough to have a sufficient effect and provides an avenue for future research into method optimisation. We would also like to emphasize that due to the nature and sensitivity of the organisms to oxygen, the objective was to minimise the extraction time to maintain sample integrity. RT against −20 ◦C (Figure 3a) produced a range of metabolite concentration ratios between 0.79 and 1.29. There were therefore no consistent, significant results and this was reproducible, suggesting that neither RT nor −20 ◦C was more successful. In past studies on human vein tissue and *C. elegans*, incubation at RT has been successfully performed [10,30], and similar experiments using *A. thaliana* demonstrated that successful extractions had been performed at −20 ◦C.

In the 60 ◦C incubation against the −20 ◦C incubation (Figure 3b), all of the extracted metabolite concentration ratios were between 0.64 and 1.06. All of the ratios were reproducible between the samples and there was no significant difference determined between them. For the number metabolites extracted ratios were both below 1.0, suggesting that −20 ◦C incubation was a more efficient incubation temperature to perform metabolite extraction than 60 ◦C. As RT was shown to be of similar efficacy to −20 ◦C, RT was selected as the best and most practical incubation temperature.

In summary, the most effective protocol determined by this study is shown in Figure 4. To summarize, this included methanol as the extraction solvent, accompanied by bead bashing and incubation at room temperature. Lyophilisation was used in each trial as a drying method and appeared to be a clean, consistent and successful drying technique. Although many of the results were reproducible, there were numerous outliers and, in some cases, only two reproducible results were produced amongst triplicates. For this reason, future work will aim to include more repeats in order to increase the reliability of the data. Therefore, for our final protocol quintuplets will be used, thus allowing the dismissal of one outlier, if necessary, to have successful triplicates.

The metabolites extracted by this protocol include amino acids such as alanine and leucine and molecules involved in energy metabolism such as acetate and succinate (Table S5, Supplementary Information). Additionally, a wide range of other molecules involved in biological processes such as betaine and malonate were present. The protocols trialled produced a range of metabolites numbering between 25 and 65. These were all polar molecules, as the solvents used target polar metabolites specifically. In the only other metabolomic study of a protozoan parasite, Vermathen et al. detected 31 different metabolites in *Giardia lamblia* using 1H HR-MAS NMR. However, they detected 22 amino acids (18 proteogenic and 4 non-proteogenic) which is at a higher abundance than what was detected here in *Blastocystis* [4]. However, molecules such as betaine and succinate which are involved in biological processes were not detected in *G. Lamblia* [4] but were detected in quite a high abundance in *Blastocystis*. This could be because of the two organisms' different metabolisms, but also may be due to *Blastocystis* morphing into the cyst form and altering its metabolism subject to environmental changes.

Other NMR metabolomics studies of eukaryotic cells have demonstrated a similar number of metabolites to that extracted from *Blastocystis* at high concentrations. In a study on *Caenorhabditis Elegans* by Geier et al., 32 metabolites were detected at concentrations ranging between 2.48 mM and 5.73 mM [7]. Furthermore, in a study by Geier et al. on the avian liver, 52 polar metabolites were detected [10], and in a study on the rat liver by Lee et al., 30 metabolites were detected at concentrations ranging between 13.6 μM and 5.28 mM using methanol as an extraction solvent [8]. Bruno et al. extracted 38 metabolites from skeletal muscle using methanol and chloroform [9]. Methanol and chloroform form a two-layered solution with chloroform on top and methanol on the bottom. The polar metabolites migrate towards the methanol layer and the non-polar metabolites migrate towards the chloroform layer [9].

Even though we were unable to analyse a wider range of molecules, our established methodology was determined to be the most efficacious process from this study to use for the extraction steps for future metabolomics studies on *Blastocystis*. There are a wide range of metabolites which were not detected in this study which have been detected in past studies to map *Blastocystis*' metabolism. Malate, oxaloacetate and succinyl-coA, for example, are all involved in *Blastocystis* energy metabolism and ATP generation, but were not detected using this extraction method [1]. Additionally, the production of amino acids isoleucine and serine have also been detected in past studies [1] but were not detected using this method. This could be down to *Blastocystis* morphing into cyst form and its metabolism becoming dormant but could also be down to the inefficiency of this method for extracting those specific metabolites.

## **5. Conclusions**

In this study, we developed an efficient and robust protocol to extract and analyse polar metabolites from *Blastocystis*. We generated many 1H-NMR spectra to provide detail on the efficacy of each step of the protocol. This is the first extraction method described for NMR metabolomics analysis of *Blastocystis* species and it will spearhead future investigations to determine the metabolome of other *Blastocystis* subtypes, both in vitro, but also in vivo (e.g., stool metabolomic profiles). As such, this easy-to-use procedure could be applied to establish biomarkers in stool samples that could be subsequently used for (infectious) disease diagnosis.

**Supplementary Materials:** The following are available online. Table S1: C μME/M for a selection of metabolites extracted from the ethanol vs. methanol experiment; Table S2: C μME/M for a selection of metabolites extracted from the sonication vs bead bashing experiment; Table S3. C μME/M for a selection of metabolites extracted from the RT vs. −20 ◦C incubation experiment; Table S4. C μME/M for a selection of metabolites extracted from the 60 ◦C vs. −20 ◦C incubation experiment; Table S5. Metabolites extracted by the optimal extraction method and their concentrations. Figure S1: Reproducibility of the ethanol extractions against the methanol extractions-C.Vs of ethanol vs. methanol metabolites extracted with outliers and without outliers; Figure S2: Reproducibility of sonication lysis against bead-bashing lysis. C.Vs of sonication vs. bead bashing metabolites extracted with outliers and without outliers. Figure S3. Spectrum obtained from the optimal extraction protocol deduced from this study.

**Author Contributions:** Conceptualisation, J.M.N. and A.D.T.; methodology, J.M.N., E.L.B., L.Y. and G.S.T.; software, G.S.T.; validation, G.S.T., E.L.B. and J.O.R.; formal analysis, J.M.N., L.Y. and E.L.B.; investigation, J.M.N.; resources, A.D.T.; data curation, J.M.N.; writing—original draft preparation, J.M.N.; writing—review and editing, G.S.T., J.O.R. and A.D.T.; supervision, G.S.T. and A.D.T.; project administration, A.D.T.; funding acquisition, A.D.T. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research was funded by Biotechnology and Biological Sciences Research Council, grant number: BB/M009971/1. The NMR facility is supported by a Wellcome Trust Equipment Grant 091163/Z/10/Z (UK). J.M.N. was supported by a Kent-Health studentship and E.L.B. is supported by a GTA studentship from the School of Biosciences, University of Kent.

**Institutional Review Board Statement:** Not applicable.

**Informed Consent Statement:** Not applicable.

**Conflicts of Interest:** The funders had no role in the design of the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript, or in the decision to publish the results.

**Sample Availability:** Not Available.
