Pangenome-based design of strain-specific primers allows the specific monitoring of engraftment in different habitats

doi:10.21203/rs.3.rs-4182624/v1

Download PDF

Article

Pangenome-based design of strain-specific primers allows the specific monitoring of engraftment in different habitats

https://doi.org/10.21203/rs.3.rs-4182624/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

Lactic acid bacteria (LAB), including many well-known beneficial bacteria, have seen a rise in the number of applications with specific strains across various areas, including live biotherapeutic products (LBPs). The most extensively researched strains belong to the Lactobacillaceae. Assessing the survival and persistence of specific strains in different niches is still an important challenge, while selective monitoring techniques are often lacking at strain level. Here, we show a robust pangenome-based approach for detecting singletons, which can be used to develop strain-specific primers. We developed selective and specific primers for six strains across different LAB species. The primers for the widely-used probiotic L. rhamnosus GG and L. plantarum WCFS1 were validated in in vivo studies and showed that these strains can persist in and on other habitats such as the human skin, upper respiratory tract and fermented vegetables. In conclusion, the selection of unique genes derived from the pangenome of a species resulted in a specific and sensitive method based on qPCR to detect and monitor strains in different habitats. This approach can be readily extended to other bacterial strains on other families for any type applications in research and industry.

Biological sciences/Microbiology/Applied microbiology

Biological sciences/Microbiology/Microbial genetics

Microbial applications are increasingly explored and implemented in different domains, such as probiotics¹, medicine (live biotherapeutic products or LBPs²), food applications (fermentation starter cultures³), and agriculture (biocontrol⁴ and biostimulant⁵) applications. A key challenge in all these applications with live micro-organisms is obtaining sufficient survival and engraftment of the specific strains so that they can optimally exert their activity and efficacy⁶. For instance, the effective and selective monitoring of the applied strains is crucial in areas, such as probiotic and LBP applications. Probiotics are defined as “live microorganisms that, when administered in adequate amounts, confer a health benefit on the host”⁷, with well-known examples belonging to the family of the Lactobacillaceae, such as Lacticaseibacillus rhamnosus GG⁸, and Lactiplantibacillus plantarum WCFS1⁹. This is a broad scientific definition, and probiotics can enter the market under different regulatory statuses¹⁰. The term live biotherapeutic product (LBP) is dedicated to live micro-organisms for which the intended use is prevention, treatment, or cure of a disease or condition in human beings¹¹. The market for probiotics and LBPs is exponentially growing worldwide, with also the type of formulations and applications expanding beyond gastro-intestinal applications with only oral supplements or fermented dairy products^2,12,13. Furthermore, the bacterial strains used as probiotics and LBPs are rapidly diversifying, indicating the need for specific detection of bacterial strains in the target niches of the diverse applications (gut, skin, vagina, oral cavity, URT and beyond). In addition to LBPs and probiotics, the applications and markets for food fermentations with starter cultures are rapidly expanding. With a growing focus on (regulatory) awareness and research and development, monitoring strain survival, persistence and engraftment has become essential for quality and safety^14,15.

While selective culture conditions used to be sufficient for these monitoring objectives, they can only detect taxa in general levels such as ‘total number of lactic acid bacteria’ after plating on selective growth media. In addition, when absolute numbers of the bacteria in a certain niche are low, they will not be picked up with cultivation methods, because they are rapidly overgrown by fast-growing microorganisms¹⁶. Thus, strain-specific approaches that can better discriminate the supplemented strains from the native and naturally present bacteria in the target microbiomes is essential. Various tools exist for strain monitoring and tracking, which have evolved significantly in the genomic and genetic engineering era. Sensitive tools are often based on fluorescence¹⁷, bioluminescence¹⁸ and DNA barcoding¹⁹. However, these methods all have the major drawback of requiring modifications and the introduction of exogenous DNA, resulting in genetically modified organisms (GMO). These strategies are thus not widely applicable from a regulatory point of view for probiotics or LBPs or for the monitoring of food fermentation starter cultures.

Since the sequencing of the first genome in 1995²⁰, over 36.000 genomes have been sequenced and made available online (based on V214 of the Genome Taxonomy Database²¹). This vast pool of genomic information has revolutionized science, allowing for innovative ways to study bacteria, including clarifying the definition of a strain and monitoring of strains. The definition of a strain has been and still is the subject of debate^22–25 with no uniform definition yet. In the present study, we have defined a strain as having at least an average nucleotide identity (ANI) lower than 99.99% with any other strain.

Non-GMO methods are thus a preferred alternative approach for LBP and related strain monitoring in industry and real-life applications. Strain-level metagenome approaches are being developed²⁶, but these approaches are still expensive as they require very deep sequencing to obtain sufficient genome coverage. We propose that the development of strain-specific primers combined with qPCR can be used in strain monitoring in a cost-effective, quantitative and rapid method^12,13. While many studies have developed and implemented such primers, their development has not been adequately validated across a wide diversity of closely related strains, resulting in concerns about their specificity. For example, in the placebo-controlled Lactin-V study where application of Lactobacillus crispatus CTV-05 was investigated in 228 patients (152 verum, 76 placebo), the strain was also detected in 2–6% of the placebo group versus 48–84% of the treated group, while the placebo group should have never come into contact with the strain, hypothesizing that other strains of L. crispatus (which are commonly present in women at high microbial loads) could also bind to the primers¹². In one of our own human studies with applied probiotic strains, cross-reactivity was also shown for primers that were designed on the srr gene (encoding fimbriae) in Lacticaseibacillus casei AMBR2, an accessory but non-unique gene for this species¹³. While this might not be a problem for habitats where lactobacilli are not dominant members, such as the skin² or respiratory tract²⁷, other habitats such as the human vagina²⁸ and fermented foods²⁹ have a high prevalence of various closely related lactobacilli, making it crucial to find more unique gene primers, as illustrated in the Lactin-V study¹². Hereto, the wealth of LAB genome information that is now publicly available can be leveraged.

The goal of this study was to develop a pipeline that is widely applicable for LAB in different habitats and that depends on unique target genes of the strains of interest and qPCR. We first designed an in silico method to find unique accessory genes (singletons) in a species’ pangenome in order to develop strain-specific primers. We then validated this method for six LAB strains: two widely used probiotic strains, i.e. Lacticaseibacillus rhamnosus GG and Lactiplantibacillus plantarum WCFS1 and four newly isolated strains of interest as novel probiotics or LBPs, i.e. Limosilactobacillus reuteri AMBV038, Limosilactobacillus reuteri AMBF471, Streptococcus salivarius AMBR024, and Streptococcus salivarius AMBR158. Cross-reactivity and qPCR efficiency were experimentally validated. Additionally, the applicability to monitor the persistence and engraftment of strains of interest in different habitats and environments was evaluated for L. rhamnosus GG and L. plantarum WCFS1, two strains that were used in placebo- controlled human intervention studies and starter cultures for vegetable fermentations.

Strain selection

Strains to validate the method were selected based on probiotic potential for different applications. We aimed for genus and species variability in our selection to test the broader application of our approach (

Table ). In addition, we selected strains with high genome accessibility and high safety profiles to make it possible to directly evaluate our approach in vivo. To validate the specificity and selectivity of this approach, additional strains were selected from the same species to test cross-reactivity (Table 1).

Table 1

T = Selection of the different target strains used in this study. C = Strains used to test for cross-reactivity in this study
	Species	Strain	Genome accession number	Isolation source and reference
T	Lacticaseibacillus rhamnosus	GG	GCF_000026505.1	Human gastrointestinal tract ³⁰, model probiotic strain
C		GR-1	GCF_900604925.1	Female urethra ³¹, model probiotic strain
C		AMBR3	GCA_901830385.1	Human respiratory tract ²⁷
C		AMBR4	GCA_901830395.1	Human respiratory tract ²⁷
T	Lactiplantibacillus plantarum	WCFS1	GCF_000203855.3	Human saliva ³², model starter
C		CMPG5300	GCF_000762955.1	Human vagina ³³
C		5057		Human ileostomy effluent ³⁴
C		ATCC8014	GCF_002370965.1	Corn silage ³⁵
T	Limosilactobacillus reuteri	AMBV038	This paper	Human vagina ³⁶
T		AMBF471	GCA_012275185.1	Chicken feces, model cereal fermentation strain
C		RC-14	GCF_002762415.1	Female urogenital tract ³⁷
C		DSM 20016	GCF_000016825.1	Human feces ³⁸
T	Streptococcus salivarius	AMBR024	GCF_905071825.1	Human respiratory tract ³⁹, model potential probiotic strain
T		AMBR158	GCF_905071905.1	Human respiratory tract ³⁹
C		HSISS4	GCF_000448685.2	Human gastrointestinal tract ⁴⁰
C		ATCC25975	GCF_002094975.1	Saliva ⁴¹

Pangenome construction and selection of unique genes

All genomes per species were selected using the genome taxonomy database v214⁴² and downloaded from NCBI GenBank (Supplementary info X or in Github). Proteins were predicted with Prodigal version 2.6.3 (github.com/hyattpd/Prodigal) and pangenomes were inferred using SCARAP (github.com/swittouck/SCARAP). Genome content similarity was checked using fastANI⁴³. Based on the pangenome, singleton genes were selected for the strains of interest, while excluding genomes that were virtually identical to those of the strains of interest. The tool is made accessible on GitHub (github.com/TomEile/uniortho).

Microbial RNA and DNA extraction

Bacterial DNA was extracted using the PowerSoil Pro DNA Isolation Kit (Qiagen, Hilden, Germany), according to the instructions of the manufacturer. Next, DNA concentrations were measured using the Qubit 3.0 Fluorometer (Life Technologies, Ledeberg, Belgium). For the primer efficiency and validation, DNA extraction was started from overnight cultures of all selected bacterial strains. Strains were grown in deMAN, Rogosa and Sharpe (MRS) broth (Difco) or Brain Hearth Infusion (BHI) broth (Neogen Culture Media, USA). Subsequently, 10⁸ CFU/mL of the overnight cultures was centrifuged and dissolved in phosphate-buffered saline (PBS), 500µl was then used for DNA extraction. For the application of the primers on clinical samples, 500µL of eNAT swab buffer was used for DNA extraction. For the fermentation cultures, bacterial RNA was extracted using the RNeasy powermicrobiome (Qiagen, Germany) including an on-column DNase digestion. cDNA was subsequently made using Readyscript (Sigma Aldrich, USA).

Primer design and in vitro validation

Primer sequences were designed to amplify the unique sequences and tested for cross-reactivity with other strains of the same species. We started with colony PCR for a first evaluation of possible cross-reactivity. PCR primers were designed in Geneious and synthesized by Integrated DNA Technologies (IDT, Leuven, Belgium). Of the different primer sequences designed by Geneious, the best primer pairs were selected based on the lowest hairpin, self-dimer, cross-dimer and melt temperature mismatching. 10 µl DNA sample was added to 15 µl mastermix (2.5 µl 10x VWR buffer, 2.5 µl 10 µM primers, 0.5 µl dNTPS 10mM, 0.2 µl Taq polymerase, 6.8 µl molecular grade water per sample), primers as indicated in Table 3. The following PCR conditions were used: denaturation at 95°C for 2 min., followed by 30 cycles at 95°C for 30 sec., 55°C for 30 sec. and 72°C for 1 min./kb, with the final extension at 72°C for 5 min. PCR products were visualized on a 1% agarose gel.

Table 2

PCR primer sequences for the different target strains used in this paper
Species	Strain	ID	Primer sequence
Lacticaseibacillus rhamnosus	GG	SL983	CGGCTTGACAGAGAATGCTA	F	PCR
		SL984	CCAAAGGCTCCGAAGTTGAA	R	PCR
		SL993	CTGGCACTCATGAATCCTTACA	F	qPCR
		SL994	CCATTCGGTAGGCTACTTCTTC	R	qPCR
Lactiplantibacillus plantarum	WCFS1	SL975	CAGAGCTGTACCGCTTGTTA	F	PCR
		SL976	CTACGGCAATGCATTGTCCT	R	PCR
		SL604	GCCACAACACTTCAGCAATAC	F	qPCR
		SL605	GTGCCATACACCCTGGTAAG	R	qPCR
Limosilactobacillus reuteri	AMBV038	SL998	CAGGTCAGTAACTTATCAGC	F	PCR
		SL997	CTTGCTGAACTTGCGCTAGT	R	PCR
		SL988	TGGTCAAGACTGGCAAATGA	F	qPCR
		SL987	CTGTGCTGAGGTGTTCCATAA	R	qPCR
Limosilactobacillus reuteri	AMBF471	SL965	CGTGAGATTCTTGACGCCAT	F	PCR
		SL966	TTAGTCGTTGTCAGTGTCCG	R	PCR
		SL589	CGTGAGATTCTTGACGCCATAA	F	qPCR
		SL590	CCGCTGAATATCTTGGACAACT	R	qPCR
Streptococcus salivarius	AMBR024	SL967	GCGATTCCTGCTCTACATAC	F	PCR
		SL968	CTAGCTCTTGAAGCACCAAC	R	PCR
		SL591	ATGCGATTCCTGCTCTACATAC	F	qPCR
		SL592	TCCCTGCTCCTCCTTGAATA	R	qPCR
Streptococcus salivarius	AMBR158	SL981	ACGAAGAATAGTCGAGCGGA	F	PCR
		SL982	GGTAAATGTGTCTTACACCC	R	PCR
		SL637	TCGAGGAAGTACAGAGTTTGATG	F	qPCR
		SL638	GAACTCTTGCAAATCCAACACA	R	qPCR

After evaluation of the absence of cross-reactivity for the selected unique sequences with colony PCR, qPCR primers were designed based on the selected unique sequences and primer design for intercalating dyes was performed in PrimerQuest™ Tool (IDT), using the default option for qPCR primers with intercalating dyes. Of the different primer sequences designed by PrimerQuest™ Tool the best primer pairs were selected based on the lowest hairpin, self-dimer, cross-dimer and melt temperature mismatching. These primer pairs were then first tested on primer efficiency whereafter the best primer pairs were selected to test on cross-reactivity. Primers were chemically synthesized by Integrated DNA Technologies (IDT, Leuven, Belgium), primer sequences in Table 4.

To test the sensitivity of the developed qPCR primers, standard curves were derived from serially diluted bacterial DNA and were subsequently used to calculate the primer efficiency (Figure S1) and estimate bacterial DNA concentrations in the samples. All primer efficiencies were between the required numbers of 90 and 110%, meaning that for each round of replication, the number of fragments was approximately doubled. Bacterial concentration was determined by plating out a serial dilution on MRS agar. qPCR was performed as described before⁴⁴. The following PCR conditions were used: pre-incubation at 50°C for 10 min., denaturation at 95°C for 20 sec., followed by 40 cycles at 95°C for 15 sec., and 60°C for 30 sec., with the melting curve at 95°C for 15 sec., 60°C for 1 min., 95°C for 15 sec. Primer efficiency was calculated using the formula by Bustin et al.⁴⁵.

The specificity of the unique genes was validated by testing cross-reactivity of the primers for different strains from the respective species.

In situ validation of the primers

To evaluate the application potential of strain-specific L. rhamnosus GG and L. plantarum WCFS1 qPCR primers, their absolute abundances were evaluated in three different habitats: skin, URT and starter culture fermentation samples. 40 skin samples were obtained from a placebo-controlled clinical trial with a topical probiotic cream containing 10^9–10^10 CFU/gram L. rhamnosus GG and L. plantarum WCFS1². Secondly, the absolute abundance of L. rhamnosus GG and L. plantarum WCFS1 was monitored in the URT using 12 samples from a clinical trial using a probiotic throat spray, containing 9.5x10^8 CFU/mL L. plantarum WCFS1¹³. For both skin and URT human intervention studies, methods for study design, human subjects, sample collection, bacterial DNA extraction, and product formulation as described by respectively Lebeer et al. (2022)² and De Boeck et al. (2022)¹³.

To evaluate the application potential of strain-specific L. rhamnosus GG and L. plantarum WCFS1 qPCR primers, their abundances were evaluated in starter culture fermentation samples. Carrots were rinsed with H₂O and processed with Solis Juice Fountain Pro Type 843 centrifuge. Subsequently, 2.5% (w/v) NaCl was added to the carrot juice and was divided into 0.25 L Weck jars. The optical density (OD) of the overnight culture of the starter culture was measured at 600 nm and a volume containing ca. 10⁸ CFU was used to prepare starter culture solution. The pelleted cells were then washed twice using sterile PBS (phosphate buffered saline) at 4000 g for 10 minutes and resuspended in fresh carrot juice. This final solution was then added to the Weck jars containing 250mL fresh carrot juice (2.5% NaCl), leading to a starting concentration of ca. 10⁵ CFU/mL of the starter culture. Weck jars were then closed and fermented at 20°C for 15 days. Each 400µl sample was frozen in liquid nitrogen and kept on -80°C before subsequent RNA extraction. Total RNA was extracted using the RNeasy PowerMicrobiome kit (Qiagen), which included an on-column Dnase digestion step. After RNA extraction, the presence of DNA was checked using a PCR on the 16S region (with 27F and 1492R), and the PCR product was visualized on a 1% agarose gel. Complementary DNA (cDNA) was synthesized using Readyscript (Sigma-Aldrich). After RNA extraction and cDNA synthesis, the primers for L. rhamnosus GG and L. plantarum WCFS1 were used to monitor the survival and growth of this strain in carrot fermentations, where this strain was added as a starter culture at 10⁵ CFU/mL and monitored for 15 days.

Pangenome analyses can reveal strain-unique orthogroups

A pipeline for the selection of unique accessory genes was designed to process the genomes of strains of interest and compile a list of all genes that are exclusive/unique to each strain within its respective species, along with their corresponding gene families (Fig. 1A). This pangenome analysis revealed a substantial number of putative singleton orthogroups for each species, indicating the presence of additional species for which singleton orthogroups exist, suitable for strain monitoring (Fig. 1B).

We then zoomed in on specific strains of interest and investigated whether they had unique genes. We performed this analysis for 6 strains from 4 species of interest for ongoing probiotic/fermented food applications in our lab, with L. plantarum WCFS1 as model strain for vegetable food applications and L. rhamnosus GG as our model strain as probiotic/LBP (Table 1). Among these strains analyzed, only the newly isolated strains L. reuteri AMBV038, L. reuteri AMBF471, S. salivarius AMBR024, and S. salivarius AMBR158, showed 3, 14, 19 and 7, unique orthogroups, respectively. For L. plantarum WCFS1 and L. rhamnosus GG, we did not find unique orthogroups, indicating that the publicly available genomes submitted for these strains were not unique or too similar, which we confirmed using fastANI (Fig. 1C). L. plantarum WCFS1 had 2 genomes with an ani larger than 99.99 and L. rhamnosus GG had 5 such submissions of potential clones sequenced several times (Supplementary table S1). Clones with an ANI above 99.99% from the analysis to determine unique orthogroups were excluded (Fig. 1C).

Validation of unique orthogroups

To validate the strategy of the unique orthogroups as targets for strain-specific primers, one of the unique orthogroups was selected for each strain and PCR primers were developed to experimentally check selectivity and sensitivity. For each target strain, this was tested against three other control strains per species (Table 1). A first evaluation of cross-reactivity was evaluated using the strain-specific PCR primers. For all unique accessory genes, the designed PCR primers (Table 2) tested negative for cross-reactivity against three other strains of the same species (Fig. 2A), indicating the selectivity of the unique accessory genes.

Next, qPCR primers were designed on the same unique accessory genes, which were generally smaller, ranging from 80–200 base pairs (Table 2). The specificity of the qPCR primers was evaluated on three other strains within the same species (Fig. 2B). Strain specificity of the qPCR primers was confirmed when the difference in Ct value between the target and other strains was 10 or higher and the strain specificity for all primer pairs could be confirmed by these results.

Unique orthogroups approach can viably monitor the presence in different niches

To validate the application potential of the strain-specific qPCR primers to monitor strains, we tested the primers for L. rhamnosus GG and L. plantarum WCFS1 in three different niches: skin, URT, and starter culture vegetable fermentations.

For the skin samples, L. rhamnosus GG and L. plantarum WCFS1 were topically applied in a placebo-controlled intervention study and monitored over time. High concentrations of both L. rhamnosus GG and L. plantarum WCFS1 were detected in the skin samples with the probiotic intervention ((3.1 ± 0.2)x10⁵ CFU/mL, (8.5 ± 0.5)x10⁵ CFU/mL). In contrast, the placebo skin samples, showed lower to undetectable concentrations of both strains (mean: (4.42 ± 0.08)x10¹ CFU/mL, (8.3 ± 0.2)x10¹ CFU/mL). Also in URT samples from a placebo-controlled trial, the strain-specific qPCR primers showed good specificity and efficiency. High concentrations of both strains were measured (1.9 ± 0.6)x10⁵ CFU/mL and (4.9 ± 0.8)x10⁶ CFU/mL for L. rhamnosus GG and L. plantarum WCFS1, respectively), with undetectable concentrations for L. rhamnosus GG and L. plantarum in the placebo group.

For the starter cultures, L. plantarum WCFS1 and L. rhamnosus GG were added at the start of carrot juice fermentations and monitored at various timepoints. We could observe that L. plantarum WCFS1 shows promising behavior as potential starter culture for these vegetable fermentations, reaching relative high concentrations at all time points (3.2 ± 0.2)x10³ CFU/mL, whereas for L. rhamnosus GG was not promising as starter culture, because only detected in the vegetable fermentations on day 1 (5.62 ± 0.07) *10² CFU/mL. For L. plantarum WCFS1, all control samples were negative, pointing towards good selectivity. Similarly for L. rhamnosus GG, all control samples and the later stages of the fermentation were negative.

The advancements in whole genome sequencing technologies, as well as the reduction in sequencing costs, have significantly increased the number of available high-quality sequenced genomes. This increase enables more in-depth analysis of intraspecies variation through pangenome analysis. The pangenome represents the entire set of genes of for example a species or genus, consisting of a core genome – containing sequences shared between individuals of the species or genus – and the accessory genome. In this study, we used pangenome analysis to show that there is a vast intraspecies diversity in different LAB of interest for LBP, probiotic and fermented food applications. We developed a novel approach to track individual strains through the use of unique accessory genes or singletons in the pangenome.

Our results demonstrated that, with a well-designed primer set, qPCR can easily be used as a technique to monitor bacterial concentrations at the strain level if enough genomes are available. However, the minimum threshold of available genomes should still be evaluated in follow-up research, as well as the use of closely related species in case not enough genomes are available, which can be a limitation for rather rare species. Related to the availability of sufficient genomes, it should also be highlighted that an important measure we implemented is to dereplicate genomes, because some strains are sequenced multiple times and added multiple times in GTDB. For instance, L. rhamnosus GG is a widely used probiotic strain ^30, and likely sequenced multiple times, as also mentioned in Wuyts et al (2017)⁴⁶. Similarly, for L. plantarum, one genome with an ANI above 99.99% compared to L. plantarum WCFS1 is available. A significant challenge in developing strain-specific primers is the debate on the criteria that differentiate two strains from one another^22–25. We used a pragmatic strain definition similar to Dijkshoorn et al. (2000)²³. Nevertheless, a clear definition and a general consensus of strain-level classification can help to expand the use of the approach proposed in this paper.

After the detection of unique orthogroups and strain-specific primer development, the application potential of these strain-specific qPCR primers was validated on human microbiome samples from different niches after LBP treatment, and on fermentation samples. The newly obtained results confirmed that the primers were selective to measure the concentration of the target strains on skin and URT swabs after topical application of these strains. In addition, the data confirmed that the two probiotic strains could at least temporarily engraft on the skin and in the URT when applied. However, there was also a low detection in the control skin samples, which could be explained because of the natural presence of these strains on the skin² and/or because of cross-contamination during laboratory practices, which is even more important for low bacterial biomass niches such as the skin and respiratory tract. Overall, the results indicate the strain-specific primers were also efficient and selective for mixed microbial DNA samples. Using this approach, we could also monitor L. plantarum WCFS1 and L. rhamnosus GG in niches were these species are present in high abundances, such as carrot juice fermentations. L. plantarum could be specifically and selectively discriminated from the endogenous L. plantarum strains, confirming its specificity and indicating that this tool might be the right tool to study starter cultures. For the probiotic strain L. rhamnosus GG, we only detected L. rhamnosus GG on day 1, indicating that they are not able to dominate the carrot juice fermentation. This is important for the addition of specific probiotics to novel food matrices, since probiotics should have viable counts present in these matrices ⁷ and our primers combined with qPCR could be a suitable approach. Overall, the results indicate the strain-specific primers were also efficient and selective for mixed microbial DNA samples.

Our approach can also be applied more broadly. The high interest in studying LAB allowed us to immediately include the developed primers for in vitro and in vivo testing in various research projects^47,48. For instance, the primers targeting L. reuteri AMBV038, L. reuteri AMBF471, S. salivarius AMBR024 and S. salivarius AMBR158 were used to evaluate the effect of these strains on in vitro periodontal biofilms, allowing us to real-time track the colonization potential of probiotic candidate strains within the biofilms, which is an important selection criterion⁴⁸. Moreover, our method was previously used to design primers targeting L. rhamnosus GG to measure its colonization in vivo in the URT during and after probiotic intervention⁴⁷. These results indicate the potential of this approach for different microbiome studies, e.g. human microbiome, food microbiology and environmental microbiomes. qPCR is a valuable technique to investigate absolute abundances of certain strains or species in microbiological DNA samples in these niches. The integration of qPCR data with amplicon and metagenomic sequencing data can overcome sequencing limitations, by providing absolute abundance data and give crucial insights into the community dynamics and interactions. Finally, this method, combined with qPCR can serve as a simple, fast and relatively cheap method compared to currently used methods for strain monitoring, such as fluorescence¹⁷, bioluminescence¹⁸ and DNA barcoding¹⁹. In future research, it might even be possible to employ this technique of unique orthogroups for the creation of a curated database for metagenomics and metatranscriptomics.

In summary, this study provides a versatile and easy-to-use pipeline for developing strain-specific primers, enabling the tracking absolute abundances at strain-level in samples of many different applications. The primers were subsequently validated with PCR and qPCR on non strain-specific cross-reactivity and application potential on human microbiome and vegetable fermentation samples. This tool is of interest to study efficacy and potency of strains, e.g. survival and performance as starter cultures or LBPs, and could potentially be expanded to different families and even phyla for use in other applications.

Competing interests

S.L. received funding from different probiotic companies that were not involved in this research. L.D. was funded by VLAIO through a Baekeland mandate in collaboration with YUN NV. T.E. is partially funded through an industrial research VLAIO grant not related to this work. The remaining authors have no conflicts of interest to declare.

Funding

TE and SL were funded by the European Research Council (project 852600). TE is partially funded by VLAIO (HBC.2022.1000). LD was funded by Baekeland from VLAIO (HBC.2020.2873). IDB, JVM, WVB, SW were funded by grants from Research Foundation – Flanders (FWO, respectively grants 12S4222N, 12AZ624N, 1224923N, and 1S08523N).

Acknowledgements

MK, CRD for the extra experiments. WVH for using the primers in his research. EO and IC for supplying skin samples of the acne intervention study. Kingsley C. Anukam and Chidozi N.E. Ibezim for supplying the chicken feces samples in the framework of the 2 months research stay of Chido Ibezem in our lab

Data Availability

Pipeline is publicly available on github (github.com/tomeile/uniortho). All genomes used were publicly available, the in-house genome of AMBV039 is published on ENA via project PRJEB74196.

Larsen, I. S. et al. Experimental diets dictate the metabolic benefits of probiotics in obesity. Gut Microbes 15, (2023).
Lebeer, S. et al. Selective targeting of skin pathobionts and inflammation with topically applied lactobacilli. Cell Reports Med. 3, (2022).
Leroy, F. & De Vuyst, L. Lactic acid bacteria as functional starter cultures for the food fermentation industry. Trends Food Sci. Technol. 15, 67–78 (2004).
Legein, M. et al. Modes of Action of Microbial Biocontrol in the Phyllosphere. Front. Microbiol. 11, 544057 (2020).
Backer, R. et al. Plant Growth-Promoting Rhizobacteria: Context, Mechanisms of Action, and Roadmap to Commercialization of Biostimulants for Sustainable Agriculture. Front. Plant Sci. 9, 402666 (2018).
Wendel, U. Assessing Viability and Stress Tolerance of Probiotics—A Review. Front. Microbiol. 12, 818468 (2022).
Hill, C. et al. The International Scientific Association for Probiotics and Prebiotics consensus statement on the scope and appropriate use of the term probiotic. Nat. Rev. Gastroenterol. Hepatol. 11, 506–514 (2014).
Lebeer, S. et al. Functional analysis of lactobacillus rhamnosus GG pili in relation to adhesion and immunomodulatory interactions with intestinal epithelial cells. Appl. Environ. Microbiol. 78, 185–193 (2012).
de Vries, M. C., Vaughan, E. E., Kleerebezem, M. & de Vos, W. M. Lactobacillus plantarum—survival, functional and potential probiotic properties in the human intestinal tract. Int. Dairy J. 16, 1018–1028 (2006).
Hill, C. et al. The International Scientific Association for Probiotics and Prebiotics consensus statement on the scope and appropriate use of the term probiotic. Nat. Rev. Gastroenterol. Hepatol. 2014 118 11, 506–514 (2014).
FDA. Early Clinical Trials with Live Biotherapeutic Products: Chemistry, Manufacturing, and Control Information; Guidance for Industry. http://www.fda.gov/BiologicsBloodVaccines/GuidanceComplianceRegulatoryInformation/Guida (2016).
Cohen, C. R. et al. Randomized Trial of Lactin-V to Prevent Recurrence of Bacterial Vaginosis. N. Engl. J. Med. 382, 1906–1915 (2020).
De Boeck, I. et al. Randomized, Double-Blind, Placebo-Controlled Trial of a Throat Spray with Selected Lactobacilli in COVID-19 Outpatients. Microbiol. Spectr. 10, (2022).
Capozzi, V., Fragasso, M. & Russo, P. Microbiological Safety and the Management of Microbial Resources in Artisanal Foods and Beverages: The Need for a Transdisciplinary Assessment to Conciliate Actual Trends and Risks Avoidance. Microorganisms 8, (2020).
Ashaolu, T. J. Safety and quality of bacterially fermented functional foods and beverages: a mini review. Food Qual. Saf. 4, 123–127 (2020).
Bron, P. A., Grangette, C., Mercenier, A., De Vos, W. M. & Kleerebezem, M. Identification of Lactobacillus plantarum Genes That Are Induced in the Gastrointestinal Tract of Mice. J. Bacteriol. 186, 5721 (2004).
Spacova, I. et al. Expression of fluorescent proteins in Lactobacillus rhamnosus to study host–microbe and microbe–microbe interactions. Microb. Biotechnol. 11, 317–331 (2018).
Turpin, P. E., Maycroft, K. A., Bedford, J., Rowlands, C. L. & Wellington, E. M. H. A rapid luminescent-phage based MPN method for the enumeration of Salmonella typhimurium in environmental samples. Lett. Appl. Microbiol. 16, 24–27 (1993).
Fasanello, V. J., Liu, P., Botero, C. A. & Fay, J. C. High-throughput analysis of adaptation using barcoded strains of Saccharomyces cerevisiae. PeerJ 8, e10118 (2020).
Fleischmann, R. D. et al. Whole-Genome Random Sequencing and Assembly of Haemophilus influenzae Rd. Science (80-.). 269, 496–512 (1995).
Parks, D. H. et al. GTDB: an ongoing census of bacterial and archaeal diversity through a phylogenetically consistent, rank normalized and complete genome-based taxonomy. Nucleic Acids Res. 50, D785–D794 (2022).
Anderson, B. D. & Bisanz, J. E. Challenges and opportunities of strain diversity in gut microbiome research. Front. Microbiol. 14, 1–8 (2023).
Dijkshoorn, L., Ursing, B. M. & Ursing, J. B. Strain, clone and species: comments on three basic concepts of bacteriology. J. Med. Microbiol. 49, 397–401 (2000).
Hill, C. & International Scientific Association for Probiotics and Prebiotics (ISAPP). What is a strain in microbiology and why does it matter? ISAPP Science Blog (2022).
Van Rossum, T., Ferretti, P., Maistrenko, O. M. & Bork, P. Diversity within species: interpreting strains in microbiomes. Nat. Rev. Microbiol. 18, 491–506 (2020).
Pasolli, E. et al. Large-scale genome-wide analysis links lactic acid bacteria from food with the gut microbiome. Nat. Commun. 11, 1–12 (2020).
De Boeck, I. et al. Lactobacilli Have a Niche in the Human Nose. Cell Rep. 31, 107674 (2020).
Lebeer, S. et al. A citizen-science-enabled catalogue of the vaginal microbiome and associated factors. Nat. Microbiol. 8, 2183–2195 (2022).
Wuyts, S. et al. Carrot juice fermentations as man-made microbial ecosystems dominated by lactic acid bacteria. Appl. Environ. Microbiol. 84, e00134-18 (2018).
Kankainen, M. et al. Comparative genomic analysis of Lactobacillus rhamnosus GG reveals pili containing a human- mucus binding protein. Proc. Natl. Acad. Sci. 106, 17193–17198 (2009).
Petrova, M. I. et al. Comparative Genomic and Phenotypic Analysis of the Vaginal Probiotic Lactobacillus rhamnosus GR-1. Front. Microbiol. 9, (2018).
Kleerebezem, M. et al. Complete genome sequence of Lactobacillus plantarum WCFS1. Proc. Natl. Acad. Sci. 100, 1990–1995 (2003).
Malik, S. et al. High mannose-specific lectin Msl mediates key interactions of the vaginal Lactobacillus plantarum isolate CMPG5300. Sci. Reports 2016 61 6, 1–16 (2016).
Danielsen, M. Characterization of the tetracycline resistance plasmid pMD5057 from Lactobacillus plantarum 5057 reveals a composite structure. Plasmid 48, 98–103 (2002).
Leer, R. J., van Luijk, N., Posno, M. & Pouwels, P. H. Structural and functional analysis of two cryptic plasmids from Lactobacillus pentosus MD353 and Lactobacillus plantarum ATCC 8014. Mol. Gen. Genet. 234, 265–274 (1992).
Allonsius, C. N. et al. Inhibition of Candida albicans morphogenesis by chitinase from Lactobacillus rhamnosus GG. Sci. Rep. 9, 1–12 (2019).
Reid, G., Cook, R. L. & Bruce, A. W. Examination of Strains of Lactobacilli for Properties that May Influence Bacterial Interference in the Urinary Tract. J. Urol. 138, 330–335 (1987).
Sriramulu, D. D. et al. Lactobacillus reuteri DSM 20016 produces cobalamin-dependent diol dehydratase in metabolosomes and metabolizes 1,2-propanediol by disproportionation. J. Bacteriol. 190, 4559–4567 (2008).
Jörissen, J. et al. Case-Control Microbiome Study of Chronic Otitis Media with Effusion in Children Points at Streptococcus salivarius as a Pathobiont-Inhibiting Species. mSystems 6, (2021).
Mignolet, J., Fontaine, L., Kleerebezem, M. & Hols, P. Complete Genome Sequence of Streptococcus salivarius HSISS4, a Human Commensal Bacterium Highly Prevalent in the Digestive Tract. Genome Announc. 4, e01637-15 (2016).
Butler, R. R., Soomer-James, J. T. A., Frenette, M. & Pombert, J. F. Complete Genome Sequences of Two Human Oral Microbiome Commensals: Streptococcus salivarius ATCC 25975 and S. salivarius ATCC 27945. Genome Announc. 5, e00536-17 (2017).
Parks, D. H. et al. GTDB: an ongoing census of bacterial and archaeal diversity through a phylogenetically consistent, rank normalized and complete genome-based taxonomy. Nucleic Acids Res. 50, D785–D794 (2022).
Jain, C., Rodriguez-R, L. M., Phillippy, A. M., Konstantinidis, K. T. & Aluru, S. High throughput ANI analysis of 90K prokaryotic genomes reveals clear species boundaries. Nat. Commun. 2018 91 9, 1–8 (2018).
Ahannach, S. et al. Microbial enrichment and storage for metagenomics of vaginal, skin, and saliva samples. iScience 24, (2021).
Bustin, S. A. et al. The MIQE Guidelines: Minimum Information for Publication of Quantitative Real-Time PCR Experiments. Clin Chem. 55, 611–622 (2009).
Wuyts, S. et al. Large-Scale Phylogenomics of the Lactobacillus casei Group Highlights Taxonomic Inconsistencies and Reveals Novel Clade-Associated Features. mSystems 2, (2017).
De Boeck, I. et al. Randomized, Double-Blind, Placebo-Controlled Trial of a Throat Spray with Selected Lactobacilli in COVID-19 Outpatients. Microbiol. Spectr. 10, (2022).
Van Holm, W. et al. Antimicrobial potential of known and novel probiotics on in vitro periodontitis biofilms. npj Biofilms Microbiomes 2023 91 9, 1–12 (2023).

There is a conflict of interest S.L. received funding from different probiotic companies that were not involved in this research. L.D. was funded by VLAIO through a Baekeland mandate in collaboration with YUN NV. T.E. is partially funded through an industrial research VLAIO grant not related to this work. The remaining authors have no conflicts of interest to declare.

Supplementaryinformation.docx

Download PDF

Version 1

posted

You are reading this latest preprint version

Pangenome-based design of strain-specific primers allows the specific monitoring of engraftment in different habitats

Status:

Version 1

Abstract

Figures

Introduction

Material and methods

Results

Discussion

Conclusion

Declarations

Competing interests

Funding

Acknowledgements

Data Availability

References

Additional Declarations

Supplementary Files

Status:

Version 1