The Burkholderia cepacia Complex (BCC) bacteria are a group of incredibly diverse but closely related species of Gram-negative, aerobic, non-fermentative bacteria that belong to the β subclass of the phylum Proteobacteria (Lessie et al., 1996; Mahenthiralingam et al., 2008; Tavares et al., 2020). BCC bacteria are spread worldwide and are proven to proliferate or survive in oligotrophic environments for an extended period (Mahenthiralingam et al., 2000; Ahn et al., 2014, 2019). They are also considered significant opportunistic human pathogens that produce a variety of potential virulence factors (Mahenthiralingam et al., 2005) and could cause devastating infections in patients with bacteremia, septic arthritis, bacterial peritonitis, and cystic fibrosis that lead to high mortality in infected and immunocompromised patients (Mahenthiralingam et al., 2008; Gautam et al., 2011; Datta et al., 2020; Wang and Cissé, 2020). BCC bacteria that could exhibit multiple antibiotic resistances are also essential pathogens of hospital-acquired infections (Decicco et al., 2010; Sousa et al., 2011; Sfeir, 2018).

Due to their nature and broad distribution, BCC bacteria have been proposed as objectionable organisms in warning letters, safety alerts, or recalls by the United States Food and Drug Administration. Global recalls related to BCC bacteria contamination were found frequently in water-based products, especially in pharmaceutical products, personal care products, and disinfectants (Kuhn et al., 1982; Balkhy et al., 2005; De Volder and Teves, 2020; Tavares et al., 2020; Zou et al., 2020; Du et al., 2021). BCC isolates, especially B. cepacia, were identified in 22% of those recalls as the leading cause of microorganisms in non-sterile product contamination from 2004 to 2011 (Jimenez, 2007; Sutton and Jimenez, 2012). To date, 24 species have been identified and classified as the Burkholderia cepacia Complex (Coenye et al., 2001c; Sfeir, 2018), of which nine genomovars (I–IX) are commonly isolated from water-based products or clinical samples. They are known as B. cepacia (I), B. multivorans (II), B. cenocepacia (III), B. stabilis (IV), B. vietnamiensis (V), B. dolosa (VI), B. ambifaria (VII), B. anthina (VIII) and B. pyrrocinia (IX; Vandamme et al., 1997; Balandreau, 2001; Coenye et al., 2001c; Vanlaere et al., 2008; Vandamme and Dawyndt, 2011; Depoorter et al., 2020).

In 2020, the United States Pharmacopoeia (USP 43) published the official general chapter <60> to describe the test in non-sterile pharmaceutical products by a culture-based method. Indeed, this new chapter has gained a broad interest in recent years and is driving progress in quality risk management in pharmaceutical industries. However, this culture-based method is time-consuming and laborious, costing more than 4 to 7 days for a laboratory to acquire positive or negative results (Steinmetz et al., 1999; Amornchai et al., 2007). Therefore, a rapid, sensitive, accurate, and affordable detection technology for BCC bacteria is urgently needed in scientific research, survey, industries, and regulatory authorities.

Rapid molecular methods targeting specific genes provide tremendous advantages over conventional approaches (Attia et al., 2016; Sfeir, 2018). Polymerase chain reaction (PCR) is considered one of the most valuable techniques for rapid detection in decades. However, it takes ineffective time for dramatic temperature changes between each amplification cycle. Several isothermal amplification methods for nucleic acid have been introduced in forensics, animal and human pathogens over the last 10 years (Soroka and Wasowicz, 2021), such as loop-mediated isothermal amplification (LAMP), nucleic acid sequence-based amplification (NASBA), helicase-dependent amplification (HDA) and rolling circle amplification (RCA; Asiello and Baeumner, 2011; Zanoli and Spoto, 2013). In particular, the recombinase polymerase amplification (RPA) approach is carried out at constant room temperature between 38 and 42°C, which allows no need for a heating and cooling process. Therefore, it offers a simple reaction system with high sensitivity and selectivity compared to other methods (Piepenburg et al., 2006; Li et al., 2018; Lobato and O'Sullivan, 2018). These characteristics offer a tremendous molecular tool for daily primary screening (Craw and Balachandran, 2012; Zanoli and Spoto, 2013).

The 16S rDNA gene, recA gene, fur gene, and hisA gene are commonly used gene targets for DNA typing and identification of Burkholderia species due to their genetic polymorphism (Campbell et al., 1995; Karlin et al., 1995; Eisen, 1996; Vandamme et al., 1996; Vermis et al., 2002; Lynch and Dennis, 2008; Papaleo et al., 2010; Devanga Ragupathi and Veeraraghavan, 2019). With the continuous discovery of novel species, unknown BCC bacteria tested by assays based on non-specific genes may cause misleading results (LiPuma et al., 1999). It is unlikely to cover all potential BCC strains based on highly polymorphic genes with one screening method. The lack of specific genes with fewer variants has significantly influenced the development of rapid detection for BCC bacteria in the pharmaceutical and cosmetics industries. With the robust growth of whole genome sequencing and bioinformatics analysis, the resources of microbial proteomes and genomes are available for high throughput comparative approaches to mine molecular biomarkers in a considerably short time, which are more economical and convenient for practical uses (Yu et al., 2011; Chen et al., 2019; Zhang et al., 2020; Srijuntongsiri et al., 2022). Therefore, it is necessary and possible to find a suitable molecular target for the complicated BCC bacteria using new mining strategies to avoid most of the deficiencies of false negative and false positive results (Yu et al., 2011).

In this study, we aimed to mine and discover a BCC-specific molecular target through an automatic high-throughput proteomic comparison and screening framework. After in silico evaluation with our database, the core conserved sequence of the protein-coding gene was introduced as a suitable biomarker of BCC bacteria for rapid real-time PRA assay. Furthermore, the screening testing method was developed and validated using artificially contaminated non-sterile pharmaceutical and personal care products.

The taxonomy of the genus Burkholderia is extraordinarily versatile and diverse, and it contains more than 120 species classified and separated from the genus Pseudomonas (Yabuuchi et al., 1992; Coenye et al., 2001c; Scoffone and Trespidi, 2021). BCC bacteria are genetically distinct but phenotypically similar bacteria within the genus Burkholderia (Yabuuchi et al., 1992; Mahenthiralingam et al., 2005). Because of the multi-drug resistance and a remarkable ability to survive in solutions (Geftic et al., 1979), BCC bacteria are at significant risk of infection in susceptible or immunosuppressed populations (Manu-Tawiah et al., 2001). BCC strains are frequently isolated in water-based products, especially in oral liquid solutions, nasal sprays, purified water, baby wipes, lotions, hand soaps, and mouthwashes (Kutty et al., 2007; Sutton and Jimenez, 2012; Marquez et al., 2017; Shaban et al., 2017; Solaimalai et al., 2019). Therefore, National drug regulatory authorities worldwide require manufacturers to monitor those objective microorganisms during production for microbiological risk assessment. However, due to the minor biochemical phenotypic differences, culture-based methods may give problematic detection and identification results with neither sensitivity nor specificity (Devanga Ragupathi and Veeraraghavan, 2019).

Molecular-based screening techniques present multiple advantages in rapid pathogen detection, such as rapidness, accuracy, and high sensitivity. It is a notable and influential improvement to the traditional methods in national or international standards. A well-studied specific molecular target is fundamental for methodology design. Evidence from DNA analysis showed that the sequence similarity within BCC species varies from 30 to 60% (Vandamme et al., 1997; Coenye et al., 2001a,b). Finding an ideal molecular detection target to cover all identified BCC bacteria is time-consuming, laborious, and frustrating.

Those commonly known conserved housekeeping genes are generally helpful for differentiating BCC species using various PCR, sequencing, or multi-locus sequence typing schemes (Coenye and Vandamme, 2003; Baldwin et al., 2005; Mahenthiralingam et al., 2008). For example, the recA gene shows 94 to 95% similarity between the genomovars of BCC bacteria (Mahenthiralingam et al., 2000), and species-level resolutions are efficiently achieved through the recA gene analysis for some problematic BCC strains (Cesarini et al., 2009; Vanlaere et al., 2009). In similar, the rpsU gene (Frickmann et al., 2014), fur gene (Lynch and Dennis, 2008), opcL gene (Plesa et al., 2004), and hisA gene (Papaleo et al., 2010) are well achieved and applied in the literature. However, those housekeeping genes are found through a long-term selection and knowledge accumulation of gene or protein function studies in the literature by biological experts, which may require more than half a century (Lederberg and Tatum, 1946; Smith, 2012). The specific sequences as biomarkers with cell active functions for daily practice remain limited.

The bioinformatics comparison allows us to virtually validate and assess the characterization and performance of assays on a broader scale and in an easier way, especially for comparing newly emerged or re-classified species for an instant. In our study, the genomic data of newly reported BCC species are also introduced, such as the novel species B. aenigmatica (Depoorter et al., 2020), B. paludis (Shion et al., 2016), and B. catarinensis (Bach et al., 2017). Our automatic screening and evaluation framework is robust to mine biomarkers related to a particular function for BCC bacteria with lower synonymous mutations. The preprotein translocase subunit, protein SecY, is essential in forming an entirely proteinaceous pathway through which secreted proteins pass during membrane transit (Joly and Wickner, 1993; Driessen et al., 1998). Although some closely related non-BCC species and other bacteria contain the secY gene in their genomes according to bio-information analysis, such as B. mallei, B. gladioli, B. plantarii, B. pseudomallei, B. glumae, Paraburkholderia rhizosphaerae, Paraburkholderia humisilvae, Paraburkholderia hospital, and Paraburkholderia steynii, evolutionary analysis shows that their secY gene sequences differ significantly from those of BCC strains. Therefore, it demonstrates that the secY gene specific to BCC bacteria can distinguish BCC bacteria at a complex level and can assist in identifying new species of BCC bacteria.

Several previous sensitive molecular detection methods are based on the principles of half-nest PCR (Attia et al., 2016), real-time PCR (Jimenez et al., 2018), and lateral flow PRA (Peng et al., 2019) for detecting one or more BCC species. The primers or probes used in any detection method are suffered from limited experimental tests, and none of them are validated by all known species of BCC bacteria (Campbell 3rd et al., 1995; Jimenez et al., 2018). Insufficient validation will mislead industries and consumers with an uncertain risk of false negative results in process monitoring and control. Therefore, continuously re-validating these developed DNA-based methods to keep them up-to-date with increasing and newly emerged biodiversity should be considered (Coenye et al., 2001c). Although some molecular methods have been introduced and combined with DNA sequencing for accurate identification, confirmation, and discrimination among BCC bacteria (Mahenthiralingam et al., 2000; Payne et al., 2005; Vonberg et al., 2006; Pimentel et al., 2007; Martinucci et al., 2016), they are unsuitable for primary screening in daily practice.

In this study, we developed a novel BCC-specific biomarker mining strategy for the scientific community to screen reliable targets for microorganisms in metadata. We also provided an isothermal real-time RPA assay targeting the secY gene to rapidly screen suspicious BCC strains. The method has high specificity, accuracy, sensitivity, and low requirements for instrumentation. It is suitable for the risk control of BCC bacteria contamination in pharmaceutical and personal care products.

The original contributions presented in the study are included in the article/Supplementary material, further inquiries can be directed to the corresponding author.

YLF: Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Resources, Validation, Visualization, Writing – original draft. SJW: Formal analysis, Validation, Writing – original draft. MS: Resources, Writing – original draft. LLZ: Methodology, Software, Writing – review & editing. CZL: Validation, Writing – review & editing. YY: Validation, Writing – review & editing. SJY: Resources, Writing – review & editing. MCY: Funding acquisition, Project administration, Resources, Supervision, Writing – review & editing. All authors contributed to the article and approved the submitted version.

The author(s) declare financial support was received for the research, authorship, and/or publication of this article. This work was supported by the Science and Technology Commission of Shanghai Municipality (22142201600 and 20DZ2293600), the National Key R&D Program of China (no. 2018YFC1603901), the Standard Improvement Project of Chinese Pharmacopoeia Commission (2022Y21), and the Scientific Research Project of Zhejiang Medical Products Administration (2023016).

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.