The lone star tick (Amblyomma americanum) is an aggressive and generalist hematophagous species known as a vector of a variety of viral (Heartland virus) and bacterial pathogens (Ehrlichia chaffeensis, Ehrlichia ewingii, Francisella tularensis, and Borrelia lonestari) and alpha-gal syndrome (AGS) (Crispell et al., 2019; Sharma et al., 2021). Microorganisms that occupy an arthropod tick vector are collectively called the tick microbiome; however, the collection of commensals, symbiotic, and pathogenic microbes associated with ticks is more precisely termed the “pathobiome”. Past investigations of the pathobiome of the lone star tick primarily focused on field-collected tick samples (nymphs or adults) using the 16S ribosomal RNA (16S rRNA) sequencing approach (Menchaca et al., 2013; Williams-Newkirk et al., 2014; Fryxell and DeBruyn, 2016; Brinkerhoff et al., 2020). Microbes coexisting association with known pathogens within the ticks might influence pathogen infection and transmission. For instance, rickettsial endosymbionts are thought to alter the transmission of other rickettsial pathogens, as reported by the inverse relationship between the infection prevalence of Rickettsia rickettsii (pathogen) and Rickettsia peacockii (symbiont) in Dermacentor andersoni (Burgdorfer et al., 1983). Similarly, the presence of Coxiella-related symbionts in the salivary glands (SGs) of Am. americanum ticks impairs the transmission of E. chaffeensis (Klyachko et al., 2007). In addition to symbionts, ticks maintain a natural bacterial flora predominantly that comprised bacteria from the Proteobacteria, Firmicutes, and Bacteroides phyla (Budachetri et al., 2014; Narasimhan et al., 2014; Budachetri et al., 2016; Budachetri et al., 2017; Karim et al., 2017; Adegoke et al., 2020), which have also been implicated in interference with tick pathogen infection. For example, Ixodes scapularis ticks hatched and raised in a sterile environment showed an altered microbiota, impaired gut integrity, and a reduced ability for Borrelia burgdorferi to colonize (Narasimhan et al., 2014). As demonstrated in other arthropod vectors (Cirimotich et al., 2011), altering the tick microbiome may also result in a modulated immune response that could also interfere with pathogen infection and transmission. As microbiomes are an integral part of tick biology, a detailed understanding of (1) microbiome composition, (2) stability, and (3) tripartite interaction with the tick, pathogen, and host will facilitate the discovery of new interventional strategies to limit tick-borne diseases (TBDs).

Most microbiome studies have been performed in whole ticks except in a few studies where tick tissues have been used (Andreotti et al., 2011; Budachetri et al., 2014; Narasimhan et al., 2014; Qiu et al., 2014; Budachetri et al., 2018). Microbiome profiling using whole ticks has limitations, such as difficulty profiling a specific microbial niche at the tissue level (Schabereiter-Gurtner et al., 2003). Profiling of the microbiome associated with vital tissues is necessary for deriving functional inferences of tick biology, TBDs, and interactions (e.g., tick-pathogen, pathogen-endosymbiont) (Narasimhan and Fikrig, 2015). Furthermore, exoskeleton-associated environmental contaminants may not be differentiated from bona fide tick tissue–associated bacteria (Narasimhan et al., 2014). Previous studies have also shed light on the prevalence of certain endosymbionts within tick samples. A previously published study showed that approximately one-third of the tick species examined either lacked Coxiella-like endosymbiont (CLE) or harbored lower frequencies than expected of an obligate endosymbiont (Duron et al., 2014; Duron et al., 2015). Tick species deficient in CLE are scattered among major tick families and genera, suggesting its frequent and repeated loss during tick evolution (Duron et al., 2015). The possible role of CLE in nutrition, osmoregulation or excretion, synthesis of amino acids, and major B vitamins has been proposed (Klyachko et al., 2007; Machado-Ferreira et al., 2011; Lalzar et al., 2014). Exclusion symbiosis and acquisition of Francisella-like endosymbiont (FLE) and Rickettsia spp. in ticks could also be speculated as a more efficient alternative for synthesizing B₇ and B₉ vitamins (Hunter et al., 2015; Duron et al., 2017) because of the reduced genome of CLE (Gerhart et al., 2016). Other possible factors could be geography, lateral gene transfer, or tick feeding on a shared vertebrate host. There is a gap in our understanding of the ecological factors that could facilitate horizontal transfer (HT) of endosymbionts among tick species. The examination of internal tick organs including SGs has indicated a high concentration of CLEs (Klyachko et al., 2007; Qiu et al., 2014); therefore, feeding on a shared vertebrate host could be an important determinant for HT.

In this study, the microbiome of the lone star tick was determined by using the 16S rRNA sequencing approach. The aims of this study were (a) to determine the composition of the microbiome across the life cycle in unfed (UF) and blood-fed immature and mature tick developmental stages, (b) to assess the stability of the microbiome in UF and blood-fed individually dissected tissues (SGs, midgut [MG], and ovary), (c) to assess the microbiome composition of field-collected ticks and, (d) to test and validate the hypothesis of acquisition of FLE at the expense of CLE.

The consistent occurrence of Francisella in both our laboratory-grown and field-collected tick samples suggests its systemic association with ticks. Our analysis has assigned abundant sequences to Francisella (70%–75%), followed by Midichloria (14%–17%) and Rickettsia (6%–9%), excluding UF-OV tissue. The richness of Francisella in immature and mature life developmental stages and individually dissected tissues (except for the UF-OV) supports the notion that it sustains an obligatory association with all feeding stages and outcompetes other bacteria. The relative abundance of Midichloria and Rickettsia was considerably lower than Francisella, signifying their link with the tick as facultative endosymbionts. However, competitive interactions among endosymbionts, pathogenic microbes, have been shown to increase virulence of pathogenic microbes, and complications in vertical transmission were suggested to influence obligatory or facultative endosymbionts in ticks (Mira and Moran, 2002). Consistent dominance of Francisella in tick life stages and tissues suggests its significance in Am. americanum development and other biological processes. Perhaps the Am. americanum ticks are sheltering Francisella by creating a favorable niche within various tissues vector for the colonization, propagation, and trafficking of Francisella at the expense of other endosymbionts and microbial communities. The tick vector sequestered vital nutrients from Francisella for development and in return provided the niche for Francisella to survive and thrive within the tick vector. Synergistic and antagonistic interactions among pathogenic and nonpathogenic rickettsial species could also play an important role in the prevalence of microbes within the tick vector. The interspecies competition among Midichloria and Rickettsial species could also cause their decreased prevalence within the arthropod vector. For example, dominant R. peacockii blocked vertical transmission of R. rickettsii in the ovary of D. andersoni (Burgdorfer et al., 1981; Niebylski et al., 1997). The dominance of FLEs in Am. americanum might be location-specific, as demonstrated for Coxiella and Rickettsia (Institute of Medicine (US) Committee on Lyme Disease and Other Tick-Borne Diseases, 2011; Ponnusamy et al., 2014) and of fundamental importance because of its occurrence in several other tick species such as Dermacentor occidentalis; Dermacentor albipictus (Packard); Dermacentor nitens; D. andersoni; Dermacentor hunteri; and Dermacentor variabilis (Niebylski et al., 1996; Sun et al., 2000; Scoles, 2004; Institute of Medicine (US) Committee on Lyme Disease and Other Tick-Borne Diseases, 2011; Ponnusamy et al., 2014). Still, the nature of their symbiotic relationship is yet to be explained. In this study, field-collected tick tissues from Illinois, Delaware, and Maryland also demonstrated the presence and dominance of FLEs, suggesting their establishment in field populations of Am. americanum ticks, although their presence seems much prominent in laboratory-grown colonies ( Figure 10 ).

We found differences in the diversity metrics between laboratory-raised and field-collected ticks colonies. The observed_OTUs metrics, a quantitative measure of the total number of bacteria, showed that field-collected ticks have approximately 100 folds more bacteria than laboratory-raised ticks irrespective of sample type ( Figure S1 ). We also found that field-collected ticks have a higher diverse microbiome regardless of the tick location than laboratory-raised tick colonies. To further strengthen this observation, despite the fact Francisellaceae was the dominant bacterial family in both tick groups, field-collected ticks also harbored other bacterial families at a relatively high abundance contributing to a more diverse microbiome. This finding implies that the tick environment plays a significant role in enriching certain bacteria members of the tick microbiome. The laboratory-raised ticks have been maintained under strict laboratory conditions with limited vertebrate hosts and have less richness in their microbial communities. It seems possible that the differences in the diversity observed among life stages of field-collected and laboratory-raised ticks are due solely to the continuous exposure to environmental bacteria encountered by questing ticks in the field. A closer look at bacterial profile in Figures 2 and 3 clearly indicates that with the exception of the presence of Francisellaceae in both laboratory-raised and field-collected ticks, several bacterial family were exclusively detected in field-collected ticks such as Propionibacteriaceae, Nocardiaceae, Dermabacteraceae, Rickettsiales, and Staphylococcaceae, all of which are known to be environmental residents in soils and are predominant on both human and animal skin. These findings provide further support for the hypothesis that the tick microbiome may play a contributing factor to the incidence of the alpha-gal phenomenon across different region in the United States.

An earlier study conducted on field-collected Am. americanum ticks from Indiana showed approximately 40% of CLEs, followed by ~5% rickettsial endosymbionts (Institute of Medicine (US) Committee on Lyme Disease and Other Tick-Borne Diseases, 2011). A recent microbiome study of Am. americanum from Kansas reported a dominance of endosymbiotic genera belonging to Coxiella and Rickettsia (>97%) (Maldonado-Ruiz et al., 2021). However, ticks collected from North Carolina harbored a preponderance of Rickettsia-like endosymbionts over Coxiella (Ponnusamy et al., 2014). In the current study, the most abundant bacterial family is Francisella at the organismal level in both immature and adult tick life stages and individually dissected tissues (excluding UF-OV), suggesting that the microbiome composition of this tick varies with geographical distribution. The presence of Coxiella in UF-OV tissue indicates its obligate symbiont nature, and another study has shown its presence in the ovary (to show its maternal transmission) and Malpighian tubules to suggest its possible role in osmoregulation and excretion (Klyachko et al., 2007; Machado-Ferreira et al., 2011; Lalzar et al., 2014). Surprisingly, CLE was detected in neither tick developmental life stages nor tissue samples, except in the UF-OV. CLE, an obligatory endosymbiont, is required for tick survival and reproductive fitness (Zhong et al., 2007). CLE has been implicated for its possible role in synthesizing amino acids and several B vitamins (Klyachko et al., 2007; Machado-Ferreira et al., 2011; Lalzar et al., 2014). Possible reasons for the replacement of CLE in Am. americanum could be the acquisition of FLE and Rickettsiae established as alternative obligate symbionts in some tick species (Duron et al., 2017). All genes required for vitamin B₉ and B₇ biosynthesis are also present in Rickettsia and FLE endosymbionts, respectively. In instances where multiple endosymbionts provide a similar advantage to the host, the presence and maintenance of all endosymbionts are not expected; it does not add additional help to the tick host (Vautrin and Vavre, 2009).

It was a common understanding that hematophagous niches are commonly driven by evolutionarily stable symbiotic interactions in several arthropods (Moran et al., 2008; Wernegreen, 2012). Contrary to that, an elegant study proposed that obligate symbioses are relatively unstable in obligate hematophagous ticks (Duron et al., 2017). In that study, six genera of distinct endosymbionts were present in the castor bean tick (Ixodes ricinus) and the African blue tick (Rhipicephalus decoloratus), but no symbiotic community structure was found fixed and stable across the tick phylogeny (Duron et al., 2017). The success of horizontal and vertical transmission patterns in ticks could also modify symbiotic interactions (Duron et al., 2017). This fact could also explain why there is low evolutionary stability of the symbiosis between Am. americanum ticks and CLE, or CLEs are missing from most of our tick samples in the current study.

CLE follows two different evolutionary strategies. Some CLEs are highly specialized to the tick host from ancient times, followed by codiversification. For example, the Rhipicephalus genus and CLE lineages have codiversified together. The emergence of the Rhipicephalus genus and original CLE infection co-occurred, approximately 14 million years ago (Murrell et al., 2001; Duron et al., 2017; Ben-Yosef et al., 2020). On the other hand, other CLEs appear to be acquired through HT from unrelated or accidental host species. Such a pattern has also been observed in other endosymbionts such as Wolbachia, which occurs frequently in insects, similar to CLE in ticks. A recent study also supports the hypothesis of frequent replacement of obligate symbiont CLE in ticks (Duron et al., 2017). Our data provide strong evidence that CLE replacement by FLE occurs in Am. americanum populations; otherwise, FLEs are rare in ticks and not frequently found in arthropods (Niebylski et al., 1997; Scoles, 2004; Goethert and Telford, 2005; Clayton et al., 2015; Gerhart et al., 2016; Duron et al., 2017).

Earlier studies proposed that among Amblyomma tick species, CLEs have reduced genomes, a feature associated with microbes that are long-term or early evolved endosymbionts (McCutcheon and Moran, 2011). A wide distribution of genetically differentiated strains of CLE across the tick phylogeny also specifies its ancient symbiotic relation with ticks (Duron et al., 2017). Unlike the highly reduced genome of CLEs, FLEs have minimal genome reduction and evolved recently from a mammalian pathogen F. tularensis (Gerhart et al., 2016) and have more superior biosynthetic metabolic capabilities than CLEs. Therefore, it is highly likely that it has replaced ancestral CLEs with reduced functionality in Am. americanum developmental stages and tissue samples, as evident from the data presented here. A recent study has also suggested replacement of endosymbiont CLEs with FLEs in another Amblyomma tick species, the Gulf Coast tick (Amblyomma maculatum), for the same reason (Gerhart et al., 2016). It is also possible that Am. americanum replaced CLE at the expense of functionally important symbiotic genes (via lateral gene transfer), with FLEs that are functionally more efficient than CLEs because CLE has a reduced genome and is less efficient in metabolic and biosynthetic capabilities, including that of B vitamin synthesis. Perhaps, this is the key evolutionary mechanism of how ticks retain their capacity to synthesize vitamin B even without containing endosymbionts. Previously, this pattern has been reported from some filarial nematodes, which demonstrated the ability to live and reproduce without obligate symbionts through lateral gene transfer (McNulty et al., 2010). A close examination of I. scapularis, a tick species deficient in CLE, did not show any sign or evidence of lateral gene transfer (Gulia-Nuss et al., 2016). However, it contained a rickettsial endosymbiont, Rickettsia buchneri, which could synthesize the B₉ vitamin (Hunter et al., 2015). The presence of R. buchneri provides a possible explanation for both the absence of CLE and no evidence of lateral gene transfer. An unexplored ecological pathway area facilitating the HT of endosymbionts among tick species needs to be thoroughly investigated. Cofeeding of different tick species on a shared vertebrate host could be an important determinant. Reports have also shown that the SGs of blood-fed ticks contain high levels of CLE in some tick species (Klyachko et al., 2007; Qiu et al., 2014). The geographic expansion of Am. americanum in new territories, including the southern coastal areas along with the Atlantic coast and the Gulf of Mexico and several northern states in the United States, has been reported and predicted (Molaei et al., 2019; Raghavan et al., 2019). The lone star tick (Am. americanum) and American dog tick (D. variabilis) are historically prevalent in these areas and naturally harbor FLEs (Budachettri et al., 2014; Travanty et al., 2019). A variety of reservoir hosts play an essential role in the maintenance of these endosymbionts. For example, wild rabbits, hare, and muskrats are known reservoirs for Francisella species (Nigrovic and Wingerter, 2008). Hence, a change in host feeding due to new expansion in these areas could be one reason for establishing FLEs in Am. americanum. As cofeeding is reported as one significant factor for exchanging virome among D. variabilis, Am. americanum, and I. scapularis (Tokarz et al., 2014). Thus, the cofeeding of tick species on a shared vertebrate host could serve as an ecological arena for exchanging endosymbionts. Earlier studies have suggested that tick pathogens also colonize with closely related endosymbionts, and these endosymbionts also appear to promote closely related pathogen acquisition and transmission (Bonnet et al., 2017). The FLEs are known to have evolved from the pathogen F. tularensis (Ft); however, unlike virulent Ft, its transmission and virulence in humans are enigmas (Ivanov et al., 2011; Rounds et al., 2012; Wójcik-Fatla et al., 2015). A wide distribution of FLEs in other tick species has been further highlighted by the reports such as >94% of FLE-positives among D. variabilis, D. andersoni, and D. occidentalis ticks in the Western United States (Niebylski et al., 1997; Rounds et al., 2012) and 41% FLE-positives in D. occidentalis ticks from California (Western United States) but without Ft infection (Gurfield et al., 2017). In another earlier study, D. andersoni ticks collected from Oregon and Montana (northwestern United States) showed FLE and Ft accounted for 80% (60% FLEs and 20% Ft) of MG microbiome (Gall et al., 2016), suggesting that genetic similarity of FLE and Ft and geography are both probably contributing to inflating Ft infection rates. More studies are needed to clarify the involvement of FLEs in Ft infection and how factors such as geography and genetic similarity are involved.

Our network analysis showed a strong positive correlation between Rickettsiaceae and Midichloriaceae families in laboratory-raised tick life stages and between Francisellaceae, Rickettsiaceae, and Midichloriaceae in tissues isolated from laboratory-raised ticks. Interestingly, the presence of the Rickettsiaceae family was shown to be positively correlated to Midichloriaceae across different life stages of laboratory-raised tick colonies, indicative of a potential synergistic relationship between bacteria belonging to these two families. This observation was also reported in another study (Budachetri et al., 2018). It showed the Rickettsia parkeri colonization of the tick tissue facilitates replication of the endosymbiont Candidatus Midichloria mitochondrii (CMM) in Am. maculatum ticks. Lejal et al. (2021) also identified a substantial prevalence of CMM in I. ricinus ticks that were positive with bacteria in the Rickettsia genus.

Coxiellaceae is replaced with Francisellaceae in Am. americanum ticks, and this pattern was further strengthened by analyzing the correlations between bacterial families identified in tissues of laboratory-raised tick colonies. While we observed no direct correlation among bacteria in the Coxiellaceae family and Rickettsiaceae, Francisellaceae, and Midichloriaceae, the Coxiellaceae family was positively correlated with Rhizobiaceae and Xanthobacteraceae, both of which were negatively correlated to Rickettsiaceae, Francisellaceae, and Midichloriaceae bacterial families. Both Rhizobiaceae and Xanthobacteraceae are nonresident, opportunistic bacteria that ticks acquire from their environment (Sun et al., 2000; Scoles, 2004; Andreotti et al., 2011; Carpi et al., 2011; Institute of Medicine (US) Committee on Lyme Disease and Other Tick-Borne Diseases, 2011; Ponnusamy et al., 2014; Clayton et al., 2015; Galili, 2016; Gurfield et al., 2017). Lejal and colleagues (2021) detected a significant abundance of a Rhizobiaceae-Multi_1 in Rickettsia-positive I. ricinus ticks, suggesting a completely different observation of our findings that Rhizobiaceae was negatively correlated with the Rickettsiaceae, but positively correlated with Coxiellaceae. This finding could present a potential interaction between known bacterial endosymbionts and possible environmental bacteria transiently acquired by ticks. An understanding of how these environmental microbes change the dynamics of the tick microbiome requires further attention.

The presence of specific microbes in tick SGs is vital in the context of AGS as tick bites are believed to be responsible for causing AGS in humans. The lone star tick possesses alpha-gal antigens in its SG and saliva, which are hypothesized to be prime culprits of AGS (Crispell et al., 2019). The gene that codes for the enzyme α-1,3-galactosyltransferase has not been identified in a tick so far, indicating that the microbiome of the tick SG could be one major contributor of alpha-gal antigens (Sharma and Karim, 2021; Sharma et al., 2021). Several pieces of evidence support this hypothesis. One important tick-borne bacterial pathogen, Anaplasma phagocytophilum, is linked with an increase of alpha-gal antigen (Cabezas-Cruz et al., 2019). Similarly, Francisella species are also reported to be involved in modulating glycosylation genes such as glycosyltransferase and glycosidase in the host (Chacko et al., 2011). Hence, it can be speculated that the establishment of FLE can contribute toward the rise of alpha-gal signatures in ticks. The recent change of guard within Am. americanum (CLE replaced by FLE) points to the potential role of the tick microbiome in the emergence of AGS. The exclusive replacement of the Coxiellaceae with Francisellaceae in both laboratory-raised and field-collected Am. americanum ticks could shed more light on the role played by this tick in the emergence of AGS. Although there are no current studies on when the pathogenic F. tularensis switches to an endosymbiont, Gerhart et al. (2016) argued that an FLE of Am. maculatum recently evolved to becoming an endosymbiont. Similarly, it is worth noting that the emergence of AGS in humans and the incrimination of Am. americanum in inducing this condition are recent development. Understanding whether these two phenomena took place concurrently or simultaneously will significantly fill a huge gap in our understanding of the microbiome–tick vector interaction in the context of AGS.

An elegant recent study conducted on the human gut microbiome reported gut resident microbes from the Enterobacteriaceae family, including Escherichia coli, Pasteurellaceae genera, Haemophilus influenzae, and Lactobacillus species, can exhibit α-1,3-galactosyltransferase activity, which indicates that the presence of an enzyme in these microbes could contribute to the devolvement of an alpha-gal antigen (Montassier et al., 2020). This study also identified specific genes exhibiting α-1,3-galactosyltransferase bacterial sequences in their shotgun sequencing data (Montassier et al., 2020). An earlier study reported microbes from Rhizobiaceae and Caulobacteriaceae families possessing novel lipid A a-(1,1)-GalA transferase gene(rgtF) (Brown et al., 2013). This enzyme could also be a potential source of an α- gal antigen. In this study, we identified microbes from Rhizobiaceae and Caulobacteriaceae family in Am. americanum SG samples. The microbes from these families could contribute to the synthesis of an alpha-gal antigen or overall alpha-gal signature in tick SGs. The tick microbiome possessing an uncharacterized enzyme with a glycoside hydrolase, glycosyltransferase, or similar function as α-1,3-galactosyltransferase, is yet to be investigated.

In this study, a detailed and comprehensive 16S rRNA sequencing revealed a stable microbiome composition in all developmental stages of the lone star tick (Am. americanum), and the traditionally associated CLE was absent in these developmental stages (excluding the UF-OV), thus pointing to a solid niche for vertical transmission to the next generation. Interestingly, the Am. americanum ticks investigated appear to have acquired another tick endosymbiont, FLE, at the expense of CLE. Levels of FLE were significantly higher in laboratory-maintained ticks than field-collected ticks, possibly due to exposure to limited host species used for tick rearing compared with more diverse vertebrate hosts in the wild and/or controlled laboratory conditions compared with field environmental conditions. Historically, CLE is known to benefit its tick host, synthesizing vitamin B and cofactors that aid in osmoregulation, excretion, and reproduction. The possible rationale for acquiring and establishing FLE at the expense of CLE could be the reduced genome of CLE leading to impaired vitamin B synthesis when compared with FLE. FLE coverage in tick life stages was ~70% of total reads, suggesting its significance in tick physiology. An inverse relationship between the number of FLEs and spotted fever group Rickettsia (SFGR) has been demonstrated in D. occidentalis, suggesting that FLEs can interfere with SFGR colonization. It has also been shown that a high abundance of FLEs is positively correlated with the acquisition of pathogen Francisella novicida. Could this positive correlation between FLE and pathogen F. novicida lead to a possible infection of F. novicida in lone star ticks in the east-central United States soon?

The datasets presented in this study can be found in online repositories. The names of the repository/repositories and accession number(s) can be found in the article/ Supplementary Material .

The animal study was reviewed and approved by University of Southern Mississippi IACUC.

SK conceptualized the study. DK, SS, AA, and SK were responsible for data curation and formal analysis. Funding acquisition was done by SK. Investigation was performed by SS, DK, AA, and SK. SS, DK, AA, AK, HT, AL, and SK were responsible for the methodology of the study. SK was responsible for project administration, resources, and supervision. Validation was performed by DK, SS, and SK. Visualization was performed by SS, DK, AA, and SK. DK, SS, AA, and SK were responsible for writing of the original draft. DK, SS, AA, AK, HT, AL, and SK were responsible for writing, review, and editing. All authors contributed to the article and approved the submitted version.

This research was principally supported by USDA NIFA award #2017-67017-26171; the National Institutes of Allergy and Infectious Diseases award, RO1 AI135049; Pakistan-US Science and Technology Cooperation award (US Department of State); the Mississippi INBRE [an institutional Award (IDeA) from the National Institute of General Medical Sciences of the National Institutes of Health under award P20GM103476]. The funders played no role in the study design, data collection and analysis, decision to publish, or preparation of the manuscript.

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.