Glycopeptide Antibiotic Resistance Genes: Distribution and Function in the Producer Actinomycetes

Glycopeptide antibiotics (GPAs) are considered drugs of “last resort” for the treatment of life-threatening infections caused by relevant Gram-positive pathogens (enterococci, staphylococci, and clostridia). Driven by the issue of the never-stopping evolution of bacterial antibiotic resistance, research on GPA biosynthesis and resistance is developing fast in modern “post-genomic” era. It is today widely accepted that resistance mechanisms emerging in pathogens have been acquired from the soil-dwelling antibiotic-producing actinomycetes, which use them to avoid suicide during production, rather than being orchestrated de novo by pathogen bacteria upon continued treatment. Actually, more and more genomes of GPA producers are being unraveled, carrying a broad collection of differently arranged GPA resistance (named van) genes. In the producer actinomycetes, van genes are generally associated with the antibiotic biosynthetic gene clusters (BGCs) deputed to GPA biosynthesis, being probably transferred/arranged together, favoring a possible co-regulation between antibiotic production and self-resistance. GPA BGC-associated van genes have been also found mining public databases of bacterial genomic and metagenomic sequences. Interestingly, some BGCs for antibiotics, seemingly unrelated to GPAs (e.g., feglymycin), carry van gene homologues. Herein, we would like to cover the recent advances on the distribution of GPA resistance genes in genomic and metagenomics datasets related to GPA potential/proved producer microorganisms. A thorough understanding of GPA resistance in the producing microorganisms may prove useful in the future surveillance of emerging mechanisms of resistance to this clinically relevant antibiotic class.

antimicrobial activity and clinical use, Zeng et al., 2016). GPAs inhibit bacterial cell wall synthesis in Gram-positive bacteria by binding to d-alanyl-d-alanine (d-Ala-d-Ala) dipeptide terminus of peptidoglycan (PG) precursors, sequestering the substrate from transpeptidation and transglycosylation reactions in the late extracellular stages of PG cross-linking. Thus, GPA action ultimately results in destabilizing cell wall integrity, causing bacterial cell death (Perkins and Nieto, 1974). Gram-negative microorganisms are intrinsically resistant to GPAs, because of their outer membrane, which prevents these molecules entering into the periplasm. In Gram-positive bacteria, the onset of vancomycin resistance was long-delayed in comparison to other antibiotic classes. The first vancomycinresistant clinical isolate -an Enterococcus faecium strain -was reported in 1987, more than 30 years after the clinical introduction of vancomycin (Leclercq et al., 1988, Miller et al., 2016. Unfortunately, today a vast majority of E. faecium isolates harbor vancomycin resistance genes (van) (Vehreschild et al., 2019). The first vancomycin-resistant S. aureus (VRSA) isolate was reported in 2002 as a result of horizontal gene transfer from resistant enterococci (Bartley, 2002;Weigel et al., 2003); nowadays, 52 VRSA strains have been described worldwide (Cong et al., 2020).
The GPA resistance mechanisms in Gram-positive pathogens were intensively studied starting from the pioneering work published in the 1990s (Arthur et al., 1992(Arthur et al., , 1996. Gram-positive pathogens escape GPA action by reprogramming PG precursor biosynthesis, replacing the terminal d-Ala with d-lactate (d-Alad-Lac) or d-serine (d-Ala-d-Ser), thus reducing the affinity for cellular targets (Arthur et al., 1992(Arthur et al., , 1996Courvalin, 2006). In enterococci, many different GPA-resistant phenotypes have been described according to their van gene operon organization (for a review, see Binda et al., 2014): in vanA, vanB, vanD, and vanM the key ligase determines the replacement of the terminus d-Ala with d-Lac, whereas in vanC, vanE, vanG, vanL, and vanN d-Ala. The d-Ala-d-Lac-type operons are located either on plasmids or on chromosomes, whereas the d-Ala-d-Ser-type ones are exclusively on the bacterial chromosome, except the case of vanN found on a plasmid in E. faecium. Operon expression could be inducible by GPAs (vanA, vanB, vanG, vanE, vanL, and vanM) or constitutive (vanC, vanD, and vanN) (Reynolds and Courvalin, 2005;Depardieu et al., 2007;Binda et al., 2014). The most clinically relevant manifestation of GPA resistance occurs in VanA enterococci and staphylococci, and in VanB enterococci. The first group is highly resistant to both vancomycin and teicoplanin, whereas the second group only to vancomycin. In both of them, resistance is mediated by the GPA-induced expression of the transposone-located vanHAX gene operon under the transcriptional control of the VanR/VanS two-component system (TCS). VanS is a membrane-associated sensor that in VanA bacteria is activated by the presence of either vancomycin or teicoplanin, whereas in VanB it is activated only by vancomycin. Consequently, VanB enterococci are sensitive to teicoplanin (Arthur et al., 1997(Arthur et al., , 1999Arthur and Quintiliani, 2001). Activated VanS transfers a phosphoryl group to VanR, which is the response regulator that controls the co-transcription of the vanH, vanA, vanX, and vanY genes (Wright et al., 1993;Arthur et al., 1997Arthur et al., , 1999Arthur and Quintilliani, 2001). VanH is a dehydrogenase that reduces pyruvate to d-lactate; VanA is the key ligase that catalyzes the formation of the d-Alad-Lac resistant depsipeptide (Bugg et al., 1991;Arthur et al., 1992); VanX is a d,d-dipeptidase, which removes the intracellular pool of d-Ala-d-Ala produced by the native enterococcal ligase, ensuring that d-Ala-d-Lac is incorporated into PG precursors (Reynolds et al., 1994;Wu et al., 1995); and finally VanY has an ancillary role as a d,d-carboxypeptidase cleaving the last d-Ala from the residual pentapeptide PG precursors terminating in d-Ala-d-Ala (Arthur et al., 1998). Among the d-Ala-d-Sertype operons, the better investigated was the vanC. It encodes for a racemase (VanT) that converts l-Ser to d-Ser, a ligase (VanC) that synthesizes d-Ala-d-Ser, and a bi-functional d,d-dipeptidase/d,d-carboxypeptidase (VanXYc) that cleaves the residual pools of d-Ala-d-Ala (Billot-Klein et al., 1994;Reynolds and Courvalin, 2005). In VanC phenotype, the TCS VanRcSc is located downstream the operon, but the resistance is constitutive due to mutations in the sensor VanSc (Healy et al., 2000;Hong et al., 2008;Koteva et al., 2010). VanC enterococci are intrinsically resistant to low levels of vancomycin, although they remain sensitive to teicoplanin.
Additional variants of these van gene operons were found in other Gram-positive pathogens including Listeria spp., streptococci, clostridia (Biavasco et al., 1996;Poyart et al., 1997;Peltier et al., 2013), and also in nonpathogenic Gram-positives, including Bacillus circulans, Oerskovia spp., Corynebacterium spp., and Streptomyces coelicolor (Power et al., 1995;Fontana et al., 1997;Hong et al., 2004). A novel vancomycin vanF operon (vanY F Z F H F FX F ) was described in Paenibacillus popilliae, an environmental bacteria used as biopesticide to counteract beetle larvae that caused milky disease in Japan (Patel et al., 2000;Ahmed and Baptiste, 2018). The dissemination of GPA resistance more recently reached zoonotic pathogens such as the emergent Streptococcus suis, where the low level of vancomycinresistance is due to the presence of a vanG-like operon (Huang et al., 2018). Herein, we focus our attention on van genes distribution and function in the GPA-producing actinomycetes, which are considered the putative primary source of the variety of GPA-resistant determinants occurring in environmental bacteria and pathogens (Marshall et al., 1998;Beltrametti et al., 2007;Marcone et al., 2010Marcone et al., , 2014Schäberle et al., 2011).

UPDATING THE GLYCOPEPTIDE RESISTANCE PARADIGM FOR THE GPA-PRODUCING STRAINS: VAN GENES AND THEIR ORGANIZATION IN KNOWN AND PUTATIVE GPA BGCs
Actinomycetes are Gram-positive soil-dwelling bacteria, which produce about two-thirds of the naturally derived antibiotics with clinical use (Bérdy, 2012;Barka et al., 2015), including GPAs (Nicolaou et al., 1999). Clinically relevant GPAs are produced by Amycolatopsis orientalis (vancomycin), Actinoplanes Frontiers in Microbiology | www.frontiersin.org teichomyceticus (teicoplanin), and Nonomuraea gerenzanensis (dalbavancin precursor -A40926) (Zeng et al., 2016;Marcone et al., 2018). GPA producers require self-resistance mechanisms to avoid suicide during antibiotic production and, like in pathogens, such resistance is due to van genes, whose description dates back to the end of the 1990s, one decade later than in pathogens (Marshall et al., 1997(Marshall et al., , 1998. Sequence and operon structure similarities of van genes between pathogens and GPA-producers are significant (Hong et al., 2008;Binda et al., 2014). The intriguing aspect is that in GPA producers, van genes are usually located within the GPA biosynthetic gene clusters (BGCs) deputed to the antibiotic biosynthesis (Pootoolal et al., 2002;Beltrametti et al., 2007;Marcone et al., 2010Marcone et al., , 2014Schäberle et al., 2011). In the last two decades, multiple novel GPA BGCs from actinomycetes were sequenced and annotated, and each of them (with few exceptions, see below) contains van genes (Figure 1).
GPAs are also produced by few Streptomyces species (Figure 1). Interestingly, functional van genes were also found in S. coelicolor, which is not a GPA producer (Hong et al., 2004). A47934 BGC from Streptomyces toyocaensis NRRL 15009 contains vanH St A St X St and vanR St S St operons, together with staO and staP orthologues to S. coelicolor vanJ and vanK, respectively (Pootoolal et al., 2002). Pekiskomycin BGC from Streptomyces sp. WAC1420 contains vanY, vanJ as well as vanHAX and vanRS homologues, but pekiskomycin BGC from Streptomyces sp. WAC4229 lacks vanRS homologues (Thaker et al., 2013). No homologues of van genes were found in complestatin (type V GPA) BGC from S. lavendulae SANK 60477 (?) (Chiu et al., 2001), although this antibiotic possesses a moderate antibacterial activity. However, complestatin was shown to inhibit the fatty acid biosynthesis in Gram-positive bacteria (Kwon et al., 2015), therefore the producer may require no cell wall remodeling for complestatin self-resistance. Finally, feglymycin BGC from Streptomyces sp. DSM 11171 (Figure 1) encodes for a 13-mer peptide antibiotic acting on bacterial cell wall biosynthesis by inhibiting MurA and MurC. Albeit the structure and the mode of action of feglymycin differs from the ones of GPAs, feglymycin BGC shares a high level of similarity with GPA BGCs (Gonsior et al., 2015;Yushchuk et al., 2020a), including the presence of vanRS-like genes -fegM and fegN.

UPDATING ON WHAT IS KNOWN ABOUT THE IN VIVO FUNCTION OF VAN GENES IN GPA-PRODUCING STRAINS
Although van genes were found in multiple GPA BGCs, only for few of them the function was experimentally proven. Balhimycin resistance in Am. balhimycina is likely the most deeply investigated model among GPA producers (Figure 2A). vanH Ab A Ab X Ab , that is located outside the BGC (Figure 1), was shown to be constitutively expressed through all the periods of growth and during balhimycin production (Schäberle et al., 2011). Deletion of vanH Ab A Ab X Ab genes makes Am. balhimycina significantly more sensitive to its own product, decreasing its MIC from 5 to 0.25 mg/ml, and causing an earlier expression of the BGC-situated vanY Ab (Frasch et al., 2015). However, vanY Ab itself does not play a decisive role in GPA-resistance since its deletion did not alter the GPA resistance phenotype Frontiers in Microbiology | www.frontiersin.org (Frasch et al., 2015). Double vanH Ab A Ab X Ab and vanY Ab knocked-out mutants showed the same GPA resistance phenotype as ΔvanH Ab A Ab X Ab . PG precursors ending in d-Ala-d-Lac were still found in the single ΔvanH Ab A Ab X Ab and in the double ΔvanH Ab A Ab X Ab ΔvanY Ab mutants together with d-Ala-d-Ala ending PG precursors and tetrapeptides (Frasch et al., 2015). The residual GPA resistance in these mutants is probably due to an accessory Ddl1 Ab , a putative d-Ala-d-Lac ligase encoded in the genome of Am. balhimycina, which shares 72% of amino acid sequence identity with VanA Ab (Frasch et al., 2015). Ddl1 Ab might add d-Lac to the tetrapeptide PG precursors generated by the d,d-carboxypetidase VanY Ab (although the presence of some other d-Ala-d-Ala carboxypeptidases encoded in the genome cannot be completely ruled out considering the resistant phenotype of the ΔvanY Ab mutant). In the absence of VanH Ab , d-Lac for this reaction is probably obtained from the primary metabolic pool. Expression of vanH Ab A Ab X Ab was demonstrated to be independent from the BGC-situated regulator vlnR Ab (Kilian et al., 2016). However, VlnR Ab is important for the activation of the BGC-situated vanY Ab expression (Kilian et al., 2016). Heterologous expression of vlnR Ab S Ab in S. coelicolor ΔvanRS mutants indicated that both VlnR Ab and VlnS Ab are active and able to replace their counterparts VanR and VanS, which in S. coelicolor control the expression of vanHAX genes in response to vancomycin (Hong et al., 2004), restoring resistance to both balhimycin and teicoplanin in the complemented strains (Kilian et al., 2016). Overall, it seems that the BGC-situated vlnR Ab S Ab -vanY Ab regulatory circuit is functional, but does not play a major role in balhimycin resistance, which is mostly determined by vanH Ab A Ab X Ab expression. It would be interesting to test GPA resistance in ddl1 Ab knocked-out mutant generated in Am. balhimycina ΔvanH Ab A Ab X Ab ΔvanY Ab to better understand the role of this accessory ligase and its connection with the d,d-carboxypetidase activity of VanY Ab (or of some other stillunknown carboxypeptidases). Differently from Am. balhimycina, in Act. teichomyceticus vanHAX orthologues -tei7-6-5 -are located within the tei BGC together with vanRS orthologues -tei2-3 ( Figure 2B). tei7-6-5 expression determines the production of PG precursors ending in d-Ala-d-Lac, conferring a GPA-resistant phenotype to Act. teichomyceticus (Beltrametti et al., 2007;Binda et al., 2018). Interestingly, the expression of tei7-6-5 operon is constant during the growth curve and in teicoplanin production conditions (Beltrametti et al., 2007;Yushchuk et al., 2019) and the VanX d,d-dipeptidase activity was detectable in cellular extracts independently from the addition of teicoplanin . One probable reason for the constitutive expression of tei7-6-5 is the non-inducibility of the sensor histidine kinase Tei3, due to its point mutations previously known to confer a constitutive kinase activity to S. coelicolor VanS (Beltrametti et al., 2007). Also, the expression of vanRS orthologues -tei2-3 -was also found constitutive under teicoplanin production conditions (Yushchuk et al., 2019) and these genes are co-expressed with tei4 -coding for a dehydrofolate reductase with no obvious role in teicoplaninresistance (Yushchuk et al., 2020b). Moreover, tei2-3-4 expression is independent from tei cluster-encoded transcriptional regulators -Tei15* and Tei16* (Yushchuk et al., 2019(Yushchuk et al., , 2020b. Constitutive expression of tei2-3-4 could be granted by tei2 promoter, which was shown to be highly active in Act. teichomyceticus, starting from the very early stage of spore germination (Yushchuk et al., 2020b). More investigations are required for a complete understanding of teicoplaninresistance in Act. teichomyceticus. Study of the Tei3 properties is among the most interesting tasks.
N. gerenzanensis (Lo Grasso et al., 2015). Unfortunately, the GPA resistance phenotype of this mutant was not described. On the other side, transcriptional analysis of dbv genes indicated that the expression of dbv5-6-7 and dbv23-22 operons is rather constitutive (Alduina et al., 2007). Although the presence of other GPA-sensitive TCS beyond the borders of A40926 BGC cannot be ruled out, role of Dbv6 and Dbv22 in N. gerenzanensis A40926 self-resistance merits further investigations.
Finally, S. toyocaensis possesses, perhaps, the most straightforward resistance mechanism among all the investigated GPA producers (Figure 2D), reminding the situation in S. coelicolor (Hong et al., 2004). The BGC-located vanH St A St X St operon was shown to be crucial for A47934 resistance and vanA St knockout made S. toyocaensis completely sensitive to A47934 (Pootoolal et al., 2002). At the same time, vanHAXgenes from the vancomycin producer Am. orientalis C329.2 were able to restore A47934 resistance phenotype in the knocked-out mutant (Pootoolal et al., 2002). Functions of VanR St and VanS St (both present in the A47934 BGC, Figures 1, 2D) were also studied in detail, showing that VanS St has a remarkable specificity for A47934 and it is unable to sense teicoplanin or vancomycin (Koteva et al., 2010;Novotna et al., 2016). Moreover, also the interaction between VanR St and VanS St was found to be very specific, since VanR St could not be phosphorylated by a non-cognate sensor-histidine kinase (Novotna et al., 2016). The roles (if there are any) of staP and staO (orthologues of S. coelicolor vanK and vanJ) in S. toyocaensis A47934 selfresistance were not investigated, thus the importance of these auxiliary resistance genes remains to be proved.

A B
C D FIGURE 2 | Knowledge-based conceptual schemes for the GPA self-resistance mechanisms and their regulation in model GPA-producers: (A) Am. balhimycina, the producer of balhimycin, (B) Act. teichomyceticus, the producer of teicoplanin, (C) N. gerenzanensis, the producer of A40926, and (D) S. toyocaensis, the producer of A47934. Please refer to main text for more details on genes involved in GPA self-resistance. Please note that tei4 in Act. teichomyceticus, dbv5 and dbv23 in N. gerenzanensis, and murX in S. toyocaensis, colored in light gray, are not van gene orthologues, but they are reported since they are close to van genes or part of their operons.
Frontiers in Microbiology | www.frontiersin.org OUTLOOK Soil GPA producers are considered the putative source of GPA resistance determinants, which might had been recruited and differently combined in pathogens. The goal of this mini review is to update the knowledge on the occurrence and role of van genes in producing microorganisms. It emerges that more in silico, in vitro, and in vivo investigations on their function and regulation are required to shed light on the intriguing issue of their origin and role. Overall, a detailed phylogenetic analysis would be useful to illuminate the evolution of GPA-resistant determinants in GPA producers and from them to pathogens. A recent pioneering work on the reconstruction of GPA BGC phylogeny (Waglechner et al., 2019) reported on the possible origin and evolution of GPA cluster-situated van-genes. According to these authors, vanA had likely originated within Amycolatopsis genus, whereas vanH, vanX, and vanRS within Actinoplanes; and vanY probably originated within genus Nonomuraea and it was then distributed among GPA BGCs by multiple transfer events. Combination of these genes in pathogens is today determining the urgent clinical need for new drugs to combat multi-drug resistant Grampositive pathogens.

AUTHOR CONTRIBUTIONS
EB and OY collected data and papers and co-wrote the review. OY prepared the figures. FM and EB supervised the work.

FUNDING
This work was supported by grant "Fondo di Ateneo per la Ricerca" 2017, 2018 to FM, and Consorzio Italbiotec to EB.