New Results

A Borrelia burgdorferi LptD Homolog Facilitates Flipping of Surface Lipoproteins Through the Spirochetal Outer Membrane

Huan He, Ankita S. Pramanik, Selene K. Swanson, David K. Johnson, Laurence Florens, Wolfram R. Zückert
doi: https://doi.org/10.1101/2022.12.21.521298

SUMMARY

Borrelia spirochetes are unique among diderm bacteria in their lack of lipopolysaccharide (LPS) in the outer membrane (OM) and their abundance of surface-exposed lipoproteins with major roles in transmission, virulence, and pathogenesis. Despite their importance, little is known about how surface lipoproteins are translocated through the periplasm and the OM. In this study, we characterized Borrelia burgdorferi BB0838, a distant homolog of the OM LPS assembly protein LptD. Using a CRISPR interference approach, we showed that BB0838 is essential for cell growth. Upon BB0838 knockdown, sentinel surface lipoprotein OspA was retained in the inner leaflet of the OM, as determined by its inaccessibility to in situ proteolysis but its presence in OM vesicles. The secretion, insertion and topology of the B. burgdorferi OM porin P66 remained unaffected. MudPIT quantitative mass spectrometry analysis of the B. burgdorferi membrane-associated proteome further confirmed the selective periplasmic retention of surface lipoproteins under BB0838 knockdown conditions. Alphafold Multimer modeling predicted a B. burgdorferi LptB2FGCAD complex spanning the periplasm. Together, this indicates that BB0838 facilitates the essential terminal step in a distinctive spirochetal lipoprotein secretion pathway that evolved in parallel to the LPS secretion pathway in gram-negative bacteria. Hence, BB0838/LptDBb represents an attractive target for novel antimicrobials.

INTRODUCTION

Lipoproteins are ubiquitous membrane-associated proteins in monoderm (e.g., gram-positive) and diderm (e.g., gram-negative) bacteria. After Sec-mediated export through the cytoplasmic or inner membrane (IM), they are posttranslationally modified in a multistep process, yielding an acylated N-terminal cysteine at the start of a disordered tether peptide that anchors them peripherally in leaflets of the membrane lipid bilayer. In diderm bacteria, lipoproteins destined for the outer membrane (OM), such as the prototypical Brown’s lipoprotein Lpp, are known to be extracted by the Lol pathway and transported through the periplasm to be anchored in the periplasmic leaflet of the OM (Tokuda & Matsuyama, 2004, Okuda & Tokuda, 2011, Zückert, 2014). IM lipoproteins avoid recognition by the Lol machinery through interaction with IM phospholipids (Hara et al., 2003).

Long thought to be primarily confined to the diderm periplasm, lipoproteins have emerged as important virulence factors at the bacterial surface, i.e., the pathogen-host interface in various gram-negative genera, being involved in biological functions ranging from nutrient acquisition to immune evasion (reviewed in (Wilson & Bernstein, 2016). Prototypical surface display pathways range from the classical type 2 secretion system (T2SS) first described for Klebsiella pullulanase PulA (Pugsley et al., 1986, d’Enfert et al., 1987, D’Enfert & Pugsley, 1989, Pugsley et al., 1990, Sauvonnet & Pugsley, 1996, Francetic & Pugsley, 2005), the Neisseria T5SS or “autotransporter” pathway secreting NalP (van Ulsen et al., 2003, Roussel-Jazede et al., 2010, Roussel-Jazede et al., 2013), or the SLAM machinery mediating secretion of a family of Neisseria surface lipoproteins (Hooda et al., 2016). Another prominent example is Escherichia coli RcsF, an OM lipoprotein involved in envelope stress signaling, that has been shown to be partially surface exposed while it is sequestered within the cavity of an outer membrane protein (OMP) (Konovalova et al., 2014, Cho et al., 2014, Rodriguez-Alonso et al., 2020).

Borrelia spirochetes, the causative agents of vector-borne relapsing fever and Lyme borreliosis, are diderm bacteria with a rather distinct envelope structure. Unlike the gram-negatives – and even some other spirochetal genera such as Leptospira and Brachyspira – Borrelia are deficient in lipopolysaccharide (LPS) biosynthesis pathways and consequently lack LPS in the surface leaflet of the OM (Takayama et al., 1987, Fraser et al., 1997, Zückert, 2019). Instead, their bacterial surface is dominated by abundant, immunogenic, and serotype-defining surface lipoproteins that are differentially expressed and play major roles throughout subsequent phases of the vector-host transmission cycle (reviewed in (Coburn et al., 2021)). Our recent study, using the genetically tractable Lyme disease spirochete Borrelia burgdorferi as a model organism, demonstrated that two-thirds of the B. burgdorferi type strain B31 lipoproteome localizes to the surface (Dowdell et al., 2017). Tether peptide protease accessibility studies for selected model surface lipoproteins indicated that these surface lipoproteins are indeed anchored in the surface leaflet of the OM (Chen et al., 2011).

The molecular mechanisms guiding this diverse cohort of about 90 B. burgdorferi lipoproteins to the spirochetal surface have slowly come into focus. Using B. burgdorferi monomeric OspA and dimeric OspC and Borrelia turicatae Vsp1 as model surface lipoproteins in conjunction with a fluorescent localization reporter, mRFPΔ4 (Schulze et al., 2010), we showed earlier that their sorting determinants are encoded within the disordered N-terminal tether peptides; yet, sorting rules established for gram-negative periplasmic lipoproteins did not apply (Schulze & Zückert, 2006, Schulze et al., 2010, Kumru et al., 2010, Kumru et al., 2011). Mutations within tether peptides generally led to the mislocalization of surface lipoproteins to the periplasmic leaflet of the OM (Schulze & Zückert, 2006, Schulze et al., 2010, Kumru et al., 2010, Kumru et al., 2011). However, this mislocalization could be rescued by introducing mutations or conditions that destabilized the fold of OspA or OspA-calmodulin fusions, respectively (Schulze et al., 2010, Chen & Zückert, 2011). This indicated that surface lipoproteins crossed the OM in an at least partially unfolded conformation. Fittingly, dimeric lipoproteins were shown to assemble on the bacterial surface (Kumru et al., 2011). Together, this indicated the requirement for an OM “lipoprotein flippase” machinery that translocated both the polypeptide through the OM while also facilitating the flipping of the lipid anchor from the periplasmic to the surface leaflet (Zückert, 2019).

In their studies of the B. burgdorferi OM proteome, Akins and colleagues showed that depletion of the B. burgdorferi beta-barrel assembly machinery (BAM) protein BamA led to a significant growth defect and also reduced the abundance of surface lipoproteins in the OM (Lenhart & Akins, 2010). This indicated that proper localization of OM lipoproteins is dependent on BAM, most likely due to the BAM-mediated assembly of an essential integral OMP serving as the OM lipoprotein flippase. This led us to query the predicted essential B. burgdorferi OMPeome for any potential function in surface lipoprotein localization using a single plasmid-based non-toxic CRISPR interference (CRISPRi) system (Murphy et al., 2022). Here, we show that depletion of BB0838, a distant B. burgdorferi homolog of the gram-negative OM LPS translocase LptD, leads to selective retention of surface lipoproteins on the periplasmic side of the OM, while assembly of a sentinel beta-barrel OMP remains unaffected. This indicates that BB0838/LptDBb functions downstream of BAM in facilitating OM translocation of B. burgdorferi surface lipoproteins, thereby expanding the capacity of LptD homologs for lipidated substrates.

RESULTS

Identification of BB0838 as a putative B. burgdorferi OM lipoprotein LptD homolog

To overcome the biophysical obstacle of moving an amphipathic molecule across a lipid bilayer membrane, an OM lipoprotein flippase must have two separate domains: a hydrophilic transmembrane lumen that accommodates polypeptides of various dimensions, and a hydrophobic cavity that shelters the lipid moiety. To narrow down our list of candidate proteins, we used Phyre2 (Kelley et al., 2015) to generate structural models of 41 B. burgdorferi proteins that had been bioinformatically predicted to localize to the OM and/or assume a β -barrel structure (Kenedy et al., 2016). Among this set were BB0795/BamABb (Lenhart & Akins, 2010) and BB0838, at the time an OMP of unknown function (Kenedy et al., 2016).

We modeled the structure of BB0838 using i-Tasser and AlphaFold 2 (Yang & Zhang, 2015, Jumper & Hassabis, 2022) (Fig. 1). i-Tasser built the top model on the Shigella flexneri complex of LptD and LptE (C-score -2.22, T-score 0.45 ± 0.15, RMSD 14.9 ± 3.6 Å on PDB accession number 4Q35) despite less than 20% amino acid identity between them (Fig. 1A). In gram-negative bacteria, LptD together with lipoprotein LptE form the OM translocon for transporting LPS across the OM and displaying it on the cell surface (reviewed in (Konovalova et al., 2017)). The LptD N terminus folds into a β -jellyroll domain with 20 antiparallel β -strands forming a hydrophobic groove extending into the periplasm while the LptD C terminus forms a β -barrel transmembrane domain with 26 antiparallel β -strands (Dong et al., 2014). The LPS lipid A moiety is directly inserted into the OM via LptD’s N-terminal hydrophobic core while the hydrophilic oligosaccharide core and polysaccharide O antigen moiety of LPS enter the OM through the hydrophilic transmembrane lumen of the LptD β -barrel. LptE is anchored in the periplasmic leaflet of the OM and inserts into the LptD β -barrel lumen to facilitate the re-orientation of LPS to the bacterial surface (Chimalakonda et al., 2011, Chng et al., 2010, Freinkman et al., 2011, Ruiz et al., 2010, Wu et al., 2006).

FIG. 1.
FIG. 1. Structural Modeling of BB0838. (A) i-Tasser and AlphaFold models of BB0838.

Both algorithms predict a BB0838 fold similar to LptD. In both models, the N-terminal periplasmic β -jellyroll domain is colored in yellow, the central β -barrel domain in orange, and the C-terminal extensions forming either a coiled “plug” or β -barrel extension domains in purple. The i-Tasser model is shown in a side view emphazising the predicted plug domain, and the Alphafold model is shown in a front view emphasizing its separate N-terminal α-bundle and β -jellyroll domains. Both top views are rotated 90° along the vertical axis. The amino and carboxy termini of the protein are indicated by N and C, respectively. Stipled brackets indicate the approximate position of the OM lipid bilayer. (B) Comparison of predicted domain structures of spirochetal LptD homologs to gram-negative LptD. Color coding is as in panel A. Sf, S. flexneri; Bb, B. burgdorferi; Tp, T. pallidum; and Li, L. interrogans. ORF numbers for each homolog are indicated below each domain bar, and numbers indicate total preprotein length in amino acids. Note that only LPS-containing L. interrogans is predicted to have an LptE homolog.

The BB0838 model predicted a slightly shortened N-terminal β -jelly roll domain composed of 18 antiparallel β -strands preceded by a coiled coil. Compared to S. flexneri LptD, the β -barrel domain expanded to 28 antiparallel β -strands due to 94 additional amino acids, suggesting an expanded transmembrane lumen that may accommodate larger cargo. In addition to a predicted larger β -barrel, BB0838 has an additional 362 amino acid C-terminal extension that i-Tasser modeled as a coiled coil on LptE within the LptD lumen (Fig. 1). The AlphaFold model of BB0838 (Fig. 1A, Supplemental Table S3) similarly predicted a β -jelly roll domain of 18 antiparallel β -strands, but the extreme N terminus assumed an additional domain consisting of a cluster of four short α-helices connected to the β -jelly roll via a flexible linker. In another contrast to the i-Tasser model, BB0838’s C-terminal extension was incorporated into the β -barrel as additional β -strands, leading to a 32-stranded β -barrel that was partially restricted by a large periplasmic loop. In both models, the β -barrel had a lateral opening aligned with the periplasmic β -jelly roll domain. We therefore concluded that BB0838 indeed represents a structural LptD homolog that–in the absence of lipidated polysaccharides within the system–may have evolved to translocate lipidated proteins instead. Intriguingly, even distant LptD homologs of other spirochetal model organisms such as Treponema pallidum and Leptospira interrogans show a similar domain structure with a C-terminal extension, but only the LPS-containing L. interrogans also encodes for an LptE homolog (Hawley et al., 2021) (Fig. 1B, Supplemental Table S1).

Generation of a conditional BB0838 knockdown strain using CRISPR interference

A B. burgdorferi Tn mutagenesis screen indicated that BB0838 is essential for cell viability (Lin et al., 2012, Lin et al., 2014). BB0838 is the last gene in a 3-gene operon downstream of uvrB and uvrA (Fig. 2A), which appear non-essential for growth based on the analysis of a ΔuvrA strain (Sambir et al., 2011). We therefore generated a conditional bb0838 knockdown strain using a fully inducible, non-toxic CRISPR interference (CRISPRi) system encoded by a single recombinant E. coli/B. burgdorferi shuttle vector, pJJW101 (Murphy et al., 2022). pJJW101 carries a B. burgdorferi codon-optimized version of dCas9-myc under standard PQE30 (T5/lac hybrid promoter) control, which avoids toxicity of the original Streptococcus pyogenes dCas9 in B. burgdorferi under higher induction conditions (Takacs et al., 2020). Stringency is further maximized by placing the single guide RNA (sgRNA) module under trc (trp/lac hybrid) promoter control. Gene-specific target sgRNA sequences were designed with the web-based CRISPy-web interface on an artificial contig assembly of the B. burgdorferi B31 chromosome and plasmids to eliminate potential off-target effects as described (Murphy et al., 2022, Blin et al., 2016). Four target sgRNAs, complementary to either the preferred non-template (NT) strand or to the non-preferred template (T) strand (Fig. 2B and Table 1), were then ligated into the pJJW101 sgRNA module. NT1 and NT2 sgRNAs targeting an overlapping sequence close to BB0838’s 5’ end led to a marked in vitro growth defect upon CRISPRi induction, whereas T1 and T2 sgRNAs targeting separate sequences further downstream appeared to grow normally (data not shown).

FIG. 2.
FIG. 2. BB0838 operon structure and sgRNAs selected for CRISPRi.

(A) BB0838 is the last gene in a 3-gene operon with uvrB and uvrA. uvrA and BB0838 coding sequences overlap by 46 bps, i.e., solely within the signal I leader peptide of BB0838. Arrow heads show the position of all designed sgRNAs within the coding sequence for BB0838’s N-terminal β -jellyroll domain. (B) NT1, NT2, T1 and T2 sgRNA sequences are shown in bolded letters, with their start and end nucleotide positions within BB0838 indicated. Associated PAM sequences are underlined.

View this table:
Table 1. Oligonucleotide primers used in this study

To evaluate the efficiency of the NT1 and NT2 CRISPRi-mediated BB0838 knockdowns, we used quantitative reverse transcription polymerase chain reaction (qRT-PCR) assays, comparing transcription levels of BB0838 and a flaB normalization control in both the mock control strain (transformed with a sgRNA-deficient “empty” pJJW101 plasmid) and in the BB0838 knockdown strains (transformed with the respective sgRNA-containing pJJW101 vectors). IPTG-driven induction of dCas9 and these two sgRNAs led to an 84- and 12-fold knockdown of BB0838 mRNA transcript compared to the mock control, respectively, in total RNA samples collected at day 2 (Fig. 3). Due to its higher knockdown efficiency, we selected the NT1 sgRNA-expressing clone as the BB0838 knockdown (KD) strain for further study.

FIG. 3.
FIG. 3. Transcriptional BB0838 knockdown efficiency of NT1 and NT2 sgRNAs.

Total RNA samples obtained from B. burgdorferi cells harvested at 2 days post-inoculation were assayed by qRT-PCR using specific primers for BB0838 and flaB, which was used as a normalization control. Mock, cells harboring “empty” pJJW101 without sgRNA; NT1 and NT2, cells harboring pJJW101 with NT1 or NT2 sgRNAs, respectively. Results are from 3 biological replicates; error bars indicate mean ± SD.

BB0838 is essential for cell viability

To test for the expected role of BB0838 in cell viability, we monitored cell growth and phenotypic changes of B. burgdorferi BB0838 KD cells over time, using non-depleted and wild type cells as a control. 1 × 105 cells/ml were inoculated in complete BSK-II medium with or without 0.25 mM IPTG and followed for 3 days post-inoculation by phase contrast microscopy. As shown by the growth curves in Fig. 3A, a significant growth defect in the induced BB0838 KD cultures emerged after day 1, with a 10-fold reduction in cell numbers at day 2 compared to WT that increased to a 100-fold drop by day 3. Higher inducer concentrations did not lead to more severe growth defects (data not shown). Phase contrast micrographs (Fig. 3B) showed that depletion of BB0838 led to envelope disturbances and blebbing. The abnormal envelope structure is most likely caused by the envelope stress response resulting from the accumulation of mislocalized surface lipoprotein in the absence of BB0838. Together, these data suggested that dilution of BB0838 to about 10-20% of its WT level is sufficient to disrupt envelope biogenesis and leads to a cessation of bacterial growth.

BB0838 depletion affects the localization of a major outer surface lipoprotein

To test our hypothesis that BB0838 plays a crucial role in surface lipoprotein translocation, we used proteolytic shaving to assess any changes in surface exposure of lipoproteins in both mock control and BB0838 KD cells. Cultures inoculated with freshly cultured stationary phase cells at 1×106 cells/ml final concentration were grown in selective BSK-II medium with or without 0.25 mM IPTG for 2 days. Harvested and washed intact cells were then treated in situ with proteinase K to remove surface-exposed proteins or peptides as described (Bunikis & Barbour, 1999, Zückert et al., 2004, Schulze & Zückert, 2006, Chen et al., 2011). As shown by Western immunoblotting (Fig. 4A) surface lipoprotein OspA was accessible to proteinase K and degraded in both mock and uninduced BB0838 KD cells, while it remained largely protected in the induced BB0838 KD. At the same time, the topology of the OM porin P66, as assessed by protease accessibility of a surface-exposed loop (Bunikis et al., 1996, Skare et al., 1997, Bunikis & Barbour, 1999, Kenedy et al., 2014), remained unaffected. Periplasmic FlaB and cytosolic protein dCas9-Myc served as internal cell integrity controls.

Fig. 4.
Fig. 4. Assessment of BB0838 knockdown-dependent growth defects.

(A) Growth curve of wild-type (WT), mock control (with “empty” pJJW101) and BB0838 knockdown (KD) cells under inducing (0.25 mM IPTG) and non-inducing conditions. Cells were counted under phase contrast in a Petroff-Hausser counting chamber. Growth curves are from 3 parallel biological replicates; error bars indicate mean ± SD. (B) Representative phase micrographs of wild-type (WT) and BB0838 KD B. burgdorferi cells at day 2 post-inoculation under both non-inducing and inducing (+0.25 mM IPTG) conditions. White arrowheads indicate visible disturbances in the spirochetal envelope. Size bars are 10 nm.

To assess any defects in surface lipoprotein transport from the IM to the OM under BB0838-depleting conditions, we purified and analyzed the protein content of outer membrane vesicles (OMVs). B. burgdorferi cells cultured and harvested as described above at day 2 were osmotically shocked by resuspension in hypotonic citrate buffer, and the released OMVs were purified from remaining cell material (intact cells and protoplasmic cylinders) by ultracentrifugation on a discontinuous sucrose gradient as described (Skare et al., 1995, Radolf et al., 1995, Dowdell et al., 2017). Western immunoblots showed that OspA localized to OMVs under both mock control and BB0838-depleting conditions. OMV purity was assessed by the absence of IM lipoprotein OppAIV (Fig. 4B). Together, this set of experiments indicated that depletion of BB0838 primarily blocks the translocation of surface lipoproteins through the OM, but does neither significantly affect their transport through the periplasm, nor disturb the proper topology of integral OMPs such as P66.

BB0838 depletion leads to a specific localization defect of surface lipoproteins

To evaluate if the observed BB0838-dependent localization defect of OspA extends to the surface lipoproteome in general, we used quantitative multidimensional protein identification technology (MudPIT) mass spectrometry. B. burgdorferi mock control and BB0838 KD cells were cultured, harvested, and subjected to proteolytic shaving as described above. Samples were enriched for membrane-associated proteins by Triton X-114 detergent extraction followed by a 2-step precipitation with acetone and trichloroacetic acid as described (Dowdell et al., 2017). Two biological replicates were submitted for MudPIT analysis and processed in 5 or 6 technical replicates. We detected 44 of the previously localized 85 lipoproteins encoded by B. burgdorferi clone B31-e2 (Dowdell et al., 2017). Ten of the undetected lipoproteins (BB0456, BB0475, BB0735, BB0844, BBA33, BBA65, BBB08, BBJ01, BBP39, BBS41) were previously shown to be not transcribed in mid-exponential phase (Arnold et al., 2016), and the expression level of the remaining 31 lipoproteins was most likely below the MudPIT detection limit. 13 periplasmic IM, 5 periplasmic OM, and 10 surface lipoproteins were detected in at least two replicates of the uninduced control (838C) and analyzed further (Supplemental Table S2).

Based on the normalized spectral abundance factor (dNSAF) as a measure of protein abundance, we calculated the average lipoprotein surface exposure by dividing each protein’s dNSAF in the untreated (–pK) samples by the dNSAF in the corresponding protease-treated (+pK) samples, resulting in a-pK/+pK dNSAF ratio for each protein. Lipoproteins that localize to the periplasm, i.e., are protected from proteinase K, have close to identical dNSAF values in both the -pK and +pK samples and thus dNSAF ratios close to 1, whereas protease-accessible surface lipoproteins show a dNSAF ratio larger than 1 due to lower dNSAF values in the +pK sample (Dowdell et al., 2017). Conditional retention of a surface lipoprotein in the periplasm could therefore be followed by a drop of its -pK/+pK dNSAF ratio towards a value of 1.

As expected from our earlier studies, the dNSAF ratios for surface and periplasmic lipoproteins showed a clear separation (Fig. 6, Supplemental Table S2 and Fig. S1). dNSAF ratios for all 28 lipoproteins in the control uninduced (838C) samples generally tracked the values obtained with the proteomically more complex B. burgdorferi B31 clone B31-A3 (Dowdell et al., 2017). The 18 periplasmic IM and OM lipoproteins had dNSAF ratios around 1 (range 0.38 to 2.160. Calculated ratios for surface lipoprotein ranged between 7.12 and 97.63, with three proteins having infinite (∞) ratios due to undetectable peptides in the protease-treated samples, and one known low outlier (BBD10). In the induced BB838 knockdown (838KD) sample, the +pK/-pK dNSAF ratios for the periplasmic IM and OM lipoproteins remained stably around 1 (range 0.20 to 1.83), indicating their continued periplasmic localization. Surface lipoprotein ratios, however, collapsed from their high values in the 838C controls to a range of 0.00 to 3.98, indicating that they were now substantially retained within the periplasm. For comparison to the immunoblot data shown in Fig. 5, OspA showed a drop in +pK/-pK ratio from 45.83 to 1.81. Together, these data indicate that depletion of BB0838 leads to a protein localization defect in B. burgdorferi cells that is specific to surface lipoproteins. This implicates BB0838 as the terminal component of a surface lipoprotein secretion pathway.

Fig. 5.
Fig. 5. Assessment of BB0838-depended protein localization defects.

(A) Western immunoblots of total cellular proteins of B. burgdorferi cells harboring “empty” pJJW101 (mock) or pJJW101 with the two BB0838-specific sgRNAs (NT1 and NT2), under non-inducing (– IPTG) or inducing (+ IPTG; 0.25mM) conditions, before (– pK) or after (+ pK) in situ proteolysis with proteinase K. CRISPRi induction and expression of the C-terminally c-Myc-tagged BbdCas9 was verified using an anti-c-Myc antibody. Surface lipoprotein OspA served as the model surface lipoprotein. OMP P66 served as a OMP topology control, and periplasmic FlaB was used as a cell integrity and constitutively expressed loading control. (B) Western immunoblots of B. burgdorferi outer membrane vesicle (OM) and protoplasmic cylinder (PC) fractions from B. burgdorferi cells harboring “empty” pJJW101 (mock) or pJJW101 with the BB0838-specific NT1 sgRNA. IM lipoprotein OppAIV and conditionally expressed dCas9-c-Myc served as OMV purity controls. Note that the PC fraction also contains OM proteins such as OspA due to the partial separation of OMVs from protoplasmic cylinders by treatment of Borrelia cells with hypotonic citrate buffer (Skare et al., 1995).

Fig. 6.
Fig. 6. Quantitative proteomic analysis of BB0838-dependent lipoprotein localization defects.

Multidimensional protein identification technology (MudPIT) quantitative proteomics analysis of membrane-associated proteins in the recombinant B. burgdorferi BB0838 KD strain. The plot shows the mean ratios of normalized peptide count ratios after vs. before proteinase K treatment (+pK/-pK dNSAF ratio) for subset of previously localized IM, periplasmic OM and surface lipoproteins in B. burgdorferi B31 clone B31-A3 (Dowdell et al. 2017). Data are shown for the inducible NT1-mediated BB0838 CRISPRi knockdown under uninduced (838C, clear squares) and induced (838KD, clear inverted triangles) conditions. Data are from 5-6 technical replicates from duplicate biological replicates. Data for B31-A3 (grey filled circles) (Dowdell et al. 2017) are shown for validation. Note that the y axis is split to help visualize the lower dNSAF ratios from low-expression proteins. See also supplementary data Table S1 and Fig. S2).

Modeling of the B. burgdorferi Lpt pathway

In gram-negative bacteria, the LPS-transporting Lpt pathway forms a continuous periplasmic bridge between the IM and OM. A dimer of the cytoplasmic ATPase LptB forms an ABC transporter-like IM complex with integral membrane proteins LptF and LptG. This IM complex interacts via LptF with LptC, whose periplasmic domain is anchored in the IM via a single N-terminal transmembrane domain. LptC’s C-terminus itself interacts with the N-terminus of periplasmic LptA, which connects to the N-terminal periplasmic domain of LptD at the OM. LptC, LptA, and the N-terminus of LptD create a hydrophobic “greasy slide” within a continuous β -jelly-roll structure that allows for the periplasmic transport of LPS’s fatty acid membrane anchors. Having shown that BB0838/LptDBb plays a role in surface lipoprotein translocation without an apparent LptE homolog, we wondered whether B. burgdorferi harbors Lpt homologs upstream of LptD that would complete the pathway.

A prior search of the NCBI Protein and CDD databases by Putker et al. (Putker et al., 2015) using COG database categories (Tatusov et al., 2000) identified B. burgdorferi homologs for LptA (BB0465), LptB (BB0466), LptF (BB0807) and LptG (BB0808), but no clear LptC homolog. In other bacterial systems, LptC is often encoded within the same locus as LptA and LptB in an lptCAB operon (Putker et al., 2015). Based on a likely synteny in B. burgdorferi, we speculated that ORF BB0464 upstream of lptA (BB0465) could encode for the B. burgdorferi LptC homolog. Like LptDBb, all five predicted Lpt homologs appeared to be essential based on a lack of Tn insertions (Lin et al., 2012, Lin et al., 2014).

To evaluate the five additional Lpt pathway candidates, we generated structural models using AlphaFold (Jumper & Hassabis, 2022) (Fig. 7A). Signal peptides predicted by SignalP 6.0 (Teufel et al., 2022) were excluded. pLDDT (predicted Local Difference Distance Test) values were used as a model confidence metrics. The top model of BB0465 (LptABb) without its signal I peptide showed the expected β -jellyroll fold (pLDDT 86.4). Compared to E. coli LptA, BB0465 is 45 amino acids larger, which in the model extended the fold by four β -strands to 12 total. Both N and C termini appeared disordered, with potential for some short N-terminal α-helical structure. BB0464 (LptC) intriguingly was predicted to be a lipoprotein with a relatively short 14-amino acid N-terminal signal II peptide (SignalP6.0 probability = 0.99). Thus, LptC would be anchored in the IM via a different mechanism than gram-negative LptC, which uses an N-terminal α-helix membrane anchor (Wilson & Ruiz, 2022). The top model showed the expected overall β -jellyroll fold (pLDDT 90.3), with a short N-terminal α-helix following four disordered tether residues after the predicted N-terminal cysteine. The LptB (BB0466) model predictably assumed the structural fold of an ABC transporter ATPase (pLDDT 92.8). The LptF (BB0807) model showed six transmembrane helixes, with the third helix uniquely extending into the periplasm as a two-α-helix stalk before transitioning into a typical small periplasmic β -jelly roll domain (pLDDT 85.7). LptG (BB0808) modeled similarly to LptF, but lacked the periplasmic α-helical stalk (pLDDT 78.1). This analysis suggested that the B. burgdorferi Lpt pathway consisted of a multimeric complex of LptB2FGCAD, i.e., was missing only a recognizable LptE homolog found in other diderm bacterial systems.

Fig. 7.
Fig. 7. Model of B. burgdorferi lipoprotein transport. (A) Composite AlphaFold Multimer model of the B. burgdorferi Lpt pathway.

Overlapping Alphafold and Alphafold Multimer models of LptD, LptDNA, LptDNAC, LptCFG and LptFGB2 were aligned in PyMol (see also text). Location of the subunit N and C termini are indicated in the same color as the subunits. The periplasmic span of the modeled complex was measured in PyMOL as the distance between the top of the membrane-spanning LptGF helices and the end of the periplasmic β -jelly roll domain of LptD. Note that the kink of the complex at the IM is likely an artefact of the model. (B) Proposed role of the B. burgdorferi Lpt and Lol pathways in lipoprotein sorting and secretion. B. burgdorferi homologs of known components are indicated by their TIGR ORF number. Canonical pathway components that are apparently missing in Borrelia are indicated by red strikethroughs. Model OMPs and surface, periplasmic OM and IM lipoproteins are indicated. Our model proposes that after complete posttranslational modification by the lipoprotein processing machinery in the IM, (i) Borrelia surface lipoproteins such as OspA are recognized and extracted by the Lpt pathway and fast-tracked along a periplasmic bridge and through the OM to the bacterial surface; (ii) the Lol pathway is solely responsible for ensuring proper localization of periplasmic OM lipoproteins such as the abundant Lp6.6 or the BAM complex associated lipoproteins BamB and BamD, and (iii) IM lipoproteins such as OppAIV avoid interaction with both the Lpt or Lol pathway to remain in the IM.

To predict protein-protein interactions and gain some initial dimensional insights, we used AlphaFold-Multimer (Evans et al., 2022) to generate a first model of the B. burgdorferi Lpt pathway. In addition to overall pLDDT, the predicted TM (pTM) and interface predicted TM (ipTM) scores were used as model confidence metrics. For multimer modeling purposes, we used only the N-terminal periplasmic domain of LptD (LptDN). The overlapping multimer models of LptDN A (pLDDT 85.6, pTM 0.79, piTM 0.79), LptDN AC (76.9, 0.70, 0.62), LptAC (78.4, 0.76, 0.71), LptCFG (76.1, 0.69, 0.64) and LptFGB2 (82.4, 0.78, 0.77) were assembled with the original LptD model by alignment in PyMol, resulting in the overall structural model of the B. burgdorferi Lpt pathway shown in Fig. 7A. Modeling of the entire LptB2 FGCADN complex resulted in the same overall structure (pLDDT 75.5, 0.8×piTM+0.2×pTM=0.545; not shown). Three key features of the predicted complex are notable: First, LptC, LptA, and the N terminus of LptD form an approximately 185 Å-long periplasmic bridge in a continual head-to-tail (N-to C-terminal) orientation. AlphaFold rejected the introduction of any additional LptA subunits into this bridge, although we were able to model LptA as a head-to-head (N-to-N terminus) homodimer and tail-to-head (C-to-N terminus) homotrimers (Supplemental Table S3). Second, the small α-helical domain predicted by AlphaFold at the N terminus of LptDBb moved from the monomer model orientation to now laterally interact with LptA. This suggests that this domain might function as a “clasp” to stabilize the LptA-LptD protein junction. Third, compared to LptG and other homologs, LptF contains an insertion that is predicted to protrude as an α-helical “stalk” domain into the periplasm. While this could be an artefact of the model, the DeepTMHMM transmembrane protein topology algorithm (Hallgren et al., 2022) equally predicted this domain to be “outside”, i.e., not part of an extended transmembrane domain.

DISCUSSION

Envelope homeostasis is a fundamental process that ensures the continued growth and persistence of bacterial cells in a variety of sometimes hostile environments. In the arthropod-borne Lyme disease spirochete B. burgdorferi, this process includes the mechanisms that properly segregate over 130 lipoproteins to be either retained in the IM, transported to the periplasmic side of the OM, or secreted to the bacterial surface. Our earlier lipoproteome compartmentalization analysis showed that two-thirds, i.e. more than 87 of these lipoproteins reach the borrelial surface (Dowdell et al., 2017). This confirmed that secretion to the bacterial surface indeed can be considered the “default” for B. burgdorferi lipoproteins and suggested an efficient pathway with broad cargo specificity that connects to the posttranslational lipoprotein modification machinery in the IM and guides lipoproteins through the periplasm and the OM.

This study provided additional insights into this pathway by characterizing the biological role of B. burgdorferi BB0838. Both experimental evidence and structural modeling indicate that BB0838, already shown to be an integral OM protein (Kenedy et al., 2016), is responsible for the terminal step of surface lipoprotein secretion at the B. burgdorferi OM. First, the depletion of BB0838 specifically prevented the translocation of surface lipoproteins through the OM. This specific surface lipoprotein mislocalization phenotype went hand-in-hand with a significant growth defect, confirming that BB0838 is essential. Second, high-confidence in silico protein modeling indicated that BB0838 is a structural homolog of the gram-negative OM protein LptD, which, analogous to the secretion of amphipathic lipopolysaccharide molecules by gram-negative bacteria, would allow for the secretion of amphipathic lipoproteins in B. burgdorferi. The two top models consistently showed a periplasmic β -jelly roll linked to a relatively large, laterally open β -barrel pore domain that would allow for the passage of both hydrophobic and hydrophilic lipoprotein moieties and insertion into the surface leaflet of the OM lipid bilayer. However, the models differed in their integration of BB0838’s C-terminal extension that contributes to its significantly larger size (1146 amino acids, 120 kDa) compared to gram-negative LptD (784 amino acids, 87 kDa). The first model, built on a gram-negative LptDE complex by i-Tasser, inserted an LptE-like C-terminal plug domain from the outside into the β -barrel pore. The second model built by AlphaFold appeared less constrained and integrated the extension into a larger β -barrel with an estimated pore diameter of 3.7 nm. For comparison, B. burgdorferi outer membrane porin/adhesin P66 forms a pore with a deduced entry diameter of 1.9 nm and an internal constriction of 0.8 nm (Barcena-Uribarri et al., 2013). Thus, it is likely that BB0838 either assumes a structure that is a hybrid of the two models, or that the larger pore is obstructed by other means, e.g., by a yet-to-be-identified interacting protein or by transiting surface lipoproteins. Notably, both models are compatible with current topology data showing at least one protease-accessible surface loop as well as a protease-protected C terminus (Kenedy et al., 2016). Interestingly, LptDBb does not contain any Cys residues, indicating that unlike gram-negative LptD, it does not use disulfide bridges to coordinate conformational changes that lead to interaction with periplasmic binding partners (Ruiz et al., 2010, Chng et al., 2012).

BB0838 homologs appear conserved among pathogenic spirochetes. Based on BlastP searches, Lyme disease Borrelia homologs share over 90% amino acid identity, whereas relapsing fever Borrelia homologs (over 60% amino acid identity) and homologs found in other spirochetes such as treponemes and leptospires (around 20% amino acid identity) are more distantly related (Table S1). Structure homology searches revealed LptD homologs in Treponema pallidum (TP0515) and Leptospira interrogans (LIC1458). While of intermediate size, they are also predicted to fold into a similar three-domain structure (Hawley et al., 2021), whereas LptE homologs were only found in the LPS-displaying leptospires. It is therefore tempting to speculate that BB0838’s C-terminal domain is supporting the translocation of lipoproteins through the OM in all three spirochetal systems. At this point, we cannot exclude that spirochetal Lpt pathways transport surface glycolipids, as suggested by Hawley et al. (Hawley et al., 2021). However, this would clash with the rather specific phenotype of the BB0838 KD described here and the presence of glycolipids in lipid rafts of both the IM and OM of B. burgdorferi (Toledo et al., 2014, Toledo et al., 2015, Toledo et al., 2018a, Toledo et al., 2018b).

Multimer modeling of the predicted B. burgdorferi Lpt pathway complex suggests a continuous periplasmic bridge that is formed by head-to-tail monomers of LptC, LptA, and the N-terminus of LptD that is estimated to be about 18 nm long (Fig. 7A). This distance corresponds to the average distance between the B. burgdorferi IM and OM as determined by cryo-ET (Charon et al., 2009). Yet, periplasmic widths vary depending on the presence of flagella (Kudryashev et al., 2009), and the longitudinal dimensions of the periplasm-spanning B. burgdorferi Tol-like BesABC complex (Bunikis et al., 2008, Greene et al., 2013) and the BAM complex-associated, IM-anchored TamB (Iqbal et al., 2016) are modeled to be slightly larger. Thus, stoichiometric studies will have to determine whether additional LptABb subunits, each extending the periplasmic bridge by about 6 nm, are needed to form functional multiprotein Lpt complexes.

We propose that BB0838/LptDBb acts as the previously proposed lipoprotein flippase at the end of a separate surface lipoprotein secretion system. Our data indicate that depletion of BB0838/LptDBb leads to a specific OM translocation defect of surface lipoproteins. This defect is reminiscent of the phenotypes we obtained in our previous lipoprotein localization studies, where mutations in the N-terminal disordered tether peptides of various Borrelia surface lipoproteins led to their mislocalization to the periplasmic leaflet of the OM (Schulze & Zückert, 2006, Schulze et al., 2010, Kumru et al., 2010, Kumru et al., 2011). The ability to redirect these tether mutants to the surface by destabilizing their tertiary structures (Schulze et al., 2010, Chen et al., 2011) suggested that these mutants were equivalent to secretion intermediates in the periplasmic leaflet of the OM rather than proteins that were ejected from the Borrelia surface lipoprotein secretion pathway at the OM. In light of the involvement of BB0838/LptDBb in surface lipoprotein secretion, we need to re-evaluate these conclusions. First, the Borrelia Lpt machinery with LptDBb at its terminus would provide a continuous conduit for lipoproteins to reach the surface, with limited opportunities for the egress of stalled proteins after entry at the IM. Second, surface lipoproteins interacting with the Lpt machinery would bypass the canonical OM lipoprotein sorting Lol pathway in its entirety, which would result in distinct cargo profiles for the Lpt and Lol pathways. Recent studies in E. coli have posited that one of the Lol pathway’s roles is to alleviate envelope stress by removing mislocalized periplasmic lipoproteins from the IM to the OM (Grabowicz & Silhavy, 2017). It is therefore possible that at least some of the lipoprotein tether mutants are rejected by the Lpt pathway at the IM and rerouted through Lol to the inner leaflet of the OM.

An earlier finding that depletion of the B. burgdorferi BamA ortholog BB0795 led to a decrease of both OMPs and lipoproteins in the OM (Kenedy et al., 2016) is compatible with the expected BAM complex dependence of BB0838/LptDBb insertion and folding. Our finding that P66, a major B. burgdorferi OM porin and adhesin (Bunikis & Barbour, 1999, Bunikis et al., 1998, Bunikis et al., 1995, Bunikis et al., 1996, Coburn & Cugini, 2003, Kenedy et al., 2014, Skare et al., 1997) maintains its w.t. topology in the absence of BB0838/LptDBb further corroborates that BB0838 acts downstream of BAM and indicates that BB0838 depletion does not lead to a more generalized OM disturbance.

The fact that BB0838/LptDBb is an essential, partially surface-exposed OMP makes it an attractive therapeutic target. LptD proteins have been successfully evaluated as potential vaccinogens for Neisseria gonorrhoeae and Vibrio parahaemolyticus (Zielke et al., 2016, Zha et al., 2016). Moreover, two novel antimicrobials targeting Pseudomonas aeruginosa and E. coli LptD have been identified. The β -hairpin-like peptidomimetic L27-11 targeting P. aeruginosa LptD showed potent antimicrobial activity in a mouse septicemia model (Srinivas et al., 2010) while the JB-95 β -hairpin macrocyclic peptide bound to E. coli LptD and BamA and showed potent antimicrobial activity against a panel of clinical strains (Urfer et al., 2016). The β-hairpin structure of these peptidomimetics might interact with the N terminus β-jellyroll of LptD (Andolina et al., 2018).

In summary, we used CRISPRi-mediated protein depletion to establish BB0838 as an LptD homolog that is involved in the terminal step of lipoprotein secretion to the B. burgdorferi surface. This suggests that the substrate specificity of Lpt pathways in diderm bacteria extends beyond LPS. Supporting this is the notion that LptD orthologs were also found in T. pallidum and other diderm LPS-deficient bacteria such as Novosphingobium aromaticivorans, Deinococcus radiodurans, Thermus thermophilus, and Thermotoga maritima (Putker et al., 2015). Yet, as in other diderm bacterial systems, B. burgdorferi LptD is involved in the display of lipidated immunodominant virulence factors, and thus is part of an elemental envelope homeostasis pathway that is indispensable for microbial pathogenesis. We propose that the presence of two distinct lipoprotein transport pathways in B. burgdorferi hints at an emerging dichotomy between Lol-mediated periplasmic lipoprotein sorting and Lpt-mediated surface lipoprotein secretion (Fig. 7B). This would indicate that sorting of the diverse B. burgdorferi lipoproteome to its three possible destinations occurs at the IM: (i) surface lipoproteins such as OspA would be fast-tracked after being recognized by LptBFG and pushed along the LptCAD periplasmic bridge and through the LptD lumen and lateral opening to the bacterial surface; (ii) periplasmic OM lipoproteins such as Lp6.6 or the essential BAM associated lipoproteins BamB and BamD would avoid interaction with the LptBFG complex to be recognized by LolCDE and transported and inserted into the periplasmic leaflet of the OM by LolA; and (iii) IM lipoproteins such as the oligopeptide-binding OppAIV would avoid either pathway to remain in the IM (Fig. 7B). Proper insertion and topology of LptD as the final component in the Lpt complex would be extrinsically linked to the Sec- and Lol-dependent assembly of the BAM complex in the OM. Yet, a requirement for LptC lipidation may represent an additional early and intrinsic quality control checkpoint at the IM that prevents assembly of a functional Lpt complex if its own maturation – and that of its lipoprotein cargo – is disturbed. At the same time, a lipidated LptC would lack the recently discovered rate-modulating function of the anchoring transmembrane α-helix of gram-negative LptC (Wilson & Ruiz, 2022). Our future studies will test these hypotheses and determine structure-function relationships of the B. burgdorferi Lpt and Lol pathway components, define their individual cargo specificities and potential interactions, and reassess previously identified Borrelia lipoprotein sorting determinants.

EXPERIMENTAL PROCEDURES

Bacterial strains and culture conditions

E. coli strain NEB5 (New England Biolabs) was used for plasmid construction and propagation, and grown in LB broth or on LB agar supplemented with selective antibiotic (40 µg/ml kanamycin; Sigma) as indicated. Borrelia burgdorferi B31-e2, a non-infectious clonal derivative of type strain B31 (Babb et al., 2004) encoding for 85 lipoproteins due to its reduced plasmid content (cp26, cp32-1, cp32-3, cp32-4, lp17, lp38, and lp54), was cultured in liquid or solid BSK-II medium at 34°C under a 5% CO2 atmosphere (Barbour, 1984, Zückert, 2007). Selective BSK-II medium contained 200 μg/ml kanamycin. Protein expression from hybrid lac promoters was induced by addition of isopropyl-b-D-thiogalactopyranoside (IPTG 0.25 mM; Sigma) where indicated.

Recombinant plasmid construction and sgRNA design for CRISPRi

pJJW101 (Murphy et al., 2022) is a derivative of the E. coli-B. burgdorferi shuttle vector pJSB142 and the Mobile-CRISPRi plasmid pTn7C107 (Blevins et al., 2007, Peters et al., 2019). The plasmid encodes for an IPTG-inducible, codon-optimized non-toxic variant of SpydCas9 (BbdCas9) as well as an IPTG-inducible single guide RNA (sgRNA) cassette that allows for easy insertion of sgRNA spacer sequences. BB0838-specific sgRNA spacer sequences were designed using the web-based CRISPy-web interface (Blin et al., 2016) on an artificial genomic contig of B. burgdorferi B31 as described (Murphy et al., 2022). Four target sgRNAs, two complementary to the non-template (NT) strand, and two complementary to the template (T) strand were selected (Table 2), synthesized as complementary pairs of single-stranded oligonucleotides with BsaI site overhangs (Table 1), annealed, and directionally ligated into the cut BsaI site array within the sgRNA cassette of pJJW101 as described (Murphy et al., 2022).

View this table:
Table 2. sgRNA spacer sequences used in this study.

Growth curve assays and phase contrast microscopy

B. burgdorferi cells were grown from frozen stocks in selective liquid BSK-II liquid medium containing 200 µg/ml kanamycin to mid-exponential phase (about 2 days). Bacteria were then seeded at final concentrations of 1×105 organisms/ml into liquid medium containing kanamycin (200 µg/ml) without or with IPTG (final concentration of 0.25 mM) and incubated at 34°C for 3 days. Spirochete numbers were determined by counting in a Petroff-Hausser counting chamber under phase contrast microscopy (Nikon Eclipse E400). For phenotypic analysis, cells were collected at day 2 post-inoculation, washed with Dulbecco’s phosphate-buffered saline (PBS, pH 7.4), fixed with 4% formaldehyde for 15 min at room temperature, and washed again with PBS to remove the formaldehyde. Micrographs were taken under phase contrast using a Nikon Eclipse E600 microscope (40× 0.55 numerical aperture Ph2 phase-contrast objective) connected to an INFINITY 3 digital camera (Teledyne Lumenera).

Total RNA extraction and qRT-PCR analysis

B. burgdorferi cells grown as described above were harvested by centrifugation at 8,000 × g for 20 min at room temperature, washed once by resuspension in sterile room-temperature PBS containing 5 mM MgCl2 (PBS+Mg) and repelleted. Total RNA was extracted with TRIzol (Invitrogen) according to the manufacturer’s instructions. To remove residual DNA, the RNA samples were treated with 0.1 U DNase I (Invitrogen) for 1h at 37ºC, followed by phenol-chloroform extraction (Ambion) and ethanol precipitation overnight at -20ºC (Sambrook & Russell, 2001). The purified RNA was quantified using a Nanodrop ND1000 spectrophotometer. RNA samples were then subjected to qRT-PCR of BB0838 and flaB transcripts using the Luna Universal One-Step RT-qPCR kit (NEB) and an ABI Prism 7500 system (Applied Biosystems) according to the manufacturer’s instructions. Oligonucleotide primers are listed in Table 1. BB0838 transcript levels were validated and normalized against flaB mRNA, and fold changes were calculated using the comparative CT (2-ΔΔCT) method for quantification.

SDS-PAGE and immunoblotting

B. burgdorferi cells were harvested by centrifugation and washed with PBS+Mg as described above. Bacterial pellets were solubilized in 1× sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) sample buffer containing 50 mM dithiothreitol (DTT), boiled for 5 min, and stored at -20 °C. Whole cell proteins were separated on a 12% polyacrylamide SDS-PAGE gel (Sambrook & Russell, 2001) and visualized by Coomassie blue staining (Fisher BioReagents™ EZ-Run™ Protein Gel Staining Solution, #BP3620-1). For immunoblots, proteins proteins were electrophoretically transferred to nitrocellulose membranes (Millipore) using a Transblot semi-dry transfer cell (Bio-Rad). The membranes were rinsed in Tris-buffered saline (TBS) (20 mM Tris, 500 mM NaCl, pH 8.0). TBS with 0.05% Tween 20 (TBST) containing 5% dry milk was used for membrane blocking and subsequent incubation with primary and secondary antibodies; TBST alone was used for the intervening washes (Sambrook & Russell, 2001). Antibodies used were anti-OspA mouse monoclonal antibody H5332 (1:100 dilution) (Barbour et al., 1983), anti-c-Myc mouse monoclonal antibody (1:1000, Thermo Fisher, 9E10), anti-P66 rabbit polyclonal antibody (1:100) (Bunikis et al., 1995), and anti-OppAIV rabbit polyclonal antibody (1:100) (Bono et al., 1998). Secondary antibodies were Alkaline Phosphatase (AP)-conjugated anti-mouse (1:30,000, A3562, Sigma), and anti-rabbit antibodies (1:30,000, A3687, Sigma). Blots were developed using AP substrate CDP-Star (BioRad) for chemiluminescent detection, and signals were detected and captured using a Fujifilm LAS-4000 CCD imaging system.

Surface proteolysis of intact B. burgdorferi spirochetes

Proteolytic shaving of intact spirochetes with proteinase K was performed as described (Bunikis & Barbour, 1999, Zückert et al., 2004, Schulze & Zückert, 2006). Briefly, B. burgdorferi cells harvested and washed as described above were resuspended in PBS+Mg without or with proteinase K (Invitrogen, 200 µg/ml final concentration). Proteinase K-containing and control samples were incubated for 1 h at room temperature, and reactions were stopped after 1 h by adding phenylmethylsulfonyl fluoride (PMSF) to a final concentration of 5 mM. Subsequently, cells were pelleted by centrifugation, resuspended in 1× SDS-PAGE sample buffer, boiled for 5 min, and stored at -20°C for further analysis.

Membrane fractionations

B. burgdorferi outer membrane vesicles (OMVs) and protoplasmic cylinders (PCs) were isolated as described (Skare et al., 1995, Lenhart & Akins, 2010). Briefly, cells were harvested at room temperature by centrifugation for 20 min at 8,000 × g, washed in PBS with 0.1 % bovine serum albumin (BSA), and repelleted. The pellet was then resuspended in 38 ml ice-cold 25 mM citrate buffer (pH 3.2) with 0.1% BSA and incubated at room temperature for 2 hours with agitation and a 1-min vortexing step every 30 min. Next, cells were pelleted by centrifugation at 20,000 × g for 30 min and resuspended in 6 ml ice-cold 25 mM citrate buffer (pH 3.2) with 0.1% BSA. This sample was then layered on a discontinuous 56% (wt/wt in 4 ml), 42% (wt/wt in 15.5 ml) and 25% (wt/wt in 12.5ml) sucrose gradient in 25 mM citrate buffer in 38 ml Ultraclear tubes (Beckman-Coulter, 344058). The gradient was centrifuged at 100,000 × g for 18 h at 4°C (Beckman-Coulter XPN-80 Ultracentrifuge, SW32 Ti swinging-bucket rotor) to separate the OMV (upper band) and PC (lower band) fractions. The PC fraction was collected, diluted with PBS, repelleted at 20,000 × g for 20 min, and resuspended in 1 ml PBS with 1mM PMSF for storage at -20°C. The OMV fraction was collected, diluted with PBS, repelleted at 100,000 × g for 4 h at 4°C, and resuspended in 100 µl PBS for storage at -20°C.

Quantitative mass spectrometry (MudPIT)

B. burgdorferi cells were subjected to surface proteolysis with proteinase K as described above and harvested. Membrane-associated proteins were then enriched by overnight extraction with Triton X-114, as described (Carroll, 2010, Dowdell et al., 2017). The washed detergent extracts were then precipitated at -20 ºC overnight in final 80 % (vol/vol) acetone, resuspended in 0.1 M Tris-HCl (pH 8.5), and precipitated again overnight using 20% trichloroacetic acid (TCA). The addition of acetone precipitation in the protocol was necessary to effectively remove detergent prior to analysis by MudPIT (Dowdell et al., 2017). Desiccated frozen protein samples from three biological replicates were then submitted for MudPIT analysis (Proteomics Center, Stowers Institute for Medical Research, Kansas City, MO). Resuspended protein samples were digested with endoproteinases Lys-C (Roche) and trypsin (Promega) at 0.1 µg/µl final concentration each. The proteinase-digested samples were then analyzed by MudPIT on an LTQ linear ion trap (Thermo Scientific) coupled to a Quaternary Agilent 1100 series high-performance liquid chromatograph (HPLC) (Florens & Washburn, 2006). Protein content in mock control versus proteinase K-treated whole-cell protein preparations was analyzed by comparison of the average distributed normalized spectral abundance factor (dNSAF) for each unique protein, which correlates directly with the relative abundance of a particular protein in the sample (Zhang et al., 2010). A mean dNSAF ratio of untreated control to protease-treated sample (dNSAF -pK/+pK ratio) was calculated for each protein. All MudPIT raw datasets have been deposited in the MassIVE Repository (ftp://MSV000089990@massive.ucsd.edu) and will also be available after publication from the Stowers Original Data Repository at https://www.stowers.org/research/publications/libpb-1732.

Bioinformatics and molecular modeling

NCBI BLASTP (Johnson et al., 2008) was used to identify protein homologs among pathogenic spirochetes. The original structural model for BB0838 was generated by homology modeling using i-Tasser and the Shigella flexneri LPS-assembly protein LptD (PDB accession number 4Q35) (Qiao et al., 2014) as a template. Monomer models were generated using AlphaFold-monomer (Jumper & Hassabis, 2022) and assessed based on pLDDT scores and visual inspection. Protein complex models were generated using AlphaFold-Multimer (Evans et al., 2022) and assessed based on the weighted sum of pTM and piTM values (the AlphaFold-Multimer ranking metric), pLDDT scores, and visual inspection. Model PDB files were visualized using PyMOL (The PyMOL Molecular Graphics System, Version 2.0 Schrödinger, LLC).

AUTHOR CONTRIBUTIONS

WRZ and HH conceived and designed the study; HH, AP, SKS and DKJ acquired the data; HH, SKS, DKJ, LF and WRZ analyzed and interpreted the data; WRZ and HH wrote the manuscript.

GRAPHICAL ABSTRACT

Model figure based on Fig. 7

GRAPHICAL ABSTRACT TEXT

CRISPRi was used to efficiently knock down expression of Borrelia burgdorferi BB0838, an essential spirochetal outer membrane protein related to LptD lipopolysaccharide (LPS) transporters in gram-negative bacteria. The BB0838 knockdown showed a specific defect in translocation of surface lipoproteins through the spirochetal outer membrane. This suggests that BB0838 functions as an outer membrane lipoprotein flippase, expanding the substrate specificity of LptD family proteins in LPS-deficient systems as part of a dichotomous Lpt/Lol lipoprotein transport system.

PLAIN LANGUAGE SUMMARY

The surface of the Lyme disease spirochete Borrelia burgdorferi is covered with a variety of abundant lipoproteins, which play important roles during natural transmission and human infection following a tick bite. Here, we show that the bacterium has seemingly adopted an orphan LPS endotoxin secretion machinery found in gram-negative bacteria to secrete its complex surface lipoproteome.

SUPPLEMENTAL DATA

(see separate file)

Table S1. BB0838 conservation among pathogenic spirochetes

Table S2. MudPIT data for 28 selected B. burgdorferi lipoproteins

Table S3. AlphaFold and AlphaFold Multimer models of full-length B. burgdorferi B31 LptD and LptA homologs

Fig. S1. MudPIT data for 28 selected B. burgdorferi lipoproteins

ACKNOWLEDGMENTS

We thank Sven Bergström (Umeå University, Sweden) and Patricia Rosa (NIH/NIAID Rocky Mountain Laboratories, Hamilton, Montana, USA) for antibodies and Jacob J. Wiepen for technical assistance. This work was supported in part by a KUMC Biomedical Research Training Program fellowship to HH, as well as National Institutes of Health grants P20 GM113117 (Pilot grant) and R21AI144624 to WRZ. SKS and LF were supported by the Stowers Institute for Medical Research.

REFERENCES

  1. ↵
    Andolina, G., Bencze, L.C., Zerbe, K., Muller, M., Steinmann, J., Kocherla, H., Mondal, M., Sobek, J., Moehle, K., Malojcic, G., Wollscheid, B., and Robinson, J.A. (2018) A Peptidomimetic Antibiotic Interacts with the Periplasmic Domain of LptD from Pseudomonas aeruginosa. ACS Chem Biol 13: 666–675.
    OpenUrlCrossRef
  2. ↵
    Arnold, W.K., Savage, C.R., Brissette, C.A., Seshu, J., Livny, J., and Stevenson, B. (2016) RNA-Seq of Borrelia burgdorferi in Multiple Phases of Growth Reveals Insights into the Dynamics of Gene Expression, Transcriptome Architecture, and Noncoding RNAs. PLoS One 11: e0164165.
    OpenUrlCrossRef
  3. ↵
    Babb, K., McAlister, J., Miller, J., and Stevenson, B. (2004) Molecular characterization of Borrelia burgdorferi erp promoter/operator elements. J Bacteriol 186: 2745–2756.
    OpenUrlAbstract/FREE Full Text
  4. ↵
    Barbour, A., Tessier, S., and Todd, W. (1983) Lyme disease spirochetes and ixodid tick spirochetes share a common surface antigenic determinant defined by a monoclonal antibody. Infect Immun 41: 795–804.
    OpenUrlAbstract/FREE Full Text
  5. ↵
    Barbour, A.G. (1984) Isolation and cultivation of Lyme disease spirochetes. Yale J Biol Med 57: 521–525.
    OpenUrlPubMedWeb of Science
  6. ↵
    Barcena-Uribarri, I., Thein, M., Maier, E., Bonde, M., Bergström, S., and Benz, R. (2013) Use of nonelectrolytes reveals the channel size and oligomeric constitution of the Borrelia burgdorferi P66 porin. PLoS One 8: e78272.
    OpenUrlCrossRefPubMed
  7. ↵
    Blevins, J., Revel, A., Smith, A., Bachlani, G., and Norgard, M. (2007) Adaptation of a luciferase gene reporter and lac expression system to Borrelia burgdorferi. Appl Environ Microbiol 73: 1501–1513.
    OpenUrlAbstract/FREE Full Text
  8. ↵
    Blin, K., Pedersen, L.E., Weber, T., and Lee, S.Y. (2016) CRISPy-web: An online resource to design sgRNAs for CRISPR applications. Synth Syst Biotechnol 1: 118–121.
    OpenUrl
  9. ↵
    Bono, J., Tilly, K., Stevenson, B., Hogan, D., and Rosa, P. (1998) Oligopeptide permease in Borrelia burgdorferi: putative peptide-binding components encoded by both chromosomal and plasmid loci. Microbiology 144: 1033–1044.
    OpenUrlCrossRefPubMedWeb of Science
  10. ↵
    Bunikis, I., Denker, K., Östberg, Y., Andersen, C., Benz, R., and Bergström, S. (2008) An RND-type efflux system in Borrelia burgdorferi is involved in virulence and resistance to antimicrobial compounds. PLoS Pathog 4: e1000009.
    OpenUrlCrossRefPubMed
  11. ↵
    Bunikis, J., and Barbour, A. (1999) Access of antibody or trypsin to an integral outer membrane protein (P66) of Borrelia burgdorferi is hindered by Osp lipoproteins. Infect Immun 67: 2874–2883.
    OpenUrlAbstract/FREE Full Text
  12. ↵
    Bunikis, J., Luke, C., Bunikiene, E., Bergström, S., and Barbour, A. (1998) A surface-exposed region of a novel outer membrane protein (P66) of Borrelia spp. is variable in size and sequence. J Bacteriol 180: 1618–1623.
    OpenUrlAbstract/FREE Full Text
  13. ↵
    Bunikis, J., Noppa, L., and Bergström, S. (1995) Molecular analysis of a 66-kDa protein associated with the outer membrane of Lyme disease Borrelia. FEMS Microbiol Lett 131: 139–145.
    OpenUrlCrossRefPubMedWeb of Science
  14. ↵
    Bunikis, J., Noppa, L., Östberg, Y., Barbour, A., and Bergström, S. (1996) Surface exposure and species specificity of an immunoreactive domain of a 66-kilodalton outer membrane protein (P66) of the Borrelia spp. that cause Lyme disease. Infect Immun 64: 5111–5116.
    OpenUrlAbstract/FREE Full Text
  15. ↵
    Carroll, J.A. (2010) Methods of identifying membrane proteins in spirochetes. Curr Protoc Microbiol Chapter 2: Unit12C 12.
  16. ↵
    Charon, N., Goldstein, S., Marko, M., Hsieh, C., Gebhardt, L., Motaleb, M., Wolgemuth, C., Limberger, R., and Rowe, N. (2009) The flat-ribbon configuration of the periplasmic flagella of Borrelia burgdorferi and its relationship to motility and morphology. J Bacteriol 191: 600–607.
    OpenUrlAbstract/FREE Full Text
  17. ↵
    Chen, S., Kumru, O., and Zückert, W.R. (2011) Determination of borrelia surface lipoprotein anchor topology by surface proteolysis. J Bacteriol 193: 6379–6383.
    OpenUrlAbstract/FREE Full Text
  18. ↵
    Chen, S., and Zückert, W.R. (2011) Probing the Borrelia burgdorferi surface lipoprotein secretion pathway using a conditionally folding protein domain. J Bacteriol 193: 6724–6732.
    OpenUrlAbstract/FREE Full Text
  19. ↵
    Chimalakonda, G., Ruiz, N., Chng, S.S., Garner, R.A., Kahne, D., and Silhavy, T.J. (2011) Lipoprotein LptE is required for the assembly of LptD by the beta-barrel assembly machine in the outer membrane of Escherichia coli. Proc Natl Acad Sci U S A 108: 2492–2497.
    OpenUrlAbstract/FREE Full Text
  20. ↵
    Chng, S.S., Ruiz, N., Chimalakonda, G., Silhavy, T.J., and Kahne, D. (2010) Characterization of the two-protein complex in Escherichia coli responsible for lipopolysaccharide assembly at the outer membrane. Proc Natl Acad Sci U S A 107: 5363–5368.
    OpenUrlAbstract/FREE Full Text
  21. ↵
    Chng, S.S., Xue, M., Garner, R.A., Kadokura, H., Boyd, D., Beckwith, J., and Kahne, D. (2012) Disulfide rearrangement triggered by translocon assembly controls lipopolysaccharide export. Science 337: 1665–1668.
    OpenUrlAbstract/FREE Full Text
  22. ↵
    Cho, S.H., Szewczyk, J., Pesavento, C., Zietek, M., Banzhaf, M., Roszczenko, P., Asmar, A., Laloux, G., Hov, A.K., Leverrier, P., Van der Henst, C., Vertommen, D., Typas, A., and Collet, J.F. (2014) Detecting envelope stress by monitoring beta-barrel assembly. Cell 159: 1652–1664.
    OpenUrlCrossRefPubMed
  23. ↵
    Coburn, J., and Cugini, C. (2003) Targeted mutation of the outer membrane protein P66 disrupts attachment of the Lyme disease agent, Borrelia burgdorferi, to integrin alphavbeta3. Proc Natl Acad Sci U S A 100: 7301–7306.
    OpenUrlAbstract/FREE Full Text
  24. ↵
    Coburn, J., Garcia, B., Hu, L.T., Jewett, M.W., Kraiczy, P., Norris, S.J., and Skare, J. (2021) Lyme Disease Pathogenesis. Curr Issues Mol Biol 42: 473–518.
    OpenUrlCrossRefPubMed
  25. ↵
    D’Enfert, C., and Pugsley, A. (1989) Klebsiella pneumoniae pulS gene encodes an outer membrane lipoprotein required for pullulanase secretion. J Bacteriol 171: 3673–3679.
    OpenUrlAbstract/FREE Full Text
  26. ↵
    d’Enfert, C., Ryter, A., and Pugsley, A. (1987) Cloning and expression in Escherichia coli of the Klebsiella pneumoniae genes for production, surface localization and secretion of the lipoprotein pullulanase. EMBO J 6: 3531–3538.
    OpenUrlPubMedWeb of Science
  27. de Leeuw, B.H., Maraha, B., Hollemans, L., Sprong, H., Brandenburg, A.H., Westenend, P.J., and Kusters, J.G. (2014) Evaluation of Borrelia real time PCR DNA targeting OspA, FlaB and 5S-23S IGS and Borrelia 16S rRNA RT-qPCR. J Microbiol Methods 107: 41–46.
    OpenUrlCrossRef
  28. ↵
    Dong, H., Xiang, Q., Gu, Y., Wang, Z., Paterson, N.G., Stansfeld, P.J., He, C., Zhang, Y., Wang, W., and Dong, C. (2014) Structural basis for outer membrane lipopolysaccharide insertion. Nature 511: 52–56.
    OpenUrlCrossRefPubMedWeb of Science
  29. ↵
    Dowdell, A.S., Murphy, M.D., Azodi, C., Swanson, S.K., Florens, L., Chen, S., and Zückert, W.R. (2017) Comprehensive Spatial Analysis of the Borrelia burgdorferi Lipoproteome Reveals a Compartmentalization Bias toward the Bacterial Surface. J Bacteriol 199.
  30. ↵
    Evans, R., O’Neill, M., Pritzel, A., Antropova, N., Senior, A., Green, T., Žídek, A., Bates, R., Blackwell, S., Yim, J., Ronneberger, O., Bodenstein, S., Zielinski, M., Bridgland, A., Potapenko, A., Cowie, A., Tunyasuvunakool, K., Jain, R., Clancy, E., Kohli, P., Jumper, J., and Hassabis, D. (2022) Protein complex prediction with AlphaFold-Multimer. bioRxiv: 2021.2010.2004.463034.
  31. ↵
    Florens, L., and Washburn, M.P. (2006) Proteomic analysis by multidimensional protein identification technology. Methods Mol Biol 328: 159–175.
    OpenUrlCrossRefPubMed
  32. ↵
    Francetic, O., and Pugsley, A. (2005) Towards the identification of type II secretion signals in a nonacylated variant of pullulanase from Klebsiella oxytoca. J Bacteriol 187: 7045–7055.
    OpenUrlAbstract/FREE Full Text
  33. ↵
    Fraser, C., Casjens, S., Huang, W., Sutton, G., Clayton, R., Lathigra, R., White, O., Ketchum, K., Dodson, R., Hickey, E., Gwinn, M., Dougherty, B., Tomb, J., Fleischmann, R., Richardson, D., Peterson, J., Kerlavage, A., Quackenbush, J., Salzberg, S., Hanson, M., van Vugt, R., Palmer, N., Adams, M., Gocayne, J., Weidman, J., Utterback, T., Watthey, L., McDonald, L., Artiach, P., Bowman, C., Garland, S., Fuji, C., Cotton, M., Horst, K., Roberts, K., Hatch, B., Smith, H., and Venter, J. (1997) Genomic sequence of a Lyme disease spirochaete, Borrelia burgdorferi. Nature 390: 580–586.
    OpenUrlCrossRefPubMedWeb of Science
  34. ↵
    Freinkman, E., Chng, S., and Kahne, D. (2011) The complex that inserts lipopolysaccharide into the bacterial outer membrane forms a two-protein plug-and-barrel. Proc Natl Acad Sci U S A 108: 2486–2491.
    OpenUrlAbstract/FREE Full Text
  35. ↵
    Grabowicz, M., and Silhavy, T.J. (2017) Redefining the essential trafficking pathway for outer membrane lipoproteins. Proc Natl Acad Sci U S A 114: 4769–4774.
    OpenUrlAbstract/FREE Full Text
  36. ↵
    Greene, N.P., Hinchliffe, P., Crow, A., Ababou, A., Hughes, C., and Koronakis, V. (2013) Structure of an atypical periplasmic adaptor from a multidrug efflux pump of the spirochete Borrelia burgdorferi. FEBS Lett 587: 2984–2988.
    OpenUrl
  37. ↵
    Hallgren, J., Tsirigos, K.D., Pedersen, M.D., Almagro Armenteros, J.J., Marcatili, P., Nielsen, H., Krogh, A., and Winther, O. (2022) DeepTMHMM predicts alpha and beta transmembrane proteins using deep neural networks. bioRxiv: 2022.2004.2008.487609.
  38. ↵
    Hara, T., Matsuyama, S., and Tokuda, H. (2003) Mechanism underlying the inner membrane retention of Escherichia coli lipoproteins caused by Lol avoidance signals. J Biol Chem 278: 40408–40414.
    OpenUrlAbstract/FREE Full Text
  39. ↵
    Hawley, K.L., Montezuma-Rusca, J.M., Delgado, K.N., Singh, N., Uversky, V.N., Caimano, M.J., Radolf, J.D., and Luthra, A. (2021) Structural modeling of the Treponema pallidum OMPeome: a roadmap for deconvolution of syphilis pathogenesis and development of a syphilis vaccine. J Bacteriol.
  40. ↵
    Hooda, Y., Lai, C.C., Judd, A., Buckwalter, C.M., Shin, H.E., Gray-Owen, S.D., and Moraes, T.F. (2016) Slam is an outer membrane protein that is required for the surface display of lipidated virulence factors in Neisseria. Nat Microbiol 1: 16009.
    OpenUrl
  41. ↵
    Iqbal, H., Kenedy, M.R., Lybecker, M., and Akins, D.R. (2016) The TamB ortholog of Borrelia burgdorferi interacts with the beta-barrel assembly machine (BAM) complex protein BamA. Mol Microbiol 102: 757–774.
    OpenUrlCrossRef
  42. ↵
    Johnson, M., Zaretskaya, I., Raytselis, Y., Merezhuk, Y., McGinnis, S., and Madden, T.L. (2008) NCBI BLAST: a better web interface. Nucleic Acids Res 36: W5–9.
    OpenUrlCrossRefPubMedWeb of Science
  43. ↵
    Jumper, J., and Hassabis, D. (2022) Protein structure predictions to atomic accuracy with AlphaFold. Nat Methods 19: 11–12.
    OpenUrlCrossRef
  44. ↵
    Kelley, L.A., Mezulis, S., Yates, C.M., Wass, M.N., and Sternberg, M.J. (2015) The Phyre2 web portal for protein modeling, prediction and analysis. Nat Protoc 10: 845–858.
    OpenUrlCrossRefPubMed
  45. ↵
    Kenedy, M.R., Luthra, A., Anand, A., Dunn, J.P., Radolf, J.D., and Akins, D.R. (2014) Structural modeling and physicochemical characterization provide evidence that P66 forms a beta-barrel in the Borrelia burgdorferi outer membrane. J Bacteriol 196: 859–872.
    OpenUrlAbstract/FREE Full Text
  46. ↵
    Kenedy, M.R., Scott, E.J., 2nd, Shrestha, B., Anand, A., Iqbal, H., Radolf, J.D., Dyer, D.W., and Akins, D.R. (2016) Consensus computational network analysis for identifying candidate outer membrane proteins from Borrelia spirochetes. BMC Microbiol 16: 141.
    OpenUrlCrossRef
  47. ↵
    Konovalova, A., Kahne, D.E., and Silhavy, T.J. (2017) Outer Membrane Biogenesis. Annu Rev Microbiol 71: 539–556.
    OpenUrlCrossRefPubMed
  48. ↵
    Konovalova, A., Perlman, D.H., Cowles, C.E., and Silhavy, T.J. (2014) Transmembrane domain of surface-exposed outer membrane lipoprotein RcsF is threaded through the lumen of beta-barrel proteins. Proc Natl Acad Sci U S A 111: E4350–4358.
    OpenUrlAbstract/FREE Full Text
  49. ↵
    Kudryashev, M., Cyrklaff, M., Baumeister, W., Simon, M., Wallich, R., and Frischknecht, F. (2009) Comparative cryo-electron tomography of pathogenic Lyme disease spirochetes. Mol Microbiol 71: 1415–1434.
    OpenUrlCrossRefPubMed
  50. ↵
    Kumru, O., Schulze, R., Rodnin, M., Ladokhin, A., and Zückert, W.R. (2011) Surface Localization Determinants of Borrelia OspC/Vsp Family Lipoproteins. J Bacteriol 193: 2814–2825.
    OpenUrlAbstract/FREE Full Text
  51. ↵
    Kumru, O.S., Schulze, R.J., Slusser, J.G., and Zückert, W.R. (2010) Development and validation of a FACS-based lipoprotein localization screen in the Lyme disease spirochete Borrelia burgdorferi. BMC Microbiol 10: 277.
    OpenUrlCrossRefPubMed
  52. ↵
    Lenhart, T., and Akins, D. (2010) Borrelia burgdorferi locus BB0795 encodes a BamA orthologue required for growth and efficient localization of outer membrane proteins. Mol Microbiol 75: 692–709.
    OpenUrlCrossRefPubMed
  53. ↵
    Lin, T., Gao, L., Zhang, C., Odeh, E., Jacobs, M.B., Coutte, L., Chaconas, G., Philipp, M.T., and Norris, S.J. (2012) Analysis of an ordered, comprehensive STM mutant library in infectious Borrelia burgdorferi: insights into the genes required for mouse infectivity. PLoS One 7: e47532.
    OpenUrlCrossRefPubMed
  54. ↵
    Lin, T., Troy, E.B., Hu, L.T., Gao, L., and Norris, S.J. (2014) Transposon mutagenesis as an approach to improved understanding of Borrelia pathogenesis and biology. Front Cell Infect Microbiol 4: 63.
    OpenUrl
  55. ↵
    Murphy, B.T., Wiepen, J.J., He, H., Pramanik, A.S., Peters, J.M., Stevenson, B., and Zückert, W.R. (2022) Inducible CRISPRi-based operon silencing and selective in trans gene complementation in Borrelia burgdorferi. bioRxiv: 2022.2012.2005.519228.
  56. ↵
    Okuda, S., and Tokuda, H. (2011) Lipoprotein sorting in bacteria. Annu Rev Microbiol 65: 239–259.
    OpenUrlCrossRefPubMedWeb of Science
  57. ↵
    Peters, J.M., Koo, B.M., Patino, R., Heussler, G.E., Hearne, C.C., Qu, J., Inclan, Y.F., Hawkins, J.S., Lu, C.H.S., Silvis, M.R., Harden, M.M., Osadnik, H., Peters, J.E., Engel, J.N., Dutton, R.J., Grossman, A.D., Gross, C.A., and Rosenberg, O.S. (2019) Enabling genetic analysis of diverse bacteria with Mobile-CRISPRi. Nat Microbiol 4: 244–250.
    OpenUrl
  58. ↵
    Pugsley, A., Chapon, C., and Schwartz, M. (1986) Extracellular pullulanase of Klebsiella pneumoniae is a lipoprotein. J Bacteriol 166: 1083–1088.
    OpenUrlAbstract/FREE Full Text
  59. ↵
    Pugsley, A., Kornacker, M., and Ryter, A. (1990) Analysis of the subcellular location of pullulanase produced by Escherichia coli carrying the pulA gene from Klebsiella pneumoniae strain UNF5023. Mol Microbiol 4: 59–72.
    OpenUrlCrossRefPubMed
  60. ↵
    Putker, F., Bos, M.P., and Tommassen, J. (2015) Transport of lipopolysaccharide to the Gram-negative bacterial cell surface. FEMS Microbiol Rev 39: 985–1002.
    OpenUrlCrossRefPubMed
  61. ↵
    Qiao, S., Luo, Q., Zhao, Y., Zhang, X.C., and Huang, Y. (2014) Structural basis for lipopolysaccharide insertion in the bacterial outer membrane. Nature 511: 108–111.
    OpenUrlCrossRefPubMedWeb of Science
  62. ↵
    Radolf, J., Goldberg, M., Bourell, K., Baker, S., Jones, J., and Norgard, M. (1995) Characterization of outer membranes isolated from Borrelia burgdorferi, the Lyme disease spirochete. Infect Immun 63: 2154–2163.
    OpenUrlAbstract/FREE Full Text
  63. ↵
    Rodriguez-Alonso, R., Letoquart, J., Nguyen, V.S., Louis, G., Calabrese, A.N., Iorga, B.I., Radford, S.E., Cho, S.H., Remaut, H., and Collet, J.F. (2020) Structural insight into the formation of lipoprotein-beta-barrel complexes. Nat Chem Biol 16: 1019–1025.
    OpenUrl
  64. ↵
    Roussel-Jazede, V., Grijpstra, J., van Dam, V., Tommassen, J., and van Ulsen, P. (2013) Lipidation of the autotransporter NalP of Neisseria meningitidis is required for its function in the release of cell-surface-exposed proteins. Microbiology 159: 286–295.
    OpenUrlCrossRefPubMedWeb of Science
  65. ↵
    Roussel-Jazede, V., Jongerius, I., Bos, M., Tommassen, J., and van Ulsen, P. (2010) NalP-mediated proteolytic release of lactoferrin-binding protein B from the meningococcal cell surface. Infect Immun 78: 3083–3089.
    OpenUrlAbstract/FREE Full Text
  66. ↵
    Ruiz, N., Chng, S.S., Hiniker, A., Kahne, D., and Silhavy, T.J. (2010) Nonconsecutive disulfide bond formation in an essential integral outer membrane protein. Proc Natl Acad Sci U S A 107: 12245–12250.
    OpenUrlAbstract/FREE Full Text
  67. ↵
    Sambir, M., Ivanova, L.B., Bryksin, A.V., Godfrey, H.P., and Cabello, F.C. (2011) Functional analysis of Borrelia burgdorferi uvrA in DNA damage protection. FEMS Microbiol Lett 317: 172–180.
    OpenUrlCrossRefPubMed
  68. ↵
    Sambrook, J., and Russell, D.W., (2001) Molecular Cloning - A Laboratory Manual. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, New York.
  69. ↵
    Sauvonnet, N., and Pugsley, A. (1996) Identification of two regions of Klebsiella oxytoca pullulanase that together are capable of promoting beta-lactamase secretion by the general secretory pathway. Mol Microbiol 22: 1–7.
    OpenUrlCrossRefPubMed
  70. ↵
    Schulze, R., Chen, S., Kumru, O., and Zückert, W.R. (2010) Translocation of Borrelia burgdorferi surface lipoprotein OspA through the outer membrane requires an unfolded conformation and can initiate at the C-terminus. Mol Microbiol 76: 1266–1278.
    OpenUrlCrossRefPubMed
  71. ↵
    Schulze, R., and Zückert, W.R. (2006) Borrelia burgdorferi lipoproteins are secreted to the outer surface by default. Mol Microbiol 59: 1473–1484.
    OpenUrlCrossRefPubMed
  72. ↵
    Skare, J., Mirzabekov, T., Shang, E., Blanco, D., Erdjument-Bromage, H., Bunikis, J., Bergström, S., Tempst, P., Kagan, B., Miller, J., and Lovett, M. (1997) The Oms66 (p66) protein is a Borrelia burgdorferi porin. Infect Immun 65: 3654–3661.
    OpenUrlAbstract/FREE Full Text
  73. ↵
    Skare, J., Shang, E., Foley, D., Blanco, D., Champion, C., Mirzabekov, T., Sokolov, Y., Kagan, B., Miller, J., and Lovett, M. (1995) Virulent strain associated outer membrane proteins of Borrelia burgdorferi. J Clin Invest 96: 2380–2392.
    OpenUrlCrossRefPubMedWeb of Science
  74. ↵
    Srinivas, N., Jetter, P., Ueberbacher, B.J., Werneburg, M., Zerbe, K., Steinmann, J., Van der Meijden, B., Bernardini, F., Lederer, A., Dias, R.L., Misson, P.E., Henze, H., Zumbrunn, J., Gombert, F.O., Obrecht, D., Hunziker, P., Schauer, S., Ziegler, U., Kach, A., Eberl, L., Riedel, K., DeMarco, S.J., and Robinson, J.A. (2010) Peptidomimetic antibiotics target outer-membrane biogenesis in Pseudomonas aeruginosa. Science 327: 1010–1013.
    OpenUrlAbstract/FREE Full Text
  75. ↵
    Takacs, C.N., Scott, M., Chang, Y., Kloos, Z.A., Irnov, I., Rosa, P.A., Liu, J., and Jacobs-Wagner, C. (2020) A CRISPR interference platform for selective downregulation of gene expression in Borrelia burgdorferi. Appl Environ Microbiol 87.
  76. ↵
    Takayama, K., Rothenberg, R., and Barbour, A. (1987) Absence of lipopolysaccharide in the Lyme disease spirochete, Borrelia burgdorferi. Infect Immun 55: 2311–2313.
    OpenUrlAbstract/FREE Full Text
  77. ↵
    Tatusov, R.L., Galperin, M.Y., Natale, D.A., and Koonin, E.V. (2000) The COG database: a tool for genome-scale analysis of protein functions and evolution. Nucleic Acids Res 28: 33–36.
    OpenUrlCrossRefPubMedWeb of Science
  78. ↵
    Teufel, F., Almagro Armenteros, J.J., Johansen, A.R., Gislason, M.H., Pihl, S.I., Tsirigos, K.D., Winther, O., Brunak, S., von Heijne, G., and Nielsen, H. (2022) SignalP 6.0 predicts all five types of signal peptides using protein language models. Nat Biotechnol 40: 1023–1025.
    OpenUrl
  79. ↵
    Tokuda, H., and Matsuyama, S. (2004) Sorting of lipoproteins to the outer membrane in E. coli. Biochim Biophys Acta 1694: IN1–9.
    OpenUrlCrossRefPubMed
  80. ↵
    Toledo, A., Crowley, J.T., Coleman, J.L., LaRocca, T.J., Chiantia, S., London, E., and Benach, J.L. (2014) Selective association of outer surface lipoproteins with the lipid rafts of Borrelia burgdorferi. MBio 5: e00899–00814.
    OpenUrlPubMed
  81. ↵
    Toledo, A., Huang, Z., Benach, J.L., and London, E. (2018a) Analysis of Lipids and Lipid Rafts in Borrelia. Methods Mol Biol 1690: 69–82.
    OpenUrlCrossRef
  82. ↵
    Toledo, A., Huang, Z., Coleman, J.L., London, E., and Benach, J.L. (2018b) Lipid rafts can form in the inner and outer membranes of Borrelia burgdorferi and have different properties and associated proteins. Mol Microbiol 108: 63–76.
    OpenUrlCrossRefPubMed
  83. ↵
    Toledo, A., Perez, A., Coleman, J.L., and Benach, J.L. (2015) The lipid raft proteome of Borrelia burgdorferi. Proteomics 15: 3662–3675.
    OpenUrlCrossRef
  84. ↵
    Urfer, M., Bogdanovic, J., Lo Monte, F., Moehle, K., Zerbe, K., Omasits, U., Ahrens, C.H., Pessi, G., Eberl, L., and Robinson, J.A. (2016) A Peptidomimetic Antibiotic Targets Outer Membrane Proteins and Disrupts Selectively the Outer Membrane in Escherichia coli. J Biol Chem 291: 1921–1932.
    OpenUrlAbstract/FREE Full Text
  85. ↵
    van Ulsen, P., van Alphen, L., ten Hove, J., Fransen, F., van der Ley, P., and Tommassen, J. (2003) A Neisserial autotransporter NalP modulating the processing of other autotransporters. Mol Microbiol 50: 1017–1030.
    OpenUrlCrossRefPubMedWeb of Science
  86. ↵
    Wilson, A., and Ruiz, N. (2022) The transmembrane alpha-helix of LptC participates in LPS extraction by the LptB(2) FGC transporter. Mol Microbiol 118: 61–76.
    OpenUrl
  87. ↵
    Wilson, M.M., and Bernstein, H.D. (2016) Surface-Exposed Lipoproteins: An Emerging Secretion Phenomenon in Gram-Negative Bacteria. Trends Microbiol 24: 198–208.
    OpenUrlCrossRef
  88. ↵
    Wu, T., McCandlish, A., Gronenberg, L., Chng, S., Silhavy, T., and Kahne, D. (2006) Identification of a protein complex that assembles lipopolysaccharide in the outer membrane of Escherichia coli. Proc Natl Acad Sci U S A 103: 11754–11759.
    OpenUrlAbstract/FREE Full Text
  89. ↵
    Yang, J., and Zhang, Y. (2015) I-TASSER server: new development for protein structure and function predictions. Nucleic Acids Res 43: W174–181.
    OpenUrlCrossRefPubMed
  90. ↵
    Zha, Z., Li, C., Li, W., Ye, Z., and Pan, J. (2016) LptD is a promising vaccine antigen and potential immunotherapeutic target for protection against Vibrio species infection. Sci Rep 6: 38577.
    OpenUrl
  91. ↵
    Zhang, Y., Wen, Z., Washburn, M.P., and Florens, L. (2010) Refinements to label free proteome quantitation: how to deal with peptides shared by multiple proteins. Anal Chem 82: 2272–2281.
    OpenUrlCrossRefPubMed
  92. ↵
    Zielke, R.A., Wierzbicki, I.H., Baarda, B.I., Gafken, P.R., Soge, O.O., Holmes, K.K., Jerse, A.E., Unemo, M., and Sikora, A.E. (2016) Proteomics-driven Antigen Discovery for Development of Vaccines Against Gonorrhea. Mol Cell Proteomics 15: 2338–2355.
    OpenUrlAbstract/FREE Full Text
  93. ↵
    Zückert, W.R. (2007) Laboratory maintenance of Borrelia burgdorferi. Curr Protoc Microbiol Chapter 12: Unit 12C.11.
  94. ↵
    Zückert, W., Lloyd, J., Stewart, P., Rosa, P., and Barbour, A. (2004) Cross-species surface display of functional spirochetal lipoproteins by recombinant Borrelia burgdorferi. Infect Immun 72: 1463–1469.
    OpenUrlAbstract/FREE Full Text
  95. ↵
    Zückert, W.R. (2014) Secretion of bacterial lipoproteins: through the cytoplasmic membrane, the periplasm and beyond. Biochim Biophys Acta 1843: 1509–1516.
    OpenUrlCrossRefPubMedWeb of Science
  96. ↵
    Zückert, W.R. (2019) Protein Secretion in Spirochetes. Microbiol Spectr 7.
Back to top
Posted December 22, 2022.
Download PDF

Supplementary Material

Email
A Borrelia burgdorferi LptD Homolog Facilitates Flipping of Surface Lipoproteins Through the Spirochetal Outer Membrane
Huan He, Ankita S. Pramanik, Selene K. Swanson, David K. Johnson, Laurence Florens, Wolfram R. Zückert
bioRxiv 2022.12.21.521298; doi: https://doi.org/10.1101/2022.12.21.521298
Twitter logo Facebook logo LinkedIn logo Mendeley logo
Citation Tools

Subject Area