• We are sorry, but NCBI web applications do not support your browser and may not function properly. More information
Logo of nihpaAbout Author manuscriptsSubmit a manuscriptNIH Public Access; Author Manuscript; Accepted for publication in peer reviewed journal;
Mol Microbiol. Author manuscript; available in PMC Jul 1, 2011.
Published in final edited form as:
PMCID: PMC2909360
NIHMSID: NIHMS207682

The crystal structures of two salivary cystatins from the tick Ixodes scapularis and the effect of these inhibitors on the establishment of Borrelia burgdorferi infection in a murine model

Summary

We have previously demonstrated that two salivary cysteine protease inhibitors from the Borrelia burgdorferi (Lyme disease) vector Ixodes scapularis-namely sialostatins L and L2-play an important role in tick biology, as demonstrated by the fact that silencing of both sialostatins in tandem results in severe feeding defects. Here we show that sialostatin L2 -but not sialostatin L- facilitates the growth of Borrelia burgdorferi in murine skin. To examine the structural basis underlying these differential effects of the two sialostatins, we have determined the crystal structures of both sialostatin L and L2. This is the first structural analysis of cystatins from an invertebrate source. Sialostatin L2 crystallizes as a monomer with an ‘unusual’ conformation of the N-terminus, while sialostatin L crystallizes as a domain-swapped dimer with an N-terminal conformation similar to other cystatins. Deletion of the ‘unusual’ N-terminal five residues of sialostatin L2 results in marked changes in its selectivity, suggesting that this region is a particularly important determinant of the biochemical activity of sialostatin L2. Collectively, our results reveal the structure of two tick salivary components that facilitate vector blood feeding and that one of them also supports pathogen transmission to the vertebrate host.

Introduction

Completion of a blood meal by the tick Ixodes scapularis requires an extended period of feeding during which it can transmit the pathogenic spirochete Borrelia burgdorferi. While attached to the host, the tick must overcome both short term responses such as the hemostatic system, and long term defenses such as acquired immunity. The saliva secreted by the tick is made up of a complex mixture of proteins and peptides that have, in a number of cases, been proven to affect the host hemostatic, inflammatory and immune systems (Francischetti et al., 2009). Many of these have been identified as putative protease inhibitors based on sequence similarity with known inhibitors, but few have been functionally characterized. Their presence in the saliva suggests that they may target some of the essential processes enumerated above.

One group of inhibitors that have been characterized from tick saliva contains cystatin-like molecules that were named sialostatin L and sialostatin L2 (Kotsyfakis et al., 2007, Kotsyfakis et al., 2006). Cystatins are inhibitors of C1-type cysteine proteases including the mammalian cathepsins (Turk et al., 2008). Chicken egg white cystatin as well as the mammalian cystatins C, E/M and F also inhbit cysteine proteases of the legumain (C13) family (Chen et al., 1997). The structures of several vertebrate cystatins have been determined, either as free molecules or in complex with target proteases (Alvarez-Fernandez et al., 2005, Bode et al., 1988, Janowski et al., 2001, Jenko Kokalj et al., 2007, Jenko et al., 2003, Sanders et al., 2004, Schuttelkopf et al., 2006). The binding interface for the cystatin-cysteine protease interaction is a wedge-shaped structure made up of three parts including the N-terminus and two loops which link individual strands of a five-stranded β-sheet (Bode et al., 1988). Binding of legumain is mediated by a completely distinct interaction involving an asparagine residue at position 39 of cystatin C (Alvarez-Fernandez et al., 1999)

A number of cystatin forms have been shown to form domain-swapped dimers, and the structures of some of these have been determined. It appears that dimer formation may be related to the formation of amyloid fibrils by cystatins. Hereditary cystatin C amyloid angiopathy is a degenerative disease formed by deposition of cystatin C amyloid, and is related to a destabilizing Leu 68 to glutamine mutation in the protein (Ekiel & Abrahamson, 1996, Wei et al., 1998).

Most of the information about the function of cystatins comes from studies on the vertebrate proteins that are divided into three groups based on sequence relatedness. Type 1 cystatins are intracellular proteins also referred to as stefins (Abrahamson et al., 2003, Turk et al., 1985). Type 2 cystatins are secreted, and appear to be most similar to sialostatins L and L2. Type 3 cystatins occur as individual domains in larger proteins, known as kininogens. Although they are most similar to the type 2 cystatins, sialostatins show specificity profiles that are different than vertebrate cystatins, and also show sequence differences in regions that are normally quite conserved in the type 2 group. More specifically, both sialostatins have been shown to be potent inhibitors of cathepsin L (Kotsyfakis et al., 2007, Kotsyfakis et al., 2006), but sialostatin L also strongly inhibits cathepsin S, while sialostatin L2 is a weak inhibitor of this enzyme. Also, both sialostatins do not inhibit cathepsins B and H, thus having rather stringent target specificity (Kotsyfakis et al., 2007, Kotsyfakis et al., 2006).

Various cathepsins play important roles in antigen processing, maturation of the MHC complex, tissue remodeling and angiogenesis (Hsing & Rudensky, 2005, Obermajer et al., 2008). Accordingly, studies with murine dendritic cells, which do not express cathepsin L (Hsing & Rudensky, 2005), showed that only sialostatin L was capable of interfering with antigen presentation. Further analyses demonstrated that the processing of the invariant chain of MHC class II, a process known to be mediated by cathepsin S, was impaired in the presence of this protein (Sa-Nunes et al., 2009). Although it shows no effect in the modulation of antigen presenting cells, sialostatin L2 appears to be essential for successful feeding (Kotsyfakis et al., 2007). Expression of this gene in the salivary gland is strongly induced in feeding ticks, and its silencing by RNA interference resulted in impaired feeding, poor growth and increased mortality.

Apart from supporting tick blood feeding, the salivary secretion has been shown to facilitate the transmission of various pathogens in the vertebrate host, including the etiologic agent of Lyme disease Borrelia burgdorferi (Rupprecht et al., 2008). Although knowledge of the salivary constituents that account for this action is very important for the development of novel tools (e.g. vaccines) in the fight against borreliosis, their identification is rather time-consuming. Some salivary proteins of ixodid ticks are known to play specific roles in infection by Borrelia. For example, Salp15 facilitates infection by binding to the spirochete surface protein OspC, thereby preventing antibody-mediated interaction in vitro, as well as promoting infection in vivo (Ramamoorthi et al., 2005). The same protein has also been described as an immunosuppressant and has been shown to interact with the lymphocyte receptors CD4 and DC-SIGN (Garg et al., 2006, Hovius et al., 2008). The immunosuppressive effects of the sialostatins and their supportive role in tick feeding suggest that they could be important in influencing Borrelia infection through the inhibition of particular proteases involved in host defensive responses.

In this study, we describe for the first time the crystal structures of two invertebrate cystatins i.e., sialostatins L and L2 and relate the similarities and differences between sialostatins and vertebrate cystatins to the differences in target specificity of the proteins. We also evaluate the effect of the two salivary sialostatins on Borrelia infection, revealing sialostatin L2 as a positive regulator for pathogen establishment in the skin of mice.

Results

Crystal Structure of Sialostatin L2

The structure of sialostatin L2 was solved using SAD methods on a data set collected with the selenomethionine derivative of the L22,47,100M triple mutant protein (Table 1). The resulting model revealed a structure similar in many ways to the type 1 and 2 cystatins of vertebrates, which are characterized by a five-stranded antiparallel β-sheet curving around an α-helix oriented nearly perpendicular to the sheet. Two disulfide bonds are present in sialostatin L2, with Cys 70 being linked with Cys 81, and Cys 92 being linked with Cys 111 (Fig 1, Fig. 2A). These disulfide positions are conserved in vertebrate type 1, 2 and 3 cystatins. While the overall similarity of sialostatin L2 with other cystatins is high, some significant modifications are present that may relate to the specificity of the protein as an inhibitor of C1 cysteine proteases. In stefins and type 2 cystatins, three contact points form a wedge-like structure which sits in the substrate binding groove of the enzyme (Bode et al., 1990, Jenko et al., 2003, Stubbs et al., 1990). The first part of the interaction surface is formed by the N-terminal peptide extending to a glycine residue (Gly 6 of sialostatin L2) which acts as a hinge (Fig. 1, Fig. 2). The second structural component consists of the loop between strands 2 and 3 of the β-sheet (loop 1), and the third component is the loop between β-strands 4 and 5 (loop 2) (Fig. 2C). The loop 1 region of sialostatin L2 contains the conserved sequence motif Gln-X1-Val-X2-Gly, with X1 and X2 being Thr 53 and Ala 55. Loop 1 is quite similar in conformation to chicken egg white cystatin and cystatin F (Bode et al., 1988, Schuttelkopf et al., 2006). Loop 2 is characterized in types 2 and 3 cystatins by the presence of a Pro-Trp dipeptide that is not present in sialostatin L2, but is replaced by the sequence Asn 99-Leu 100. In comparison to other type 2 cystatins (cystatins C, D and egg white cystatin) β-strands 4 and 5 extend further toward loop 2, leaving only two residues to form a β-turn which would be less flexible and would probably extend less deeply into the peptidase binding site (Fig. 2C). Egg white cystatin, and cystatins F, D, and C (Ekiel et al., 1997, Hall et al., 1998) have four residues lying outside of the β-sheet that would present a presumably more flexible loop structure to the peptidase. In the vertebrate type 2 cystatins the apex of loop 2 contains the dipeptide Trp-(Leu, Glu, Gln), while in sialostatin L2 it has the sequence Leu-Gln (Fig. 1, Fig. 2C).

Fig. 1
Alignment of sialostatins L (SialoL) and L2 (SialoL2) with vertebrate type 2 cystatins chicken egg white cystatin (Chicken) and cystatin C (CystC). In chicken egg white cystatin, only the residues observed in the crystal structure of Bode et al. (Bode ...
Fig. 2
Structure of sialostatin L2. A. Two views of a ribbon diagram of the protein with the disulfide positions shown in stick format with the sulfur atoms colored in yellow. The positions of the N-and C-terminals as well as loops 1 and 2 of the peptidase interaction ...
Table 1
Data collection, phasing and refinement statistics for sialostatin L (sialo L) and L2 (sialo L2).

The most dramatic difference between sialostatin L2 and other cystatins lies in the N-terminal peptide which takes on an apparently novel conformation (Fig. 2B, ,3).3). Rather than forming an extended coil or loop that remains in-line with loops 1 and 2, this segment is packed against the β-sheet on the side opposite the helical segment. The structure is stabilized by hydrophobic interactions involving the side chain of Leu 2 which projects into a pocket formed by the side chains of Val 54 and Tyr 97. Additionally, the carbonyl groups of Leu 2 and Gly 6, form hydrogen bonds with the side chains of β-sheet residues Asn 58 and Gln 52, respectively (Fig 2B), further stabilizing the conformation of the N-terminal peptide. Crystal contacts may also play a role in stabilizing this conformation. Two intermolecular hydrogen bonds are also present between the peptide amide of Ala 3 and the carbonyl oxygen of Val 44 from a symmetry related molecule as well as between the amide group of Arg 5 and the carbonyl oxygen of Ala 83 of a second molecule. The packing arrangement of this segment suggests both diminished conformational flexibility, and an altered position of, what is in other cystatins, a critical part of the peptidase binding surface. Recombinant sialostatin L2 also contains a methionine (Met 0) residue at the N-terminus that is not present in the tick-derived protein after processing of the signal peptide (Fig. 3). This residue is stabilized in the crystal structure by a hydrogen bond between the terminal amino group and the carbonyl oxygen of Asn 13 of a symmetry-related molecule. The effect of this residue on the conformation of the N-terminus of sialostatin L2 is not known.

Fig. 3
Space-filling diagram of sialostatin L2 showing the packing of the N-terminal region. The two views shown are related by an approximately 90° rotation around the horizontal axis. The terminal seven residues are labeled and colored light blue.

Crystal Structure of Sialostatin L

Sialostatin L was crystallized in the space group C2221 and the structure determined by molecular replacement using the sialostatin L2 structure as a search model. The initial protein model had four monomers in the asymmetric unit, with some adjacent molecules coming into close proximity at the apex of loop 1 connecting β-strands 2 and 3 of the β-sheet. Examination of the electron density in this area showed clearly that adjacent monomeric units were in fact rearranged to form a domain-swapped dimeric structure (Fig. 4). In this configuration, the portion of the molecule corresponding to a sialostatin monomer is formed from a protein domain made up of parts of two monomers (Fig. 4). The N-terminal region, the α-helix, and β-strands 1 and 2 originate from one cystatin molecule, while the remaining three β-strands and the C-terminus are derived from the second molecule (Fig. 4). The region corresponding to β-strands 2 and 3 in the monomer forms a single β-strand extending through the dimer resulting in a long antiparallel β-sheet containing one strand derived from each monomer (Fig. 4). Superposition of the sialostatin L2 model with a single domain of the dimer structure corresponding to a cystatin monomer showed the two to be quite similar, with an RMS deviation of 0.7 Å for 85 Cα positions, indicating a high degree of conservation of the monomeric structure in each domain of the dimer (Fig. 4).

Fig. 4
Crystal structure of sialostatin L. A. Ribbon diagram of the domain swapped dimer. The two views are related by an approximately 90° rotation around the horizontal axis. The two peptide chains making up the dimer are colored blue and magenta. ...

The topology of the sialostatin L dimer is identical to the domain swapped structures reported for stefin A and cystatin C (Janowski et al., 2005, Janowski et al., 2001, Staniforth et al., 2001). The interface formed within each domain is referred to as the closed interface because the observed atomic interactions resemble those of the “closed” monomeric protein, while the interface formed by the long two-stranded β-sheet is referred to as the open interface. Cubic and tetragonal crystal forms of dimeric cystatin C show different orientations of the domains relative to one another due to varying degrees of twisting in the β-sheet of the open interface as a result of differences in crystal environment (Janowski et al., 2005, Janowski et al., 2001). The sialostatin L structure is most similar to the tetragonal form with an RMS deviation of 3.4 Å for 79 cα positions contained in the β-sheet and α-helical regions.

The domain swapped dimeric forms of cystatins are inactive as inhibitors of C1 cysteine proteases since dimerization disrupts the protease binding surface of the cystatin. We observed no significant loss of activity of sialostatin L suggesting that the monomeric form is stable in solution. Analysis by gel filtration chromatography showed that only a small fraction of sialostatin L is dimerized at a concentration of 10 mg/ml, indicating that domain-swapped dimers efficiently form under the conditions of crystallization but not in solution, even at relatively high protein concentrations (Fig. 4).

Comparison of the N-terminal portions of sialostatins L and L2 show the two to differ in a manner that may affect their specificity as inhibitors. The N-terminus of sialostatin L does not pack against the β-sheet as in sialostatin L2, and appears more similar to other cystatins whose structures have been determined. Instead, the N-terminus is turned away from the sheet, and becomes disordered from Val 3 to the N-terminus, indicating a high degree of conformational flexibility over this range (Fig. 4). The structures of sialostatins L and L2 are nearly identical over the span from Gly 6 to Arg 10 (in sialostatin L), but prior to Gly 6 the two chains diverge, with sialostatin L taking on a conformation more similar to that of other type 2 cystatins including egg white cystatin, cystatin D and cystatin F, than to sialostatin L2. A monomeric model of sialostatin L was constructed from the half-dimer structure and shows a protease interaction interface that is substantially different than that of sialostatin L2 in both surface charge and shape. Electrostatic calculations made with APBS for the two proteins using default configuration settings and an ionic concentration of 150 mM show that the broad, relatively flat surface of sialostatin L2 is more positively charged than the corresponding surface of the sialostatin L monomer (Fig. 5).

Fig. 5
Surface characteristics of sialostatins. A. Molecular surface of sialostatin L2 colored by surface potential calculated using the program APBS implemented in Pymol. Areas of the surface with negative electrostatic potential are colored in red, while those ...

Structure and Inhibitory Specificity

When a panel of potential targets was tested, the sialostatins were found to be very selective for particular C1-type cysteine proteases. Sialostatin L inhibits papain, cathepsin L and cathepsin S, while sialostatin L2 inhibits only papain and cathepsin L with high potency. Sialostatin L2 shows a similar degree of discrimination between cathepsins L and S as the small-molecule inhibitor LHVS, with sialostatin L2 inhibiting cathepsin L more strongly and LHVS inhibiting cathepsin S more strongly (Fig. 6, Table 2). Both sialostatins show no inhibition of cathepsins B and H. On the amino acid sequence level, the two proteins are approximately 75 % identical, with the sequences of loops 1 and 2 of the peptidase-binding interface being absolutely identical in sequence. Of the structural features thought to be involved in binding, the N-terminal portion segment shows the least sequence conservation, and the most structural variation suggesting that this region may dictate differences in the target specificities of the two inhibitors.

Fig. 6
Concentration dependence of the inhibition of various cysteine proteases by sialostatins L and L2, truncated sialostatin L2, chicken egg white cystatin and LHVS. All the experiments comparing the effect of these cysteine protease inhibitors on a specific ...
Table 2
Inibition of various C1 cysteine proteases by sialostatin L (sialoL), sialostatin L2 (sialoL2), truncated sialostatin L2 (sialoL2T), chicken egg white cystatin (CEWC), and LHVS. Values represent the calculated inhibitor concentration to achieve 50 % inhibition ...

In vertebrate cystatins, the N-terminal segment is known to be an important determinant of specificity and inhibitory potency (Hall et al., 1993, Hall et al., 1998, Hall et al., 1995, Machleidt et al., 1989). To test the importance of this region in sialostatins L and L2, we constructed a truncated mutant in which residues 1–5 were deleted. The modified protein was evaluated as an inhibitor of papain and cathepsins B, C, H, L and S, and compared to wild-type sialostatins L and L2, chicken egg white cystatin and LHVS (Fig. 6, Table 2). Deletion of the N-terminal peptide of sialostatin L2 had variable effects on the activity of a number of cysteine proteases. While the IC50 values for papain and cathepsin L and cathepsin S were increased by 1000, 56, and 50-fold, respectively, the IC50 value for cathepsin C was reduced by 60 times (Fig. 6, Table 2). Full-length sialostatin L2 showed no detectable inhibition of cathepsin H, but when the N-terminal fragment was shortened an IC50 of 6 μM was observed, suggesting that the N-terminal segment of sialostatin L2 sterically hinders binding with cathepsin C and cathepsin H.

Sialostatins and Infection by Borrelia burgdorferi

Sialostatin L shows pronounced immunosuppressive and anti-inflammatory effects in host animals, presumably due to the inhibition of cathepsin S. This protein, but not sialostatin L2, inhibits LPS-induced dendritic cell maturation by interfering with the processing of the MHC class II invariant chain (Sa-Nunes et al., 2009). Although it lacks an effect in antigen presentation by dendritic cells, sialostatin L2 dramatically affects the feeding success of ticks, possibly by influencing the activity of extracellular or intracellular cathepsins involved in inflammatory responses, tissue remodeling or angiogenesis (Kotsyfakis et al., 2007). The importance of these factors in the establishment of a viable infection by B. burgdorferi has not been evaluated previously. We therefore tested the effect of the two sialostatins on infection by separately injecting mice with either sialostatin L or sialostatin L2 prior to, concurrently, and after injection with infectious spirochetes. After 4 days, the numbers of spirochetes in the skin were evaluated (Fig. 7A). The levels of Borrelia infection did not differ significantly after injection of either vehicle alone, or vehicle containing sialostatin L along with the spirochete. However, when vehicle plus sialostatin L2 was injected, an increase in mean spirochete number of almost six-fold, as determined by quantitative PCR, was observed in skin samples indicating a facilitating role of this inhibitor upon B. burgdorferi infection (Fig. 7A). Sialostatin L2 alone does not appear to stimulate or inhibit in vitro proliferation of cultured spirochetes, since no effect on cell number was seen when sialostatin L2 was added to media containing spirochetes relative to a control containing an equal concentration of ovalbumin in place of sialostatin L2 (Fig. 7B).

Fig. 7
Effect of sialostatins L and L2 on Borrelia burgdorferi. A. Sialostatins and infection by B. burgdorferi in mice. Quantification of mean number of spirochetes in murine skin samples by real-time PCR in animals treated with sialostatin L, sialostatin L2 ...

The possibility that the stimulatory effect of sialostatin L2 was due to direct interaction of the protein with the spirochete in a manner similar to Salp15 (Ramamoorthi et al., 2005) was tested by incubating Borrelia suspensions on coverslips coated with sialostatin L, sialostatin L2 or chicken egg white albumin. No significant differences were detected in the numbers of B. burgdorferi cells adhering to the coverslips, suggesting that sialostatin L2, unlike Salp15, does not bind directly to spirochetes (Fig. 7C).

Discussion

Salivary cystatins from I. scapularis show relatively narrow inhibitory selectivity compared to the well studied vertebrate cystatins, cystatin C and chicken egg white cystatin. Sialostatin L2 is a potent inhibitor of cathepsin L and papain but not of other C1-type cysteine proteases, while sialostatin L inhibits cathepsin S in addition to papain and cathepsin L. The broadly selective cystatin C, on the other hand, inhibits cathepsins B, H, L and S with inhibition constants of less than 10 nM (Hall et al., 1998, Hall et al., 1995). Sialostatins L and L2 are highly similar at the amino acid sequence level but differences in the structures at the putative protease interaction interface are quite dramatic. In the sialostatin L2 structure, the N-terminal region folds back across the β-sheet rather than forming a third interaction loop in line with loops 1 and 2, creating a surface unlike that of other known cystatins. This unusual binding surface is also positively charged, thereby facilitating interaction with the strongly anionic cathepsin L substrate binding cleft (Schuttelkopf et al., 2006). Like sialostatin L2, the p41 fragment from the invariant chain of MHC class II is a potent inhibitor of cathepsin L but not of cathepsin S (Bevec et al., 1996, Ogrinc et al., 1993). This protein forms a wedge-like structure similar to the cystatins that inserts into the substrate binding groove of the protease (Mihelic et al., 2008). The lower part of the wedge contains a number of positively charged residues that are in position to interact with the negatively charged surface of the protease (Mihelic et al., 2008). Introduction of negatively charged residues into the binding grove of cathepsin S by site-directed mutagenesis results in a nearly 1000-fold enhancement in affinity, verifying the role of electrostatics in this interaction (Mihelic et al., 2008).

The structure of sialostatin L was obtained only as a domain-swapped dimer, although gel filtration experiments indicated that the protein is monomeric in solution. A variety of vertebrate cystatins form domain-swapped dimers, a process thought to produce in pathogenic amyloid fibrils in some cases. The dimers assemble into tetramers and other higher order structures that may be incorporated into fibrils (Jenko Kokalj et al., 2007, Sanders et al., 2004). The dimers formed by sialostatin L are very similar to those formed by other cystatins. However, isomerization of a conserved proline in loop 2 is critical for the assembly of tetrameric stefin B This residue is highly conserved in vertebrate cystatin forms, and may be of general importance in fibril formation. Interestingly, no proline is present at this position in Ixodes salivary cystatins. This would perhaps limit the ability of this protein to form higher-order oligomers.

Because only the domain-swapped dimer was observed in the crystal, the position of the N-terminus in the monomer cannot be determined with absolute certainty. Nevertheless the N-terminal segment of the sialostatin L dimer clearly does not fold back across the β-sheet in the manner of sialostatin L2. If the same is true of the inhibitory monomer, its N-terminal conformation is more similar to the structurally-characterized vertebrate cystatins. The interaction interface of sialostatin L forms a wedge-like structure with an in-line arrangement of the N-terminus, loop 1 and loop 2, rather than the broader, flatter surface of sialostatin L2 where the N-terminus extends along-side loops 1 and 2. This region is not as positively charged as in sialostatin L2 suggesting that electrostatic considerations may not be as important for cathepsin L binding of sialostatin L as sialostatin L2 (Fig. 5). Consistent with this, the surface potential of substrate binding cleft of cathepsin S does not carry the strong negative charge seen in cathepsin L (Schuttelkopf et al., 2006).

The importance of the N-terminal end of sialostatin L2 was tested directly in mutagenesis experiments. Truncation of sialostatin L2 at its N-terminus changes its specificity; improving its interaction with some cysteine proteases while diminishing interaction with others. Previous studies with vertebrate cystatins indicated that the region N-terminal to Gly 5 in sialostatin L2 is a particularly important determinant of both inhibitory potency and specificity (Auerswald et al., 1994, Hall et al., 1993, Hall et al., 1998, Hall et al., 1995, Machleidt et al., 1989). Notably, truncation of the N-terminus of sialostatin L2 weakens rather than improves its interaction with cathepsin S. This indicates that weakened inhibition of cathepsin S by sialostatin L2 is not simply a matter of steric occlusion due to unusual positioning of the N-terminal region. Truncation of the N-terminus of sialostatin L2 also reduces the inhibition of cathepsin L by a factor of 325 suggesting that, as with other cystatins, specific interactions with the N-terminus are important in this case, and further supporting the idea that the binding modes of sialostatins L and L2 must be somewhat different. With cathepsin C on the other hand, truncation of sialostatin L2 improved the interaction by a factor of approximately 60, making it very similar to sialostatin L in its affinity. It appears that steric hindrance of binding due to the unusual N-terminus may be important in reducing the affinity of sialostatin L2 in this case.

Ixodes scapularis produces two very similar cystatins that have different inhibition profiles. Correspondingly, the crystal structures of these proteins show significant structural differences in the putative peptidase binding interface, particularly in the N-terminal region. The electrostatic and steric features of this region appear to give the two proteins their unique inhibitory characteristics. Removal of the N-terminus of sialostatin L2 dramatically alters the inhibitory specificity of sialostatin L2. It decreases the inhibitory potency of the protein toward cathepsins S and L, which are also targets of sialostatin L, but it improves the interaction of the protein with cathepsin C. Overall, the stringent target specificity of sialostatins compares with that of LHVS, a chemical inhibitor developed as a highly specific cathepsin S inhibitor. Importantly, this inhibitor targets also cathepsin B (Figure 7) and calpain (data not shown), which is not the case for sialostatins. The unique selectivity of sialostatins can lead to potential pharmaceutical applications with respect to vertebrate immunity (Sa-Nunes et al., 2009), while the herein resolved structure -for the first time for non vertebrate cystatins- will be the basis for potential further improvement through directed mutagenesis.

Salivary proteins secreted by hard ticks are known to play a variety of anti-inflammatory and anti-hemostatic roles in the ingestion of a blood meal. While the inhibition of inflammation, immunity, angiogenesis, and tissue remodeling is demonstrably important for successful feeding, individual salivary constituents are also likely to have effects on the survival of arthropod-vectored pathogens, and have not been as thoroughly investigated in this regard. Inhibition of lysosomal proteases in antigen presenting cells can affect processing of proteins for presentation, and proteases secreted by leukocytes into the extracellular matrix are important in the activation of various defensive responses. Here we show that coadministration of sialostatin L2 but not sialostatin L exacerbates skin infections of Borrelia in a murine model.

Unlike sialostatin L, and similar to Salp15 that also affects Borrelia transmission, expression of sialostatin L2 is strongly induced during feeding, suggesting that it is particularly important in the period when spirochetes are being transmitted. Unlike Salp15, however, it does not directly bind to spirochetes. The mechanism of growth stimulation is not clear at this point, but previous studies suggest that it may not involve suppression of antigen presentation by dendritic cells, since sialostatin L2 does not inhibit this process (Sa-Nunes et al., 2009). Nevertheless, knockdown of sialostatin gene expression is known to result in a severe defect in tick feeding in that ticks fail to grow, and up to 40 % die while attached to the host when sialostatins L and L2 were silenced in tandem (Kotsyfakis et al., 2007). In this case, at least part of the effect appears to be due to an enhanced inflammatory response to tick saliva. Collectively these data combined with the data presented herein reveal an arthropod disease vector protein, sialostatin L2, that is essential for tick blood feeding and assists Borrelia transmission to the vertebrate host as well. In other words, a tick-borne pathogen takes advantage of a mechanism essential for the adaptation of its vector to the vertebrate host. However, other functions pertaining to the maintenance of a viable feeding site for the long feeding period of ixodid ticks have not been examined with respect to sialostatin L2 contributions such as epidermal homeostasis, angiogenesis and cell chemotaxis. Indeed, cathepsins L and S are important mediators of the digestion of the extracellular matrix and angiogenesis (Obermajer et al., 2008, Urbich et al., 2005). We are currently searching for additional sialostatin L2 targets and functions, as well as examining the interplay between different salivary proteins with regard to spirochete infection in order to answer these questions.

Materials and Methods

Production of Recombinant Proteins

Recombinant sialostatins L and L2 were produced as inclusion bodies in Escherichia coli as described previously (Kotsyfakis et al., 2007, Kotsyfakis et al., 2006). The proteins were dissolved in 6 M guanidinium hydrochloride and reduced by adding dithiothreitol to a concentration of 10 mM. After 30 m incubation at room temperature, the solution was added dropwise to a large excess of 20 mM Tris HCl, pH 8.0, 0.3 M arginine hydrochloride. After two days at 4°C, the refolded protein solution was concentrated by ultrafiltration to a volume of 10–20 mL. Sialostatins L and L2 were then purified by gel filtration chromatography on Sephacryl S-100 followed by ion exchange chromatography on Q-Sepharose.

Crystallization of Sialostatins

Both sialostatin L and L2 were crystallized using the hanging drop vapor diffusion method. Sialostatin L2 was crystallized from 1.8 M ammonium sulfate, 0.1 M bicine pH 9.0, while sialostatin L was crystallized from 19 % PEG 6000 0.1 M sodium citrate, pH 5.3. Sialostatin L crystals were soaked briefly in the crystallization condition containing 10 % glycerol prior to flash freezing in liquid nitrogen. Sialostatin L2 crystals were moved through three changes of 30 % PEG 6000, 0.1 M Tris HCl pH 9.0, 15 % glycerol prior to flash freezing. For phasing of the sialostatin L2 diffraction data, a selenomethionine derivative was prepared. No methionine is present in the wild-type sequence, so a triple mutant variant cDNA was synthesized. In the mutant form, leucine residues were replaced by methionine at positions 22, 47 and 100. The protein was produced in the E. coli strain B834(DE3)pLys S. Cells were grown in SelenoMet medium (Athena ES) containing selenomethionine, inclusion bodies were harvested, and protein was refolded and purified as described above. The selenomethionine derivative of the triple mutant crystallized in the same condition as the wild-type protein.

Data Collection and Structure Solution

Diffraction data for sialostatins L and L2 were collected at beamline 22-BM of the Southeast Regional Collaborative Access Team (SER-CAT) and beamline 19-BM of the Structural Biology Center at the Advanced Photon Source, Argonne National Laboratory. Sialostatin L2 crystallized in the space group I23 with a single protein molecule in the asymmetric unit, while sialostatin L crystallized in the space group C2221 with two dimeric molecules in the asymmetric unit. Diffraction images were processed using HKL-2000 (Otwinowski & Minor, 1997). The structure of sialostatin L2 was determined from the selenium derivative of the triple mutant described above by using single anomalous dispersion (SAD) methodology on a dataset collected at a wavelength near the selenium X-ray absorption edge. Selenium positions were determined using SHELXD (Schneider & Sheldrick, 2002). Phasing and density modification were performed using SHELXE (Sheldrick, 2002). Parts of the structure were built using ARP-WARP (Cohen et al., 2004), and the remainder was built manually in Coot (Emsley & Cowtan, 2004). The structure was refined using Refmac (Murshudov et al., 1997). In the latter stages of refinement a TLS model was used, incorporating a single TLS group.

The structure of sialostatin L was determined by molecular replacement using the program Phaser (McCoy et al., 2005) with the sialostatin L2 structure as a starting model. The model was refined using Refmac and CNS (Brunger et al., 1987). After the initial stages of refinement it became clear that the asymmetric unit contained two molecules of a domain-swapped dimer rather than four monomeric units. The model was corrected to reflect this fact and the refinement was continued.

Coordinates and structure factors for sialostatin L2 and sialostatin L have been deposited with the accession codes 3MWZ (selenomethionine derivative of the L 22,47,100 M mutant of sialostatin L2), 3LH4 (wild-type native sialostatin L2) and 3LI7 (sialostatin L) in the RCSB Protein Data Bank.

Functional Analyses of Sialostatins

Purified human cathepsin H, recombinant human cathepsin S, and recombinant human cathepsin L (Calbiochem);purified bovine cathepsins B and C and purified papain; (Sigma) were assayed in 100 mM sodium acetate, pH 5.5, 100 mM NaCl, 1 mM EDTA, 1 mg/ml cysteine, 0.005% Triton X-100. Calculation of the final molar concentration of the enzymes used during the assays was done as described previously (Kotsyfakis et al., 2006). IC50 values of the inhibitors for various cysteine proteases were obtained as described earlier (Kotsyfakis et al., 2006) by measuring the loss of enzymatic activity at increasing concentrations of inhibitor in the presence of a fluorogenic substrate in large excess. The assays were performed in a 96-well plate-reading fluorimeter in volumes of 50 μL and each enzyme was pre-incubated with the various inhibitors for 10 minutes at room temperature. The assays were initiated and developed by the addition of the corresponding fluorogenic substrate and the reaction was allowed to proceed at 30°C. The substrates, used in a 0.25 mM final concentration, were H-Arg-7-amino-4-methylcoumarin for cathepsin H (purchased from Calbiochem); N-carbobenzyloxy-Leu-Arg-7-amino-4-methylcoumarin for cathepsin B, L, C and papain (purchased from R & D Systems); N-carbobenzyloxy-Val-Val-Arg-7-amino-4-methylcoumarin for cathepsin S (purchased from Bachem Bioscience). The linear (r square higher than 0.95) hydrolysis rate of the substrate was followed for 20 minutes in a Spectramax Gemini XPS 96 well plate fluorescence reader (Molecular Devices) using 365 nm excitation and 450nm emission wavelength with a cutoff at 435 nm. All experiments were performed in triplicate and the results plotted as the mean of each triplicate ± standard error of mean (SEM). All the experiments comparing the effect of various cysteine protease inhibitors on a specific cysteine protease were performed in the same day and same plate under exactly the same experimental conditions. Morpholinurea-leucine-homophenylalanine-vinylsulfone-phenyl (LHVS) was purchased from Arris Pharmaceutical Inc and purified chicken egg white cystatin from Sigma. To determine the effect of the N-terminus of sialostatin L2 on inhibitory activity, a truncated mutant was constructed with the sequence encoding the first five amino acid residues deleted. The first for N-terminal amino acids of the mutant are MGGY rather than MELA of the wild type (point of sialostatin L2 truncation is highlighted in magenta in Fig. 1)

Statistical Analysis

All data are expressed as the mean ± S.E.M and each experiment was performed at least three times. Statistical significance was determined by Student’s t test; differences in multiple comparisons among different experimental groups were determined by analysis of variance using the Tukey test. A value of p < 0.05 was considered statistically significant.

Effects of Sialostatins on B. burgdorferi Infection

Specific-pathogen-free mice C3H/HeN (Charles River) were used through the experiments (at 22°C and a relative humidity of 65 %). The B31 strain of Borrelia burgdorferi sensu stricto was cultivated in BSK-H medium (Sigma) supplemented with 6 % rabbit serum and 1 % antibiotic mixture for Borrelia (Sigma) at 34 °C. The ninth passage of B31 spirochetes was used for the experiments (Machackova et al., 2006). Fifteen C3H mice were infected intradermally at the dorsal thoracic midline with a dose of 5 × 103 spirochetes in BSK-H medium, at a total volume 10 μl. Five hours prior to Borrelia inoculation, concomitant with the inoculation and 5 hours after the inoculation mice received i.d. injections of SialoL, SialoL2 or PBS (10 μg/injection) into the same sites of the back. On day 4 post infection, skin samples from the site of inoculation, urinary bladder, kidney and heart muscle were collected from each mouse. After subcutaneous fat was scraped off the skin, all tissues were weighed and stored on ice for the following isolation of DNA. Genomic DNA from murine tissues and ticks was extracted with High Pure PCR Template Preparation Kit (Roche, Germany) according to manufacturer’s instructions.

For the quantitative analysis of extracted DNA, real-time PCR was performed by amplifying a portion of the FliD gene (flagellar hook-associated protein II, GenBank accession U.66699.1), which is localized in a single copy on the chromosome of Borrelia burgdorferi sensu stricto. A species-specific pair of primers (forward primer 5′-TGG TGA CAG AGT GTA TGA TAA TGG AA-3′ (Zeidner et al., 2002) and reverse primer 5′-CCT TCC ACT TTT CTC TCT CTA TCT T-3′ (Machackova et al., 2006) were used to generate a 189 bp fragment of the FliD gene, using the internal TaqMan® probe (5′FAM-ACT TAA AAT GCT AGG AGA TTA TCT GTC GCC- BHQ 3′ (Zeidner et al., 2002). The q-PCR mixture contained 4 μl of template DNA, 0.5 U Hot Start Taq polymerase (TaKaRa), polymerase buffer (TaKaRa), primers 2μM each (Generi Biotech), 500 nM dual-labeled probe (Generi-Biotech), dNTP mixture 200 nM each (TaKaRa) and water to the final volume of 20 μl. Amplification was performed in a Rotor Gene 3000 thermocycler (Corbett Research). The standard amplification program included an initial denaturation for 10 min at 95 °C and 45 cycles consisting of a denaturation step for 15 s at 95 °C and annealing and synthesis steps for 1 min at 60 °C. All measurements were performed in triplicate. To quantify the copy number of the target gene, the standard curve was generated using several 10-fold dilutions of DNA isolated from cultured spirochetes of known concentrations. The complete PCR mixture without template DNA served as a negative control. The measured data were analyzed with Rotor Gene 5.0 software (Corbett Research).

Bactericidal Activity of Sialostatins

Sialostatins L and L2 were tested for their bactericidal activity against Borrelia burgdorferi sensu stricto B31 by dark-field microscopy. Spirochetes were harvested, counted and diluted to the final concentration 5×106/ml. Two hundred microliters of bacterial suspension were incubated at 33 °C in sterile 1.5 ml microtubes in the presence or absence of SialoL/L2 at a final concentration of 500 μg/ml. Chicken ovalbumin added to spirochetes in the same concentration served as a negative control. After 1 h, 4 h, 24 h, 48 h and 72 h of incubation, samples were examined for viable spirochetes. Ten random fields were counted per each sample. The experiment was carried out in triplicates and repeated two times.

Binding Assay

The binding capacity of Borrelia spirochetes to the sialostatins was tested as described by Brissette et al. (Brissette et al., 2009) with minor modifications. Briefly, glass microsope slides were coated with 20 μg/ml sialostatin L or sialostatin L2 in PBS. Control slides were coated similarly with chicken egg albumin (OVA, Sigma-Aldrich) or PBS only. The slides were allowed to dry overnight. The following day, slides were washed three times with PBS and then blocked by incubation with 3% (w/v) bovine serum albumin (BSA, Sigma-Aldrich) for 2 h at room temperature. Spirochetes from the B31 culture were harvested and resuspended in PBS to 2 × 106 bacteria/ml. Slides were covered with bacteria, incubated at 37 °C for 2 h and then washed 8 times with PBS. Bound bacteria were visualized by dark field microscopy. Numbers of bacteria observed in five 400× fields per slide (five slides per each protein sample) were counted in a blinded manner by two independent persons.

All animal experiments were performed under the guidelines and with the approval of the committee at the Institute of Parasitology, Biology Centre of the Academy of Sciences of Czech Republic, , České Budĕjovice Czech Republic.

Acknowledgments

The authors acknowledge Rosanne Hearn for help with protein expression, and José Ribeiro, Ivo Francischetti and Eric Calvo for helpful discussions. This work was supported by the intramural research program of NIAID, National Institutes of Health. Experiments describing effect of sialostatins on infection by Borrelia burgdorferi were supported by grant nos. KJB500960702 and IAA600960811 (GAASCR). We also thank Drs. David Garboczi, Apostolos Gittis and Kavita Singh of the NIAID Research Technologies Branch as well as the staffs of the Structural Biology Center Collaborative Access Team, and the Southeast Regional Collaborative Access Team, Advanced Photon Source, Argonne National Laboratory for assistance with X-ray data collection. Use of the Advanced Photon Source beamlines was supported by the U.S. Department of Energy, Office of Science, Office of Basic Energy Sciences, under contract no. W-31-109-Eng-38.

References

  • Abrahamson M, Alvarez-Fernandez M, Nathanson CM. Cystatins. Biochem Soc Symp. 2003:179–199. [PubMed]
  • Alvarez-Fernandez M, Barrett AJ, Gerhartz B, Dando PM, Ni J, Abrahamson M. Inhibition of mammalian legumain by some cystatins is due to a novel second reactive site. J Biol Chem. 1999;274:19195–19203. [PubMed]
  • Alvarez-Fernandez M, Liang YH, Abrahamson M, Su XD. Crystal structure of human cystatin D, a cysteine peptidase inhibitor with restricted inhibition profile. J Biol Chem. 2005;280:18221–18228. [PubMed]
  • Auerswald EA, Nagler DK, Schulze AJ, Engh RA, Genenger G, Machleidt W, Fritz H. Production, inhibitory activity, folding and conformational analysis of an N-terminal and an internal deletion variant of chicken cystatin. Eur J Biochem. 1994;224:407–415. [PubMed]
  • Bevec T, Stoka V, Pungercic G, Dolenc I, Turk V. Major histocompatibility complex class II-associated p41 invariant chain fragment is a strong inhibitor of lysosomal cathepsin L. J Exp Med. 1996;183:1331–1338. [PMC free article] [PubMed]
  • Bode W, Engh R, Musil D, Laber B, Stubbs M, Huber R, Turk V. Mechanism of interaction of cysteine proteinases and their protein inhibitors as compared to the serine proteinase-inhibitor interaction. Biol Chem Hoppe Seyler. 1990;371(Suppl):111–118. [PubMed]
  • Bode W, Engh R, Musil D, Thiele U, Huber R, Karshikov A, Brzin J, Kos J, Turk V. The 2.0 A X-ray crystal structure of chicken egg white cystatin and its possible mode of interaction with cysteine proteinases. Embo J. 1988;7:2593–2599. [PMC free article] [PubMed]
  • Brissette CA, Bykowski T, Cooley AE, Bowman A, Stevenson B. Borrelia burgdorferi RevA antigen binds host fibronectin. Infect Immun. 2009;77:2802–2812. [PMC free article] [PubMed]
  • Brunger AT, Kuriyan J, Karplus M. Crystallographic R Factor Refinement by Molecular Dynamics. Science. 1987;235:458–460. [PubMed]
  • Chen JM, Dando PM, Rawlings ND, Brown MA, Young NE, Stevens RA, Hewitt E, Watts C, Barrett AJ. Cloning, isolation, and characterization of mammalian legumain, an asparaginyl endopeptidase. J Biol Chem. 1997;272:8090–8098. [PubMed]
  • Cohen SX, Morris RJ, Fernandez FJ, Ben Jelloul M, Kakaris M, Parthasarathy V, Lamzin VS, Kleywegt GJ, Perrakis A. Towards complete validated models in the next generation of ARP/wARP. Acta Crystallogr D Biol Crystallogr. 2004;60:2222–2229. [PubMed]
  • Ekiel I, Abrahamson M. Folding-related dimerization of human cystatin C. J Biol Chem. 1996;271:1314–1321. [PubMed]
  • Ekiel I, Abrahamson M, Fulton DB, Lindahl P, Storer AC, Levadoux W, Lafrance M, Labelle S, Pomerleau Y, Groleau D, LeSauteur L, Gehring K. NMR structural studies of human cystatin C dimers and monomers. J Mol Biol. 1997;271:266–277. [PubMed]
  • Emsley P, Cowtan K. Coot: model-building tools for molecular graphics. Acta Crystallogr D Biol Crystallogr. 2004;60:2126–2132. [PubMed]
  • Francischetti IM, Sa-Nunes A, Mans BJ, Santos IM, Ribeiro JM. The role of saliva in tick feeding. Front Biosci. 2009;14:2051–2088. [PMC free article] [PubMed]
  • Garg R, Juncadella IJ, Ramamoorthi N, Ashish, Ananthanarayanan SK, Thomas V, Rincon M, Krueger JK, Fikrig E, Yengo CM, Anguita J. Cutting edge: CD4 is the receptor for the tick saliva immunosuppressor, Salp15. J Immunol. 2006;177:6579–6583. [PubMed]
  • Hall A, Dalboge H, Grubb A, Abrahamson M. Importance of the evolutionarily conserved glycine residue in the N-terminal region of human cystatin C (Gly-11) for cysteine endopeptidase inhibition. Biochem J. 1993;291(Pt 1):123–129. [PMC free article] [PubMed]
  • Hall A, Ekiel I, Mason RW, Kasprzykowski F, Grubb A, Abrahamson M. Structural basis for different inhibitory specificities of human cystatins C and D. Biochemistry. 1998;37:4071–4079. [PubMed]
  • Hall A, Hakansson K, Mason RW, Grubb A, Abrahamson M. Structural basis for the biological specificity of cystatin C. Identification of leucine 9 in the N-terminal binding region as a selectivity-conferring residue in the inhibition of mammalian cysteine peptidases. J Biol Chem. 1995;270:5115–5121. [PubMed]
  • Hovius JW, de Jong MA, den Dunnen J, Litjens M, Fikrig E, van der Poll T, Gringhuis SI, Geijtenbeek TB. Salp15 binding to DC-SIGN inhibits cytokine expression by impairing both nucleosome remodeling and mRNA stabilization. PLoS Pathog. 2008;4:e31. [PMC free article] [PubMed]
  • Hsing LC, Rudensky AY. The lysosomal cysteine proteases in MHC class II antigen presentation. Immunol Rev. 2005;207:229–241. [PubMed]
  • Janowski R, Kozak M, Abrahamson M, Grubb A, Jaskolski M. 3D domain-swapped human cystatin C with amyloidlike intermolecular beta-sheets. Proteins. 2005;61:570–578. [PubMed]
  • Janowski R, Kozak M, Jankowska E, Grzonka Z, Grubb A, Abrahamson M, Jaskolski M. Human cystatin C, an amyloidogenic protein, dimerizes through three-dimensional domain swapping. Nat Struct Biol. 2001;8:316–320. [PubMed]
  • Jenko Kokalj S, Guncar G, Stern I, Morgan G, Rabzelj S, Kenig M, Staniforth RA, Waltho JP, Zerovnik E, Turk D. Essential role of proline isomerization in stefin B tetramer formation. J Mol Biol. 2007;366:1569–1579. [PubMed]
  • Jenko S, Dolenc I, Guncar G, Dobersek A, Podobnik M, Turk D. Crystal structure of Stefin A in complex with cathepsin H: N-terminal residues of inhibitors can adapt to the active sites of endo- and exopeptidases. J Mol Biol. 2003;326:875–885. [PubMed]
  • Kotsyfakis M, Karim S, Andersen JF, Mather TN, Ribeiro JM. Selective cysteine protease inhibition contributes to blood-feeding success of the tick Ixodes scapularis. J Biol Chem. 2007;282:29256–29263. [PubMed]
  • Kotsyfakis M, Sa-Nunes A, Francischetti IM, Mather TN, Andersen JF, Ribeiro JM. Antiinflammatory and immunosuppressive activity of sialostatin L, a salivary cystatin from the tick Ixodes scapularis. J Biol Chem. 2006;281:26298–26307. [PubMed]
  • Machackova M, Obornik M, Kopecky J. Effect of salivary gland extract from Ixodes ricinus ticks on the proliferation of Borrelia burgdorferi sensu stricto in vivo. Folia Parasitol (Praha) 2006;53:153–158. [PubMed]
  • Machleidt W, Thiele U, Laber B, Assfalg-Machleidt I, Esterl A, Wiegand G, Kos J, Turk V, Bode W. Mechanism of inhibition of papain by chicken egg white cystatin. Inhibition constants of N-terminally truncated forms and cyanogen bromide fragments of the inhibitor. FEBS Lett. 1989;243:234–238. [PubMed]
  • McCoy AJ, Grosse-Kunstleve RW, Storoni LC, Read RJ. Likelihood-enhanced fast translation functions. Acta Crystallogr D Biol Crystallogr. 2005;61:458–464. [PubMed]
  • Mihelic M, Dobersek A, Guncar G, Turk D. Inhibitory fragment from the p41 form of invariant chain can regulate activity of cysteine cathepsins in antigen presentation. J Biol Chem. 2008;283:14453–14460. [PubMed]
  • Murshudov GN, Vagin AA, Dodson EJ. Refinement of macromolecular structures by the maximum-likelihood method. Acta Crystallogr D Biol Crystallogr. 1997;53:240–255. [PubMed]
  • Obermajer N, Jevnikar Z, Doljak B, Kos J. Role of cysteine cathepsins in matrix degradation and cell signalling. Connect Tissue Res. 2008;49:193–196. [PubMed]
  • Ogrinc T, Dolenc I, Ritonja A, Turk V. Purification of the complex of cathepsin L and the MHC class II-associated invariant chain fragment from human kidney. FEBS Lett. 1993;336:555–559. [PubMed]
  • Otwinowski Z, Minor W. Processing of X-ray diffraction data collected in oscillation mode. Methods Enzymol. 1997;276:307–326.
  • Ramamoorthi N, Narasimhan S, Pal U, Bao F, Yang XF, Fish D, Anguita J, Norgard MV, Kantor FS, Anderson JF, Koski RA, Fikrig E. The Lyme disease agent exploits a tick protein to infect the mammalian host. Nature. 2005;436:573–577. [PubMed]
  • Rupprecht TA, Koedel U, Fingerle V, Pfister HW. The pathogenesis of lyme neuroborreliosis: from infection to inflammation. Mol Med. 2008;14:205–212. [PMC free article] [PubMed]
  • Sa-Nunes A, Bafica A, Antonelli LR, Choi EY, Francischetti IM, Andersen JF, Shi GP, Chavakis T, Ribeiro JM, Kotsyfakis M. The immunomodulatory action of sialostatin L on dendritic cells reveals its potential to interfere with autoimmunity. J Immunol. 2009;182:7422–7429. [PMC free article] [PubMed]
  • Sanders A, Jeremy Craven C, Higgins LD, Giannini S, Conroy MJ, Hounslow AM, Waltho JP, Staniforth RA. Cystatin forms a tetramer through structural rearrangement of domain-swapped dimers prior to amyloidogenesis. J Mol Biol. 2004;336:165–178. [PubMed]
  • Schneider TR, Sheldrick GM. Substructure solution with SHELXD. Acta Crystallogr D Biol Crystallogr. 2002;58:1772–1779. [PubMed]
  • Schuttelkopf AW, Hamilton G, Watts C, van Aalten DM. Structural basis of reduction-dependent activation of human cystatin F. J Biol Chem. 2006;281:16570–16575. [PubMed]
  • Sheldrick GM. Macromolecular phasing with SHELXE. Z Kristallogr. 2002;217:644–650.
  • Staniforth RA, Giannini S, Higgins LD, Conroy MJ, Hounslow AM, Jerala R, Craven CJ, Waltho JP. Three-dimensional domain swapping in the folded and molten-globule states of cystatins, an amyloid-forming structural superfamily. Embo J. 2001;20:4774–4781. [PMC free article] [PubMed]
  • Stubbs MT, Laber B, Bode W, Huber R, Jerala R, Lenarcic B, Turk V. The refined 2.4 A X-ray crystal structure of recombinant human stefin B in complex with the cysteine proteinase papain: a novel type of proteinase inhibitor interaction. Embo J. 1990;9:1939–1947. [PMC free article] [PubMed]
  • Turk V, Brzin J, Lenarcic B, Locnikar P, Popovic T, Ritonja A, Babnik J, Bode W, Machleidt W. Structure and function of lysosomal cysteine proteinases and their protein inhibitors. Prog Clin Biol Res. 1985;180:91–103. [PubMed]
  • Turk V, Stoka V, Turk D. Cystatins: biochemical and structural properties, and medical relevance. Front Biosci. 2008;13:5406–5420. [PubMed]
  • Urbich C, Heeschen C, Aicher A, Sasaki K, Bruhl T, Farhadi MR, Vajkoczy P, Hofmann WK, Peters C, Pennacchio LA, Abolmaali ND, Chavakis E, Reinheckel T, Zeiher AM, Dimmeler S. Cathepsin L is required for endothelial progenitor cell-induced neovascularization. Nat Med. 2005;11:206–213. [PubMed]
  • Wei L, Berman Y, Castano EM, Cadene M, Beavis RC, Devi L, Levy E. Instability of the amyloidogenic cystatin C variant of hereditary cerebral hemorrhage with amyloidosis, Icelandic type. J Biol Chem. 1998;273:11806–11814. [PubMed]
  • Zeidner NS, Schneider BS, Nuncio MS, Gern L, Piesman J. Coinoculation of Borrelia spp. with tick salivary gland lysate enhances spirochete load in mice and is tick species-specific. J Parasitol. 2002;88:1276–1278. [PubMed]
PubReader format: click here to try

Formats:

Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...

Links

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...