Published Ahead of Print 21 April 2008.
2008, 76(7):2888. DOI: 10.1128/IAI.00232-08. Infect. Immun.
Nandhini Ramamoorthi, Erol Fikrig and Alje P. van Dam
Tim J. Schuijt, Joppe W. R. Hovius, Nathalie D. van Burgel,
Sensu Lato Isolates burgdorferi
Borreliathe Killing of Serum-Sensitive
The Tick Salivary Protein Salp15 Inhibits
http://iai.asm.org/content/76/7/2888
Updated information and services can be found at:
These include:
REFERENCES
http://iai.asm.org/content/76/7/2888#ref-list-1
This article cites 28 articles, 8 of which can be accessed free at:
CONTENT ALERTS
more»articles cite this article),
Receive: RSS Feeds, eTOCs, free email alerts (when new
http://iai.asm.org/site/misc/reprints.xhtmlInformation about commercial reprint orders:
http://journals.asm.org/site/subscriptions/To subscribe to to another ASM Journal go to:
o
n
D
ecem
ber 5, 2011 by guest
http://iai.asm
.org/
D
ow
nloaded from
INFECTION AND IMMUNITY, July 2008, p. 2888–2894 Vol. 76, No. 7
0019-9567/08/$08.00�0 doi:10.1128/IAI.00232-08
Copyright © 2008, American Society for Microbiology. All Rights Reserved.
The Tick Salivary Protein Salp15 Inhibits the Killing of Serum-Sensitive
Borrelia burgdorferi Sensu Lato Isolates�
Tim J. Schuijt,1* Joppe W. R. Hovius,2,3 Nathalie D. van Burgel,1 Nandhini Ramamoorthi,3
Erol Fikrig,3 and Alje P. van Dam1
Department of Medical Microbiology, Leiden University Medical Centre, Leiden, The Netherlands1; Center for Experimental and
Molecular Medicine (CEMM), University of Amsterdam, AMC, Amsterdam, The Netherlands2; and Section of Infectious Diseases,
Department of Internal Medicine, Yale University School of Medicine, New Haven, Connecticut3
Received 18 February 2008/Returned for modification 20 March 2008/Accepted 8 April 2008
Borrelia burgdorferi, the agent of Lyme disease, is transmitted by ticks. During transmission from the tick to
the host, spirochetes are delivered with tick saliva, which contains the salivary protein Salp15. Salp15 has been
shown to protect spirochetes against B. burgdorferi-specific antibodies. We now show that Salp15 from both
Ixodes ricinus and Ixodes scapularis protects serum-sensitive isolates of Borrelia burgdorferi sensu lato against
complement-mediated killing. I. ricinus Salp15 showed strong protective effects compared to those of I.
scapularis Salp15. Deposition of terminal C5b to C9 (one molecule each of C5b, C6, C7, and C8 and one or more
molecules of C9) complement complexes, part of the membrane attack complex, on the surface of B. burgdorferi
was inhibited in the presence of Salp15. In the presence of normal human serum, serum-sensitive Borrelia
burgdorferi requires protection against complement-mediated killing, which is provided, at least in part, by the
binding to the tick salivary protein Salp15.
The Lyme disease agent Borrelia burgdorferi survives in a
tick-mouse cycle. In the United States, B. burgdorferi sensu
stricto is maintained primarily in Ixodes scapularis ticks, while
the European vector of B. burgdorferi sensu stricto, B. garinii,
and B. afzelii strains are generally Ixodes ricinus ticks. Feeding
of ixodid ticks normally takes several days (2), which gives the
host immune system time to react to the arthropod. The ticks
have developed several mechanisms to evade both the innate
and adaptive host responses, which enable them to take an
effective blood meal. Tick saliva possesses proteins with immu-
nosuppressive (14, 18), anticomplement (5, 19, 27), and anti-
hemostatic (21, 22) activity. Salp15, a feeding-induced tick
salivary protein, is known to inhibit CD4� T-cell activation and
proliferation by specifically binding to the CD4 coreceptor of
the T cells (1, 6, 13). Also, Salp15 appeared to enhance the
survival of B. burgdorferi in the host after transmission by the
tick by specifically interacting with B. burgdorferi outer surface
protein C (OspC) and providing protection against borrelia-
cidal antibodies (25). Recently we found three Salp15 homo-
logues in I. ricinus ticks (12), and one of these homologues,
Salp15 Iric-1, showed 80% similarity to I. scapularis Salp15
(Iscap Salp15) at the DNA level.
The innate response, the complement system in particular,
plays a crucial role in the eradication of invading pathogens.
The complement system is important in the initiation of attack
against B. burgdorferi. The spirochetes are opsonized and also
directly killed by the formation of the lytic pore-forming mem-
brane attack complex (MAC) (3, 23). B. burgdorferi sensu
stricto, B. garinii, and B. afzelii isolates are able to activate
complement both by the classical pathway and by the alterna-
tive pathway in nonimmune human serum (NHS) in the ab-
sence of specific antibodies, but they differ in susceptibility to
complement-mediated killing (28). Serum-resistant Borrelia
strains are able to evade complement-mediated killing by bind-
ing to complement regulators of the alternative complement
pathway, i.e., binding factor H and factor H-like protein-1
(FHL-1) to CRASP-1Bb (15), CRASP-2Bb (9), OspE (10),
and/or CRASP-3Bb (16) proteins, or by expressing a CD59-like
complement inhibitory molecule (24). The split products after
complement activation are also important because of chemo-
taxis and the infiltration of immune cells in the Borrelia-in-
fected tissue. Altogether, there are several reasons for the
spirochetes to protect themselves against complement activa-
tion. In this study, we show that the tick salivary protein Salp15
plays a role in the protection of serum-sensitive B. garinii
strains and intermediately resistant B. burgdorferi strains
against direct killing by the complement system.
MATERIALS AND METHODS
Borrelia isolates and growth conditions. Serum-sensitive strains B. garinii A87S
and VSBP and intermediately resistant strains B. burgdorferi VS215 and B31
were used in this study. Both B. garinii strains are human isolates, while both B.
burgdorferi strains are tick isolates. Spirochetes were cultivated at 33°C in Bar-
bour-Stoenner-Kelley medium plus sodium bicarbonate (BSK-H medium) sup-
plemented with 6% rabbit serum (Sigma).
Purification of recombinant I. scapularis and I. ricinus Salp15. For the puri-
fication of Iscap Salp15 (GenBank accession number AAK97817), salp15 was
cloned in frame in Drosophila melanogaster cells in conjunction with a His tag, a
V5 epitope, and a resistance gene for hygromycin as described previously (1).
Salp15 Iric-1 (GenBank accession number ABU93613) was purified using Dro-
sophila cells expressing salp15 together with a resistance gene for blastomycin
(J. W. Hovius, T. J. Schuijt, K. A. de Groot, J. T. T. H. Roelofs, A. Oei, J. A.
Marquart, C. van �t Veer, T. van der Poll, N. Ramamoorthi, E. Fikrig, and A. P.
van Dam, submitted for publication). The Schneider Drosophila cells expressing
the salp15 gene from I. scapularis or I. ricinus were selected with hygromycin (500
�g ml�1) or blastomycin (25 �g ml�1), respectively, and were grown in large
spinner flasks together with penicillin and streptomycin (Invitrogen) for 3 days.
* Corresponding author. Mailing address: Leiden University Medi-
cal Centre, Department of Medical Microbiology, University of Lei-
den, Albinusdreef 2, 2333 ZA Leiden, The Netherlands. Phone: 31-
71-5261451. Fax: 31-71-5248148. E-mail: T.J.Schuijt@lumc.nl.
� Published ahead of print on 21 April 2008.
2888
o
n
D
ecem
ber 5, 2011 by guest
http://iai.asm
.org/
D
ow
nloaded from
The Drosophila cells were subsequently induced with copper sulfate with a final
concentration of 500 mM for 4 days and centrifuged at 1,000 � g for 15 min. The
supernatant was filtered using a 0.22-�m filter (Millipore). Both Salp15 Iric-1
and Iscap Salp15 were purified from the supernatant by use of the HisTrap Ni2�
column (GE Healthcare) and eluted with 100 mM imidazole. The eluted frac-
tions were filtered through a 0.22-�m filter and concentrated with a 5-kDa
concentrator (Vivascience) through centrifugal concentration.
The purity of the purified Salp15 was checked by silver staining (Bio-Rad)
according to the manufacturer’s recommendations, and the concentration was
determined with the Bradford assay.
NHS. Serum samples used were derived from one donor and were checked for
the absence of antibodies against B. burgdorferi by Western blot analysis. Heat
inactivation of NHS was achieved by incubation of the serum samples at 56°C for
30 min.
Assays for detection of complement-mediated killing of spirochetes and
Salp15 protection. Two serum-sensitive B. garinii strains, the A87S and the VSBP
strains, and two intermediately resistant B. burgdorferi sensu stricto strains,
VS215 and B31, were used (107 spirochetes ml�1). Spirochetes (2.5 � 105) were
preincubated with bovine serum albumin (BSA), Salp15 Iric-1, or Iscap Salp15
(80 �g/ml) for 30 min at 33°C. They were then incubated with NHS or heat-
inactivated NHS and examined after 1.5 h, 4.5 h, and 24 h. The two parameters
of borreliacidal effect that were recorded are immobilization and bleb formation
of the spirochetes. Immotile spirochetes were considered dead (28). The per-
centages of immotile spirochetes for 200 spirochetes per well were assessed. In
a separate titration experiment, different Salp15 concentrations, ranging from 5
�g/ml to 160 �g/ml, were also tested in the same way.
To find out if membrane-bound Salp15 protects the spirochetes against anti-
body-independent complement-mediated killing, the spirochetes were washed
twice with BSK-H medium (4,000 � g, 10 min) after incubation with Salp15
Iric-1. After removal of unbound Salp15 by washing, the spirochetes were sub-
jected to 12.5% NHS and examined for borreliacidal effect after 1.5 h, 4.5 h, and
24 h of incubation.
Subculture of B. garinii VSBP after incubation with I. ricinus Salp15. The
serum-sensitive B. garinii VSBP strain was preincubated with Salp15 Iric-1 for 60
min at 33°C. Then, spirochetes were exposed to 50% NHS for 24 h. As described
above, the surviving spirochetes were subcultured in BSK-H medium for 7 days.
This selection process was repeated twice.
Binding of I. ricinus Salp15 to B. garinii VSBP in overlay assay. B. garinii VSBP
lysates were obtained from 100-ml cultures that were grown to a density of 1 �
108 spirochetes/ml. Lysates were separated by 12.5% sodium dodecyl sulfate-
polyacrylamide gel electrophoresis and blotted onto an Immobilon-P membrane
(Millipore) to have approximately similar amounts of OspC, as determined by
Coomassie staining. After the membrane was incubated overnight in blocking
buffer (1% BSA, 3% milk in TBS-0.05% Tween), it was incubated with 1 �g/ml
purified Salp15 Iric-1 in blocking buffer for 1.5 h at room temperature, washed,
and consequently incubated with a 1:5,000 dilution of horseradish peroxidase-
conjugated V5 antibody (Invitrogen) in blocking buffer. Blots were developed by
enhanced chemiluminescence.
Binding assay of Salp15 Iric-1 to B. garinii VSBP by immunofluorescence.
Purified Salp15 Iric-1 was biotinylated by incubating 1 mg/ml Salp15 with 0.25
mg/ml sulfo-NHS-biotin (Pierce) for 45 min at 4°C. Unbound sulfo-NHS-biotin
was removed by dialysis against 25 mM lysine in phosphate-buffered saline (PBS)
overnight and four times against PBS for 15 min at 4°C.
Biotinylated purified Salp15 Iric-1 was incubated with B. garinii VSBP spiro-
chetes for 30 min at 33°C. The spirochetes were washed twice with PBS-1% BSA
and were resuspended in PBS-1% BSA, air dried on microscope slides overnight,
FIG. 1. Spirochetes are protected against complement-mediated killing in the presence of Salp15 Iric-1 or Iscap Salp15. Serum-sensitive B.
garinii strains VSBP (A) and A87S (B) were incubated with 12.5% NHS, while intermediately resistant B. burgdorferi strains VS215 (C) and B31
(D) were incubated with 50% NHS, after they had been preincubated for 30 min with BSA, Iscap Salp15, or Salp15 Iric-1. As a control, spirochetes
were also incubated with heat-inactivated NHS. After 1.5 h, 4.5 h, and 24 h of incubation with serum, the percentages of immotile spirochetes were
determined. Two hundred spirochetes were counted.
VOL. 76, 2008 Salp15 INHIBITS COMPLEMENT-MEDIATED BORRELIA KILLING 2889
o
n
D
ecem
ber 5, 2011 by guest
http://iai.asm
.org/
D
ow
nloaded from
and fixed in 100% methanol. Slides were incubated with bisbenzimide and
streptavidine-Cy3 (Sigma) and examined with a fluorescence microscope
(Axioscop 2 mot plus; Carl Zeiss).
Detection of terminal C5b-9 complement complexes. B. garinii strain VSBP
was preincubated with either BSA, Iscap Salp15, or Salp15 Iric-1 for 30 min at
33°C. Then, the spirochetes were incubated in BSK-H medium containing 12.5%
NHS for 30 min. The spirochetes were washed twice with PBS-1% BSA. They
were resuspended in PBS-1% BSA, air dried on microscope slides overnight, and
fixed in 100% methanol. Spirochetes were detected by incubation with human
serum containing antibodies against B. burgdorferi, and the C5b to C9 (C5b-9;
one molecule each of C5b, C6, C7, and C8 and one or more molecules of C9)
complement complexes were indicated with monoclonal mouse C5b-9 anti-
bodies (Dako). Slides were washed with PBS-1% BSA and incubated with an
anti-human immunoglobulin G-fluorescein isothiocyanate-labeled antibody
(BioMerieux) and an anti-mouse Cy3 antibody (Jackson). After slides were
washed and were mounted with Mowiol, they were visualized by confocal mi-
croscopy using a fluorescence microscope (Axioscop 2 mot plus; Carl Zeiss). At
least 100 spirochetes were counted, and the experiment was performed two
times.
Statistical analysis. The protection of spirochetes against complement-medi-
ated killing by Salp15 Iric-1 or Iscap Salp15 was compared to the protection by
the control protein BSA. The chi-square test was used for the analysis of pro-
portions, where absolute numbers of spirochetes were used in cross tabulations.
Crude relative risks for surviving different circumstances were estimated as odds
ratios (OR) and presented with both 95% confidence intervals (95% CI) and P
values. Tests were performed using SPSS 14 software. Calculated P values of
�0.05 were considered significant.
RESULTS
Salp15 protects serum-sensitive B. burgdorferi sensu lato
isolates against complement-mediated killing. As expected, B.
burgdorferi sensu lato isolates differed in their sensitivities to
NHS. No motile B. garinii A87S or B. garinii VSBP organisms
were seen when incubated for 24 h in 12.5% NHS (Fig. 1). In
contrast, other spirochetes required greater amounts of NHS
to induce partial killing: B. burgdorferi B31 was killed at 24 h by
50% NHS (Fig. 1) and B. burgdorferi VS215 was killed by 75%
NHS at 24 h (data not shown). All these Borrelia isolates
TABLE 1. Protective effect of Salp15 Iric-1 or Iscap Salp15 compared to that of BSA
Isolate Preincubationwitha:
Protection against immobilization afterb:
1.5 h 4.5 h 24 h
% Deadc ORd 95% CI P value % Dead OR 95% CI P value % Dead OR 95% CI P value
B. garinii VSBP BSA 57/43 82/62 93/88
Iric-1 13 8.9 5.4–14.6 �0.0001 19 19.4 11.7–32.2 �0.0001 48 14.7 8.0–27.0 �0.0001
9 7.6 6.0–9.8 �0.0001 22 5.9 4.9–7.2 �0.0001 46 8.2 6.6–10.3 �0.0001
Iscap 24 4.2 2.7–6.4 �0.0001 59 3.1 2.0–4.9 �0.0001 88 1.9 0.96–3.8 N.S.
25 2.3 1.9–2.8 �0.0001 49 1.7 1.4–2.0 �0.0001 83 1.5 1.2–1.9 0.001
B. garinii A87S BSA 70/71 87/91 97/94
Iric-1 34 4.6 3.0–7.1 �0.0001 47 7.6 4.6–12.4 �0.0001 44 35.1 15.7–78.4 �0.0001
33 5.0 4.2–6.1 �0.0001 64 5.5 4.3–7.0 �0.0001 60 10.9 8.1–14.5 �0.0001
Iscap 54 2.0 1.4–3.1 0.001 84 1.3 0.73–2.2 N.S. 90 3.1 1.3–7.4 0.01
51 2.4 2.0–2.9 �0.0001 89 1.2 .93–1.7 N.S. 89 2.1 1.5–2.9 �0.0001
B. burgdorferi VS215 BSA 75/75 77/80 79/83
Iric-1 44 3.8 2.5–5.8 �0.0001 56 2.6 1.7–3.9 �0.0001 55 3.1 2.0–4.7 �0.0001
53 2.7 2.2–3.3 �0.0001 59 2.7 2.2–3.3 �0.0001 61 3.2 2.6–4.0 �0.0001
Iscap 59 2.1 1.4–3.3 �0.0001 65 1.8 1.1–2.7 0.01 68 1.8 1.1–2.8 0.01
58 2.3 1.9–2.7 �0.0001 66 2.0 1.6–2.4 �0.0001 61 3.1 2.5–3.9 �0.0001
B. burgdorferi B31 BSA 81/86 82/90 98/89
Iric-1 65 2.2 1.4–3.5 �0.0001 75 3.0 1.6–5.4 �0.0001 93 2.9 1.0–8.3 0.03
74 2.2 1.7–2.7 �0.0001 79 2.3 1.8–3.0 �0.0001 77 2.6 2.0–3.3 �0.0001
Iscap 71 1.7 1.1–2.7 0.03 71 1.5 0.76–2.8 N.S. 93 2.9 1.0–8.3 0.03
79 1.6 1.3–2.1 �0.0001 83 1.8 1.4–2.4 �0.0001 78 2.4 1.9–3.1 �0.0001
a Iric-1 and Iscap indicate Salp15 Iric-1 and Iscap Salp15, respectively. Spirochetes were preincubated with Salp15 Iric-1 or Iscap Salp15 for 30 min before incubation
with NHS.
b Spirochetes were incubated for 1.5, 4.5, or 24 h with 12.5% NHS in the case of B. garinii VSBP and B. garinii A87S or 50% NHS in the case of B. burgdorferi B31
and VS215.
c Percentages of dead spirochetes. One hundred spirochetes were counted and the experiment was performed two times. Lightface values show the percentages of
dead spirochetes treated with Salp15 Iric-1, while boldface values indicate the percentages of dead spirochetes treated with Iscap Salp15.
d The protective effect of Salp15 compared to BSA was calculated using the chi-square test. Lightface values represent experiments with 200 spirochetes, while
boldface values indicate experiments with 1,000 spirochetes.
TABLE 2. Protective effect of Salp15 Iric-1 compared to that of Iscap Salp15
Isolate
Protection against immobilization aftera:
1.5 h 4.5 h 24 h
% Deadb ORc 95% CI P value % Dead OR 95% CI P value % Dead OR 95% CI P value
B. garinii VSBP 9/25 3.3 2.6–4.3 �0.0001 22/49 3.5 2.9–4.2 �0.0001 46/83 5.6 4.6–6.9 �0.0001
B. garinii A87S 33/51 2.1 1.8–2.5 �0.0001 64/89 4.4 3.5–5.6 �0.0001 60/89 5.2 4.2–6.6 �0.0001
a Spirochetes were incubated for 1.5, 4.5, or 24 h with 12.5% NHS.
b Percentages of dead spirochetes. One hundred spirochetes were counted and the experiment was performed two times. Lightface values show the percentages of
dead spirochetes treated with Salp15 Iric-1, while boldface values indicate the percentages of dead spirochetes treated with Iscap Salp15.
c The protective effect of Salp15 Iric-1 versus Iscap Salp15 against complement-mediated killing was calculated using the chi-square test.
2890 SCHUIJT ET AL. INFECT. IMMUN.
o
n
D
ecem
ber 5, 2011 by guest
http://iai.asm
.org/
D
ow
nloaded from
survived incubation for 1.5, 4.5, or 24 h with heat-inactivated
NHS, demonstrating the importance of complement in serum
sensitivity.
Since Salp15 has previously been shown to enhance the
capacity of spirochetes to survive in naive mice (25), we deter-
mined whether Salp15 could alter the serum sensitivity of these
B. burgdorferi isolates. Indeed, Salp15 from I. scapularis ticks,
Iscap Salp15, altered the serum sensitivity of Borrelia (Fig. 1).
The percentage of dead spirochetes significantly decreased
when Borrelia isolates were preincubated with Iscap Salp15
(Table 1). The strongest protective effect was seen with B.
garinii VSBP. We had previously cloned Salp15 from I. scapu-
laris ticks, but since the serum-sensitive Borrelia are transmit-
ted predominantly by I. ricinus ticks, we wanted to determine
whether Salp15 from these ticks was as potent as or more
potent than Salp15 from I. scapularis for selected Borrelia iso-
lates. We therefore cloned, expressed, and purified Salp15
from I. ricinus, Salp15 Iric-1, as described in Materials and
Methods. Salp15 Iric-1 afforded protection against comple-
ment-mediated killing (Fig. 1; Table 1). The protective effect
of Salp15 Iric-1 for all four Borrelia isolates was also greater
than that of Iscap Salp15 (Table 1) and the difference in pro-
tection of serum-sensitive B. garinii spirochetes by Salp15 Iric-1
compared to that by Iscap Salp15 was most apparent (Table 2).
A titration of Salp15 showed that both Iscap Salp15 and Salp15
Iric-1 had a dose-dependent protective effect (data not shown).
Salp15 Iric-1 binds to the surface of B. garinii VSBP. B.
garinii VSBP was preincubated with biotinylated Salp15 Iric-1
and membrane-bound Salp15 Iric-1 was detected using strepta-
vidin-Cy3 in