Published online before print February 4, 2005
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
,2
* Center for Immunology and Inflammatory Diseases, Massachusetts General Hospital, Charlestown;
Molecular Oncology Research Institute and
Division of Geographic Medicine and Infectious Diseases, Tufts-New England Medical Center, Boston, Massachusetts
3 Correspondence: Center for Immunology and Inflammatory Diseases, Massachusetts General Hospital, 149 13th Street, Room 8301, Charlestown, MA, 02129. E-mail: lglickstein{at}partners.org
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
from wild-type and spp1–/– macrophages alike. These results indicate that the usually critical role of osteopontin in the induction of cellular immune responses to intracellular pathogens was circumvented by the ability of the extracellular pathogen B. burgdorferi to induce macrophages directly to produce proinflammatory cytokines.
Key Words: mouse • macrophage • IL-12 • TNF-
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
(IFN-) or nitric oxide, and expression was associated with survival in human patients [3
, 4
]. Osteopontin may also contribute prominently to autoimmunity [1
, 5
, 6
]. A role for osteopontin in defense against extracellular pathogens has not yet been defined. In mice, disease manifestations following infection with the extracellular pathogen Borrelia burgdorferi vary widely depending on the host immune response. Experimental Lyme arthritis, which closely recapitulates features of acute Lyme arthritis in humans, is induced in susceptible mouse strains by infection with B. burgdorferi. Current evidence suggests that strong proinflammatory responses, coupled with bactericidal humoral responses, control the number of spirochetes in tissues. Susceptible C3H/He mice do not make the acute proinflammatory immune response associated with less severe disease manifestations in other mouse strains [7 8 9 ] despite their ability to respond to lipoproteins through Toll-like receptor 2 [10 11 12 ]. These mice were previously characterized as bearing a low expression allele of osteopontin [13 ], and their infection with B. burgdorferi consistently induces an early TH2 cytokine response [8 ]. Although these results suggested a possible role for osteopontin in host resistance to experimental Lyme arthritis, a previous study concluded that a closely linked gene on mouse chromosome 5, rather than spp1 itself, might be responsible for the host defense against B. burgdorferi in arthritis-resistant 129S7 mice [14 ].
A downside of a strong proinflammatory response against B. burgdorferi associated with type 1 cytokines is organ involvement, such as arthritis and carditis [15 , 16 ]. Although C3H mice may lack robust, early TH1 responses, a strong, late TH1 response in these mice is associated with arthritis [17 , 18 ]. A similar mechanism of delayed TH1 response might in part explain the course of the group of human patients who develops antibiotic treatment-resistant Lyme arthritis [19 ]. The inflammation that is seen with B. burgdorferi infection may not be dependent on osteopontin, as some interbred mice (B6129F1 and B6129F2) bearing genes from the C57BL/6 and 129S7 parental strains were susceptible to arthritis in the presence and absence of the spp1 gene [14 ]. These spp1–/– mice produced an immunoglobulin G2 (IgG2) humoral response, characteristic of a TH1 response, suggesting that the immune response had bypassed osteopontin.
To further characterize the possible relationship between osteopontin and inflammation in the immune response to B. burgdorferi, we studied cytokine production using ex vivo macrophages and assessed arthritis and bacterial burden in vivo following B. burgdorferi infection in arthritis-susceptible mice in the presence and absence of osteopontin.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Spirochetes and antigen preparation
Dr. Stephen Barthold (University of California, Davis) kindly provided the N40 isolate of B. burgdorferi. Derivation of clone D10/E9 was described previously [21
], and a virulent isolate was obtained by infecting mice with this clone and isolating spirochetes from tissues by outgrowth in Barbour-Stoenner-Kelly (BSK) II medium (Sigma Chemical Co., St. Louis, MO). Soluble antigens were prepared from a 500-ml culture of spirochetes by sonication, as described previously [22
]. The supernatant of soluble antigens (lysate) was stored in aliquots at –70°C.
Lysate stimulation of lymph node cells
Single-cell suspensions from pooled lymph nodes of uninfected C57BL/6J or C3H/HeJ mice were washed with RPMI-1640, supplemented with 20 mM HEPES, 1 mM sodium pyruvate, a 1x concentration of nonessential amino acids, and 2 mM l-glutamine (serum-free, complete RPMI), and were plated at 5 x 105 cells/well in 96-well plates in the presence of 10 µg/ml soluble antigen in serum-free, complete RPMI. Supernatant was harvested from cultures after 17–40 h and frozen at –70°C.
Mouse infection
Low passage isolates of B. burgdorferi strain N40 were frozen into single experiment aliquots in BSK II containing 30% glycerol. To infect the mice, frozen B. burgdorferi were thawed rapidly, transferred into 10 vol BSK II, and cultured overnight at 32°C prior to enumeration of motile spirochetes using dark field microscopy. Mice were inoculated in the right hind footpad with 5 x 104 motile spirochetes in 50 µl BSK II or were mock-infected with BSK II alone. All animal experiments were performed according to National Institutes of Health (NIH) guidelines under an Institutional Animal Care and Use Committee-approved protocol.
Lyophilized, purified recombinant mouse osteopontin was obtained from R&D Systems (Minneapolis, MN), reconstituted with sterile phosphate-buffered saline (PBS), and stored at –70°C. B6129S7tm1spp1F3 and C3H/HeJ mice were injected subcutaneously (s.c.) on days 0, 2, 4, 6, 8, and 10 postinfection with 10 µg osteopontin.
Joint swelling and inflammation
Ankle swelling and histopathologic changes are under control of different genes in mice [23
]. Tibiotarsal joint swelling was measured on a weekly basis with a spring-loaded microcaliper (Federal, Providence, RI). After sacrifice at 7 weeks postinfection, the right tibiotarsal joint was fixed in 10% formalin (Fisher Scientific, Pittsburgh, PA). Fixed tissue was decalcified, sectioned, and stained with hematoxylin and eosin by Suzanne White (Beth Israel Medical Center, Boston, MA). K. Craig-Mylius and L. Glickstein independently scored each ankle from zero (no arthritis) to four (severe arthritis) and averaged scores as described previously [22
].
Quantitative PCR (Q-PCR)
B. burgdorferi- and mock-infected mice were killed at 5 weeks postinfection, and the joint tissue was stored at –80°C until DNA extraction was performed as described [24
]. Q-PCR was performed in a spectrofluorometric thermal cycler (MX4000, Stratagene, La Jolla, CA) The final reaction concentration was 1x Brilliant SYBR Green master mix (Stratagene), 3 mM MgCl2, 0.25 µM each primer, and 200 ng target DNA. The primers used to detect B. burgdorferi recA were nTM17F (5'-GTG GAT CTA TTG TAT TAG ATG AGG CTC TCG-3') and nTM17R (5'-GCC AAA GTT CTG CAA CAT TAA CAC CTA AAG-3') [24
]. The primers used to detect mouse nidogen were nido.F (5'-CCA GCC ACA GAA TAC CAT CC-3') and nido.R (5'-GGA CAT ACT CTG CTG CCA TC-3') [24
]. The thermal profile conditions used were 50°C for 2 min, 95°C for 10 min, and then 40 cycles of 95°C for 15 s and 60°C for 1 min. Emitted fluorescence for each reaction was measured three times during the annealing/extension phase, and amplification plots were analyzed using the MX4000 software version 3.0 (Stratagene). A series of standards was prepared using DNA prepared from B. burgdorferi using the QIAmp DNA mini kit (Qiagen, Valencia, CA). The DNA concentration of the standard sample was used to calculate B. burgdorferi genome equivalents (1 µg/ml DNA=6.1x105 genomes/µl), which were then diluted into DNA from an uninfected mouse to make a standard curve. As it has been shown that 200 ng mouse DNA contains 
104 nidogen copies, the number of recA copies amplified was normalized to this as described [24
].
Spirochete infection of macrophages
Primary cultures of bone marrow-derived macrophages (BMM
) were prepared as described previously [25
]. Briefly, the BM from four femurs (two mice) per strain was pooled for each experiment. A single-cell suspension was plated at 4 x 106 cells per 100 mm untreated plastic dish in 10 ml RPMI-1640 medium, supplemented with 10% fetal bovine serum (FBS), glutamine to 2 mM, penicillin G to 100 U/ml, and streptomycin sulfate to 100 ug/ml (components from Gibco/Invitrogen, Gaithersburg, MD), and 30% L cell-conditioned medium (BMM medium) and was incubated at 37°C, 5% CO2. Plates were fed on day 3 by adding 10 ml fresh BMM medium. On day 5, the medium and nonadherent cells were discarded, and adherent cells were lifted with 2 ml PBS with 5 mM EDTA per plate. The lifted cells were then plated at 5 x 105 cells/well (six-well tissue-culture plates) in RPMI 1640 plus 10% heat-inactivated FBS, glutamine to 2 mM, penicillin G to 100 U/ml, and streptomycin sulfate to 100 ug/ml L cell and macrophage medium (L/M medium). Plates were incubated at 37°C, 5% CO2, for 2 days before infection on day 7.
B. burgdorferi strain N40 clone D10/E9 was grown to 5 x 107–1 x 108 per ml in modified Kelly-Pettenkofer medium supplemented with human serum in place of rabbit serum at 33°C in tightly capped tubes as described previously [21 , 26 ]. The cultures were then supplemented with sterile glycerol to 20% and frozen at passage 11 at –80°C until use.
Frozen spirochetes were thawed, washed twice at ambient temperature in PBS, supplemented with 0.2% bovine serum albumin (BSA), and resuspended at 1.67 x 107 spirochetes/ml [to achieve a multiplicity of infection (MOI) of 100 bacteria per macrophage] in L/M medium without antibiotics. Ten-fold serial dilutions of this suspension into the same medium achieved MOIs of 10 and 1. L/M medium alone (without antibiotics) served as the control and is designated as MOI = 0. Macrophages were infected in serum-free L/M medium for osteopontin Western blot analysis. Spirochetes were added to prewashed, plated macrophages (3 ml/well each six-well plate), and plates centrifuged (540 g) for 15 min at room temperature to maximize bacteria-macrophage contact. Plates were returned to the incubator for 3 h infection (Western blot) or 4 h infection (flow cytometry). Cell-culture supernatants were removed for Western blot, centrifuged (210 g) to remove cells and debris, and frozen at –80°C until analysis.
Western blot
Western blotting was performed after separation of 40 µl tissue-culture supernatant on nonreducing, denaturing sodium dodecyl sulfate-polyacrylamide mini-gels and transfer to nitrocellulose membranes. The membranes were blocked with 5% BSA and 0.05% Tween-20 in PBS and probed with biotinylated antiosteopontin antibody (R&D Systems), followed by streptavidin-peroxidase and luminescent detection. Purified recombinant mouse osteopontin (10 ng, R&D Systems) served as a positive control.
Flow cytometry
Monensin (3 µM, Sigma Chemical Co.) was added to six-well plates after 3 h incubation, and plates were returned to incubator for 1 h. Washed, adherent cells were harvested by digestion with 0.5 ml trypsin/EDTA (Gibco/Invitrogen) for 5 min at 37°C, scraping from plates and pooling duplicate wells. Cells (105/well) were aliquoted into a 96-well V-bottom plate for staining with the intracytoplasmic cytokine kit from BD PharMingen (San Diego, CA). Briefly, cells were fixed in Cytofix/Cytoperm for 15 min at 4°C, washed once with Perm/Wash, and then incubated with anticytokine antibody [1:100, anti-IL-12 phycoerythrin (PE) or anti-tumor necrosis factor (anti-TNF-) PE, from BD Pharmingen] for 15 min at 4°C. Cells were washed twice with Perm/Wash and resuspended in PBS (2% FBS, 0.1% sodium azide, Sigma Chemical Co.) for analysis on a BD Pharmingen FACSCalibur cytometer.
Ig enzyme-linked immunosorbent assay (ELISA)
B. burgdorferi antigen-specific ELISA was performed as described previously [22
]. Briefly, diluted sera were incubated in plates previously coated with B. burgdorferi lysate and blocked with milk buffer. Isotype-specific secondary antibodies were added after washing, and plates were developed with appropriate reagents. Microplates were read at 405 nm. Titer was calculated by extrapolation from the linear regression of the absorbance from three dilutions of each serum sample. The titer was defined as the point where this line crossed that of the mean plus 3 SD of a pool of negative control sera.
Statistics
Means of groups were compared using Student’s t-test. We tested the datasets for all groups of mice at all time-points for normal distribution according to the test by Baker [27
] as modified by David et al. [28
]. P values that were less than or equal to 0.05 were considered significant.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
 View larger version (46K): [in a new window] |
Figure 1. Genetic deficiency of osteopontin does not affect the susceptibility to B. burgdorferi-induced arthritis. (A) Time-course of ankle swelling following infection by B. burgdorferi. Arthritis index was calculated as the sum of both tibiotarsal diameters. Mean ± SD of six to 22 mice per group is shown for each time-point. Values that are significantly greater than mock-infected mice are indicated (*). (B) Ankle swelling 35 days after infection, the time of peak arthritis; each point represents an individual animal. A black bar in each column indicates mean value. (C) Histopathology scores of ankle inflammation at sacrifice (49 days) after infection by B. burgdorferi. Mean value is indicated. (D) Histologic sections of ankles of infected mice and mock-infected controls. spp1 genotype of mice indicated as homozygous wild-type (+/+), deficient (–/–), or heterozygous (+/–). B, Bone; C, cartilage; JS, joint space; M, muscle; S, synovium; T, tendon. Mock, injected with medium; Bb, infected with B. burgdorferi.
|
Exogenous osteopontin does not ameliorate experimental Lyme arthritis
It was still possible that a role for osteopontin was obscured by other susceptibility alleles expressed in these mice. To ascertain whether the cytokine plays a protective role against Lyme arthritis, we infected arthritis-susceptible spp1–/– mice and C3H/He mice with B. burgdorferi (strain N40) and injected 10 µg recombinant-purified osteopontin in PBS or PBS alone, s.c. above the right hind limb on days 0, 2, 4, 6, 8, and 10. This dose of osteopontin has previously been shown to reconstitute TH1 responses in spp1–/– mice [1
]. As expected, the C3H/He mice developed more severe arthritis than the spp1–/– mice. In neither case was a protective effect achieved by exogenous osteopontin (Fig. 2
). In fact, the only significant difference between the groups was a more rapid onset of arthritis (day 14) in the osteopontin-treated C3H/He mice.
 View larger version (13K): [in a new window] |
Figure 2. Exogenous osteopontin does not reduce ankle swelling in C3H/He or spp1–/– mice. All mice were infected with B. burgdorferi in this experiment. Indicated mice [osteopontin (OPN)] received injections of recombinant osteopontin every other day from days 0 to 10. (A) Time-course of ankle swelling following infection by B. burgdorferi; each point represents the mean swelling of four to five animals per group. Values for treated groups that differ significantly from syngeneic infected mice are shown (*). (B) Ankle swelling on day 35 after challenge; each point represents an individual animal. The black bars indicate mean value for each group.
|
 View larger version (12K): [in a new window] |
Figure 3. Osteopontin deficiency does not increase spirochete burden following infection. Spirochete burden was determined by quantification of B. burgdorferi recA DNA copies in joints by Q-PCR. DNA was prepared from joint tissue and amplified with recA and murine nidogen-specific primers. Copy number was determined for each by comparison to a standard curve, and values expressed as copies recA/104 copies nidogen. Results (•) are presented for individual spp1–/– mice and wild-type (WT) littermates; black bars indicate mean value and SD.
|
Consistently, no detectable osteopontin was found in the supernatant from the C3H/He lymph node cells or macrophages following exposure to B. burgdorferi or its antigens (data not shown). Arthritis-susceptible C3H/He mice recognize B. burgdorferi antigens [17 ], but they have the osteopontin allele (spp1b/b), which is associated with low expression levels [13 ].
Osteopontin was constitutively secreted in macrophages from the C57BL/6 strain and induced in the 129S7 strain following in vitro infection with live B. burgdorferi (Fig. 4 ). Macrophages from the C57BL/6 mice expressed a form of osteopontin that had slower electrophoretic mobility, most likely a result of differential glycosylation [29 , 30 ]. Recombinant osteopontin served as a positive control for antibody specificity.
 View larger version (19K): [in a new window] |
Figure 4. Osteopontin secretion in response to B. burgdorferi in C57BL/6 and 129S7 mice. (A) Equal numbers of macrophages from 129S7 or C57BL/6 mice were infected with 0 or 10 B. burgdorferi/cell for 4 h in serum-free media, and the supernatants from the macrophage cultures were collected for Western blot analysis. Purified recombinant mouse osteopontin (rOPN; 10 ng) served as a positive control. (B) Equal numbers of primary lymphocytes from lymph nodes of uninfected C57BL/6 mice were treated with B. burgdorferi lysate for 17–40 h in serum-free media, and the supernatants from the culture wells were collected for Western blot analysis.
|
Immune responses to B. burgdorferi do not depend on osteopontin
Class switching in the mouse, in particular, production of IgG2a, is dependent in part on IFN- [31
]. We examined B. burgdorferi antigen-specific Ig isotypes in the sera of infected mice. Although pathogen-specific IgG2a and IgG2b are considered to be indicative of TH1-driven immunity, high titers of IgG1 reflect TH2 responses. We found all isotypes to be induced after challenge with no differences between spp1–/– and wild-type mice (Fig. 5a
).
 View larger version (28K): [in a new window] |
Figure 5. B. burgdorferi induces IgG2 isotypes and proinflammatory cytokines in macrophages independently of osteopontin. (a) B. burgdorferi-specific Ig isotype ELISA. Sera from B. burgdorferi-infected (Bb) and mock-infected mice were subjected to Ig isotype-specific ELISA against plate-bound B. burgdorferi lysate, and titer was calculated from serially diluted samples. Results for individual animals ( ) and mean + SD (•) are presented for spp1–/– mice and wild-type (WT) littermates. n.d., None detected. (b) Intracellular cytokine staining. Macrophages from uninfected spp1–/– mice and WT littermates were infected ex vivo and subjected after 4 h to flow cytometry analysis for intracellular IL-12 and TNF-. Histograms are shown for MOI = 0 and MOI = 1, 10, and 100. Neg FITC, Negative fluorescein isothiocyanate.
|
in vitro with live B. burgdorferi and measured the production of the TH1 cytokine IL-12 and the arthritogenic cytokine TNF- by intracellular staining and flow cytometry. Both cytokines were induced dose-dependently in macrophages from wild-type B6129F3 or spp1–/– mice (Fig. 5b)
. The macrophages from the spp1–/– mice produced higher levels of IL-12 and TNF- in this experiment. The possibility of a negative feedback mechanism involving osteopontin might warrant further study. |
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
by macrophages, independently of osteopontin, a cellular response that was sufficient to generate humoral immunity and arthritis. Although neither other groups nor ours identified any defect in the response to B. burgdorferi infection in vivo by spp1–/– mice, exposure to B. burgdorferi antigens led to substantial osteopontin secretion from lymphocytes of C57BL/6 and 129S7 mice that bear the spp1a/a allele. It is possible that osteopontin does affect symptoms other than arthritis. Our investigations have focused on the multiplicity of organisms in situ and on the development of acute arthritis as readouts for host responses to B. burgdorferi infection. We did not examine the murine carditis or cardiac conduction abnormalities associated with B. burgdorferi infection [33 , 34 ]. Other signs of infection in humans, such as skin affliction (erythema migrans, acrodermatitis chronica atrophicans), eye, and nervous system involvement (meningoradiculoneuritis, dementia) or treatment-resistant Lyme arthritis, a syndrome of putative infection-induced autoimmunity, do not occur in mice, and thus, the potential effects of osteopontin on these disease manifestations could not be tested using the mouse model. In addition, the relatively low number of B. burgdorferi cells per mouse cell during in vivo infection might be below a critical threshold required to induce osteopontin in quantitities sufficient to alter the immune response.
Based on our results, however, we would argue that a role for osteopontin is unlikely, as a classic TH1 response with cytotoxic T lymphocytes is not the critical arm of the immune response to extracellular pathogens. In the conventional TH1 priming pathway, activated T cells secrete osteopontin, which induces macrophages to secrete IL-12 and inhibits IL-10 production, both of which drive TH1 differentiation. For intracellular pathogens, generation of a cytotoxic T cell response is critical, as the immune system faces a challenge in that the pathogens are hidden within host cells. Thus, a mechanism involving T cell recognition and osteopontin-mediated priming of macrophages is needed to induce such a response. In contrast, B. burgdorferi expresses a number of lipoproteins that are highly immunostimulatory and proinflammatory [35 36 37 ]. In this setting, in which macrophages are activated directly, there is no need for osteopontin, and cytotoxic T cells are not required to clear infection. Rather, antibody responses are critical.
Circumvention of the conventional TH1 priming pathway by direct induction of IL-12 from macrophages by lipoproteins or other spirochetal components may have implications for the pathology seen in murine and human infection. A lack of osteopontin production, as observed in susceptible C3H/He mice, might be expected to result in overproduction of IL-10. Excess IL-10 has been shown to impede the host antispirochetal immune response [38
] and may contribute to chronic infection. Furthermore, if the infection is not cleared by a successful immune response or antibiotic therapy, chronic T cell-independent IL-12 and TNF- production by macrophages and possibly other cells, induced by B. burgdorferi components, would lead to joint pathology and might prime susceptible individuals to develop autoimmunity. It is most important that dependence versus independence from osteopontin may define two variant innate-immune responses to intracellular versus extracellular bacterial infection. This may have important implications for disease manifestations, resolution, and therapeutic strategies.
2 Current address: University of Cincinnati Medical Center, College of Pharmacy, 3223 Eden Avenue, Cincinnati, OH 45267-0004.
Received June 23, 2004; revised January 10, 2005; accepted January 14, 2005.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
-1 (Osteopontin): an early component of type-1 (cell-mediated) immunity Science 287,860-864-1) gene. Definition of a novel T cell-dependent response associated with genetic resistance to bacterial infection J. Exp. Med. 170,145-161-1 cytokine and the development of systemic autoimmune disease J. Immunol. 147,2902-2906[Abstract] in modulation of immunity to Borrelia burgdorferi in mice J. Immunol. 155,2020-2028[Abstract] IIb ß 3 mediates binding of the Lyme disease agent Borrelia burgdorferi to human platelets Proc. Natl. Acad. Sci. USA 90,7059-7063 Interferon is not required for arthritis resistance in the murine Lyme disease model Infect. Immun. 69,3737-3743ß T-cell receptor transgenic mice Infect. Immun. 65,4094-4099[Abstract]This article has been cited by other articles:
 |
|
 |
|
  L. J. GLICKSTEIN and J. L. COBURN ASSOCIATION OF MACROPHAGE INFLAMMATORY RESPONSE AND CELL DEATH AFTER IN VITRO BORRELIA BURGDORFERI INFECTION WITH ARTHRITIS RESISTANCE Am J Trop Med Hyg, November 1, 2006; 75(5): 964 - 967. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Sodek, A. P. Batista Da Silva, and R. Zohar Osteopontin and Mucosal Protection Journal of Dental Research, May 1, 2006; 85(5): 404 - 415. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
