|
|
||||||||
CUTTING EDGE |


* Servicio de Inmunología, Hospital Universitario Marqués de Valdecilla, Servicio Cántabro de Salud, Santander, Spain; and
Biochemisches Institut, Christian-Albrechts-Universität Kiel, Kiel, Germany
| Abstract |
|---|
|
|
|---|
| Introduction |
|---|
|
|
|---|
To analyze the role of Ctsd in LM clearance, we used gain and loss of function experiments in fibroblasts and macrophages (MAC) followed by studies on the intracellular growth of LM, phagosome isolation, and evaluation of the viability of intraphagosomal LM. Fibroblasts and MAC from Ctsd-deficient mice were highly susceptible to LM infection and exhibited an enhanced intraphagosomal viability of this bacterium. Ctsd/ mice contained 10-fold more bacteria compared with Ctsd+/+ mice.
| Materials and Methods |
|---|
|
|
|---|
Biotinylation reagents (Pierce), transfection antibiotics (Clontech), and Abs: rabbit anti-Ctsd (11), polyclonal rabbit anti-LLO specific (Diatheva) and HRP-conjugated Abs (Jackson ImmunoResearch Laboratories). Tissue culture media were DMEM (D) supplemented with 10 mM L-glutamine, 100 mM nonessential amino acids, 50 µg/ml gentamicin, 50 µg/ml vancomycin, and 5% (D5), 10% (D10), or 20% (D20) of FCS.
Bacteria and infection in mice
LM strains used were as follows: 10403S wild-type LM,
LLO (LLO-deficient mutant, DP-L2161);
plcA/plcB (plcA and plcB-deficient mutant, DP-L1936) and
LLO
plcA/plcB (LLO, plcA and plcB-deficient mutant, DP-L2319); Escherichia coli-producing recombinant N-terminal-tagged His-LLO bacteria (12); and E. coli-producing recombinant N-terminally His-tagged EEA1 (13). LM were FITC-labeled (0.5 mg/ml FITC-PBS, 1 h, 4° C), and CFU were enumerated. Mice and cells used correspond to the Ctsdtm1Cptr/Ctsdtm1Cptr allelic phenotype and were called for simplicity, Ctsd/ or Ctsd+/+ for Ctsd-deficient or wild-type mice, respectively.
Cells, constructs, and transfections
Mouse embryonic fibroblasts (11, 14) were cultured in D10. Bone marrow MAC (BM-MAC) were cultured in D20 with 25 ng/ml macrophage-CSF for 7 days. Vectors were as follows: pcDNA3.1/Ctsd-hygro, Ctsd/peGFP, and peGFP were transfected and selected as described previously (15).
Bacterial infection assays
The infection experiments were performed as previously reported (2). Intraphagosomal viability was evaluated in isolated phagosomes as CFU and expressed as described elsewhere (6, 15).
Immunoblots, in vivo and in vitro degradation assays
Immunoblots of isolated phagosomes (30 µg/lane) (10) were incubated with rabbit anti-Ctsd (1/1,000) or monoclonal anti-Rab5a (1/5,000) Abs. Biotinylation was performed with EZ-link-LC-biotin (Pierce) (1 mg/ml, 1 h, 37°C, and 4-h incubation in D10). Immunoprecipitations were performed with a rabbit anti-LLO Ab and developed with streptavidin-HRP (1/20,000). For Rab5a-GTP assays, phagosomes were incubated with 3 µg of His-EEA1 to recover exclusively Rab5a-GTP (13), and immunoblots were developed with anti-Rab5a Ab. For in vitro degradation assays of His-LLO, 5 µg of purified Ctsd was incubated with 5 µg of His-LLO in the presence or absence of 1 mM pepstatin A.
Isolation of phagosomal membrane fractions
Fibroblasts and CHO cells were infected with LM (ratio of 20:1 bacteria:cells, 20 min) in the presence or absence of pepstatin A (5 µg/ml) and processed as described previously (10, 15). Fibroblast phagosomes contained <2% of plasma membrane, 0.2% of Golgi, and 0.25% of endosomal contamination. For calculations of percentages of phagosomal and cytosolic bacteria, LM infection was conducted at 0 h. Postnuclear supernatants (PNS) were obtained, and CFU values reflected the total of internalized LM. Phagosomes were isolated, and CFU values were expressed as percentages of the total LM. The percentage of cytosolic bacteria was calculated as (PNS phagosomal numbers/PNS) x 100.
Fluorescence microscopy
Conventional immunofluorescence microscopy was performed with cells fixed in cold methanol. Confocal microscopy was used for visualization of BM-MAC infected with FITC-LM.
Statistical analysis
For statistical analysis, the Students t test was applied.
| Results |
|---|
|
|
|---|
We first analyzed the intracellular survival of LM in Ctsd+/+ fibroblasts and observed exponential growth especially between 8 and 18 h postinfection (Fig. 1A). Ctsd played a crucial role in LM infection because the intracellular growth indexes of LM in Ctsd/ fibroblasts were 4- to 6-fold higher than in wild-type Ctsd+/+ cells at 8 h postinfection (Fig. 1B). Moreover, transfection of Ctsd/ cells with a Ctsd expression vector (Ctsd) reduced LM intracellular growth to values observed in control fibroblasts. Transfection of wild-type Ctsd+/+ fibroblasts with Ctsd decreased LM intracellular growth indexes by 4-fold (CFU ratio 9.8 ± 0.3 compared with 2.2 ± 0.5 in transfected cells). The specificity of the role of Ctsd was confirmed using fibroblasts deficient in cathepsin-L (Ctsl/) (Fig. 1B).
|
Phagosomes are critical compartments in the control of LM growth (1, 15). The number of viable LM (CFU) was 4-fold higher within Ctsd/ phagosomes compared with control phagosomes (Fig. 1C). Phagosomes from Ctsd/ cells transfected with Ctsd contained numbers of viable LM similar to those found in wild-type phagosomes. Phagosomes from Ctsd+/+ fibroblasts transfected with Ctsd contained 8-fold fewer viable intraphagosomal bacteria compared with phagosomes from mock-transfected cells (Fig. 1C). Ctsd protein levels in wild-type Ctsd+/+ phagosomes, as well as in Ctsd/-deficient fibroblasts re-expressing Ctsd, showed low amounts of the active 34-kDa mature Ctsd form (Fig. 1D). Phagosomes with low levels of intraphagosomal viable LM in Ctsd+/+ fibroblasts transfected with Ctsd displayed 15-fold higher levels of active 34-kDa Ctsd. Interestingly, Ctsd/ phagosomes exhibited very low levels of Rab5a compared with wild-type phagosomes. Moreover, transfection of Ctsd+/+ fibroblasts with Ctsd resulted in a significant increase in Rab5a levels (Fig. 1D). These results suggest that intraphagosomal viability of LM inversely correlated with active Ctsd and high levels of Rab5a in phagosomes. Phagosomes from CHO cells (15) transfected with Ctsd displayed a 5-fold lower intraphagosomal CFU compared with control cells. The listericidal potential of phagosomes from Ctsd-transfected CHO was dependent on the presence of active Ctsd, because it was pepstatin-A sensitive (Fig. 1E). These phagosomes showed high total Rab5a and Rab5a-GTP levels (Fig. 1E). These results indicate that Rab5a activation was required for the listericidal ability of Ctsd.
The listericidal effect of Ctsd is targeted to listeriolysin O
We next analyzed whether the main virulence factors of LM with reported phagosomal activity, LLO or phospholipases C (PLCs) (1, 16), are targets for Ctsd.
plcA/plcB is a mutant deficient in plcA/plcB (PLCs), and
LLO
plcA/plcB is deficient in LLO and the two PLCs (17). A
plcA/plcB LM mutant showed a 4-fold higher LM intracellular growth index in Ctsd/ cells as compared with Ctsd+/+ cells (Fig. 1B). No enhancement of LM replication indexes in Ctsd/ cells was observed with
LLO
plcA/plcB-deficient mutants (Fig. 1B). This suggests that the high LM susceptibility of Ctsd/ cells was LLO-specific and renders the involvement of PLCs unlikely.
The calculated percentages of phagosomal vs cytosolic bacteria at 0 h reflected the compartment in which each LM bacterial mutant was able to survive (1, 18) (Table I). In Ctsd+/+ cells, 30% of total wild-type LM were localized within the phagosomes, and 70% of total viable LM were in the cytosol. In Ctsd/ cells, the number of viable wild-type LM found within the phagosomes decreased to 10% of total viable LM, whereas the percentages of cytosolic bacteria increased to 90%. With
plcA/plcB-deficient mutants, phagosomal and cytosolic percentages of viable LM were similar to wild-type bacteria. However,
LLO
plcA/plcB-deficient mutants exhibited significantly reduced numbers of viable LM, all of which were phagosomally localized in both Ctsd/ and Ctsd+/+ cells.
LLO-deficient mutants did not grow intracellularly in Ctsd+/+ cells. In Ctsd/ cells, they showed an unexpected vigorous intracellular growth, which was similar to that of wild-type LM. In fact, 99% of viable
LLO-deficient mutants remained phagosomal, and only 1% localized in the cytosol.
|
LLO-deficient mutants. Intact LLO protein could be isolated from Ctsd/ fibroblasts infected with wild-type LM, but it was not detected in Ctsd+/+ fibroblasts, suggesting in vivo degradation of LLO by Ctsd (Fig. 2A).
|
Ctsd is involved in the early MAC resistance to LM
MAC are responsible for the early resistance to LM (19). Within BM-MAC of Ctsd+/+ mice, wild-type LM displayed an exponential growth preferentially between 10 and 18 h postinfection (Fig. 3A). After infection with wild-type LM, BM-MAC generated from Ctsd/ mice exhibited 4- to 5-fold higher LM replication indexes as compared with BM-MAC from Ctsd+/+ mice (Fig. 3B).
plcA/plcB LM mutants also showed increased replication indexes, but growth of
LLO
plcA/plcB mutants was fully abrogated (Fig. 3B), supporting the notion that LLO is a putative substrate for Ctsd. We observed high numbers of intracellular wild-type FITC-labeled LM in the perinuclear regions of BM-MAC from Ctsd/ mice (Fig. 3, E, G, I, and K). Confocal microscopy (Fig. 3, D and E) suggested that LM were freely localized within the cytosol as FITC-LM colocalized with tetramethylrhodamine isothiocyanate (TRITC)-phalloidin. The number of FITC-LM in BM-MAC from Ctsd+/+ mice was significantly lower (Fig. 3, D, F, H, and J). We performed in vivo experiments with Ctsd/ mice (11). Ctsd/ and Ctsd+/+ p17 mice were inoculated i.p. with a 1 LD50 LM dose of
1 x 104 CFU. Three different experiments were performed with a total of 14 Ctsd/ mice and 15 Ctsd+/+ littermates (p < 0.01). The number of CFU per spleen observed in Ctsd/ mice 72 h postinfection was
10-fold higher compared with similarly treated Ctsd+/+ mice (Fig. 3C). Similar results were observed when mice were inoculated with a 0.1 LD50 LM dose, and bacterial burdens were evaluated in spleens and livers 48 h after injection (data not shown). These results underline the in vivo role of Ctsd among the innate mechanisms that restrict LM exponential growth at early phases of listeriosis.
|
| Discussion |
|---|
|
|
|---|
In permissive cells, early interactions of phagosomes with endosomes transport Rab5a and Ctsd precursor to LM phagosomes (10) before LLO and PLCs become active (1). The low pH environment (10, 21) and Rab5a may activate certain amounts of Ctsd to cleave membrane-inserted LLO. In support of this model, LLO is degraded in Ctsd+/+ fibroblasts as well as in an in vitro degradation assay. This mechanism could diminish LM intraphagosomal viability to some extent, although the remaining viable LM might rapidly inactivate Rab5a and Ctsd (15), allowing LM escape into the cytosol. In Ctsd/ fibroblasts, the lack of Ctsd and Rab5a confers a more permissive environment and LLO action takes place without restriction, supported by the observation of higher numbers of viable intraphagosomal LM and of LM growing in the cytosol. These phagosomes resembled those of Rab5a-deficient cells that allowed growth of
LLO-deficient mutants (10) in a similar manner to Ctsd/ cells.
Because MAC are critical effector cells of the innate response to LM (19), our findings in vivo indicate that Ctsd has an important and novel role in early listeriosis. Ctsd/ mice were 10-fold more susceptible to LM infection, compared with Ctsd+/+ littermates.
This work on the role of Ctsd in the clearance of LM constitutes the first in vivo study demonstrating Ctsd as a bactericidal agent effective against intracellular bacteria.
| Acknowledgments |
|---|
| Disclosures |
|---|
|
|
|---|
| Acknowledgments |
|---|
| Footnotes |
|---|
1 This work was supported by Fondo de Investigaciones Sanitarias (FIS) Grants 01/3128, 03/1009, and 04/0324; Fundacion Marques de Valdecilla (FMV)-Instituto de Formacion e Investigacion Marques de Valdecilla (IFIMAV) Grants 01/02, 01/03, and 04/01; BIO2002-00628 and Gobierno Cántabro (to C.A.-D. and E.C.-M.); the Deutsche Forschungsgemeinschaft, SFB 679, TPC4 (to P.S.); and FIS and FMV (to F.L.-C.). C.A.-D. is the recipient of a Ramón y Cajal contract; E.C.-M. is the recipient of a FIS-Instituto de Salud Carlos II Contract 00/3073; F.M.-T. is the recipient of a FMV-IFIMAV fellowship; and L.F.-P. is supported by FIS Grant 03/1009 (to E.C.-M.). ![]()
2 E.d.C.-V., F.M.-T., P.S., and C.A.-D. contributed equally to this work. ![]()
3 Address correspondence and reprint requests to Dr. Carmen Alvarez-Dominguez, Servicio de Inmunología, Hospital Universitario Marqués de Valdecilla, Avenida de Valdecilla s/n, 39008 Santander, Spain. E-mail address: calvarez{at}humv.es ![]()
4 Abbreviations used in this paper: LM, Listeria monocytogene; Ctsd, cathepsin-D; MAC, macrophage; BM-MAC, bone marrow MAC; PNS, postnuclear supernatants; PLC, phospholipase C; TRITC, tetramethylrhodamine isothiocyanate. ![]()
Received for publication June 30, 2005. Accepted for publication November 17, 2005.
| References |
|---|
|
|
|---|
. 2000. The contribution of both oxygen and nitrogen intermediates to the intracellular killing mechanisms of C1q-opsonized Listeria monocytogenes by the macrophage-like IC-21 cell line. Immunology 101: 83-89. [Medline]
listericidal action is mediated by novel rab5a functions at the phagosomal environment. J. Biol. Chem. 276: 19059-19065.
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |