The JI
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     
 

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow Request Permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Gatto, D.
Right arrow Articles by Bachmann, M. F.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Gatto, D.
Right arrow Articles by Bachmann, M. F.
Right arrowPubmed/NCBI databases
*Substance via MeSH
The Journal of Immunology, 2007, 178: 67-76.
Copyright © 2007 by The American Association of Immunologists, Inc.

Regulation of Memory Antibody Levels: The Role of Persisting Antigen versus Plasma Cell Life Span

Dominique Gatto1, Stephen W. Martin1, Juliana Bessa, Erica Pellicioli, Philippe Saudan, Heather J. Hinton and Martin F. Bachmann2

Cytos Biotechnology, Zurich-Schlieren, Switzerland


   Abstract
Top
 Abstract
Introduction
 Materials and Methods
 Results
 Discussion
 Disclosures
 References
 
Protective Ab levels can be maintained for years upon infection or vaccination. In this study, we studied the duration of Ab responses as a function of the life span of plasma cells and tested the role of persisting Ag in maintaining B cell memory. Our analysis of B cell responses induced in mice immunized with virus-like particles demonstrates the following: 1) Ab titers are long-lived, but decline continuously with a t1/2 of ~80 days, which corresponds to the life span of plasma cells; 2) the germinal center (GC) reaction, which lasts for up to 100 days, is dependent on Ag associated with follicular dendritic cells; and 3) early GCs produce massive numbers of plasma and memory B cell precursors, whereas the late Ag-dependent GCs are dispensable for the maintenance of Ab levels and B cell memory.


   Introduction
Top
 Abstract
 Introduction
Materials and Methods
 Results
 Discussion
 Disclosures
 References
 
The success of current vaccines is based on the induction of long-lived Ab responses. Despite this, relatively little is known about how the duration of Ab responses is regulated. It has been shown that plasma cells can be very long-lived in experimental animals (1, 2). In light of this, it seems plausible that such long-lived plasma cells may maintain Ab titers for long periods of time in an Ag-independent fashion, as has been observed in individuals exposed to vaccinia virus (3). In contrast, Ag can persist on follicular dendritic cells (FDCs)3 for long periods of time, and potentially contribute to the maintenance of Ab responses in an Ag-dependent fashion (4, 5). In support of this, it has been shown that increasing levels of persisting Ag can result in increased Ab titers, and long-lived germinal centers (GC) associated with Ag-carrying FDCs have been observed (6, 7). Ag may also persist in the absence of FDCs, for example during latent viral infections or after vaccination, in which the use of adjuvants results in local Ag depots (8). Thus, it is difficult to create experimental conditions that exclude Ag persistence. A further mechanism that may be used to maintain Ab memory is the nonspecific stimulation of memory B cells by TLR ligands, which activates memory B cells to differentiate into Ab-producing plasma cells (9). Hence, the relative contribution of long-lived bone marrow plasma cells and continually differentiating memory B cells to the maintenance of humoral memory remains controversial (8, 10, 11).

In the present study, we used virus-like particles (VLPs) derived from the bacteriophage Qbeta (12) to study the regulation of B cell and Ab memory. We have shown previously that these VLPs are highly immunogenic in the absence of adjuvant and efficiently persist on FDCs (13, 14). Thus, VLPs are able to induce strong B cell responses in the absence of infection or long-term Ag deposits, other than those persisting on FDCs. BrdU-labeling experiments were performed to study the life span of memory B cells and plasma cells and relate their turnover with the maintenance of specific IgG Ab titers. To modulate Ag persistence, we depleted FDCs at various time points after immunization using a lymphotoxin (LT) beta receptor (LTbetaR)-Ig fusion protein. LT{alpha}beta signaling is required for the maintenance of mature FDC networks via LTbetaR expressed by FDCs and LT{alpha}1/beta2 expressed by B cells. Inhibition of LT{alpha}1/beta2 signaling by the injection of LTbetaR-Ig fusion proteins has been shown to cause a rapid disappearance of functional FDCs and the markers specific of this population, such as FDC-M1, FDC-M2, and CR1 (15, 16). Furthermore, LTbetaR-Ig treatment has been demonstrated to prevent the trapping of newly formed immune complexes, as well as to eliminate previously sequestered Ags (15). Earlier studies have reported that injection of LTbetaR-Ig before immunization abolished GC formation and led to impaired Ab responses to SRBC (16). However, the effect of LTbetaR-Ig treatment after the B cell response has been induced and, in particular, at late time points after immunization has not yet been addressed.

In this study, we report that the maintenance of the GC reaction, early B cell proliferation, and the development of B cell memory were highly Ag dependent, whereas persisting Ag was not essential for the maintenance of B cell and Ab memory in the late phase of the response. GCs had a high output of memory B cells and plasma cells within the first month after immunization. At later time points, the contribution of the ongoing GC reaction to the pool of long-lived memory B cells and bone marrow plasma cells became negligible. In this late phase of the response, Ab titers declined with a t1/2 of ~3 mo. The kinetics of this decline was dictated by the t1/2 of the plasma cell population, which was found to be in a similar range. These findings suggest a major role for long-lived plasma cells residing in the bone marrow in maintaining memory Ab levels and attribute an important role to persisting Ag in the early, but not late phase of the B cell response.


   Materials and Methods
Top
 Abstract
 Introduction
 Materials and Methods
Results
 Discussion
 Disclosures
 References
 
Mice and Ags

C57BL/6 mice (Harlan) were immunized i.v. with 10 µg of Qbeta. Animal protocols used were reviewed and approved by the Swiss Federal Veterinary Office.

Qbeta capsids were expressed using the vector pQbeta10 and purified, as described (12).

Depletion of FDCs and retained Ag

LTbetaR-Ig was prepared by fusing the coding region of the extracellular domain of the LTbetaR to the C region of human IgG1. The construct was transfected into EBV-encoded nuclear Ag cells (Invitrogen Life Technologies) using Lipofectamine Plus (Invitrogen Life Technologies) and cultured in serum-free medium. The LTbetaR-Ig fusion protein was purified using protein G-Sepharose columns (Pharmacia).

Mice were injected i.p. with 300 µg of LTbetaR-Ig on days 9 and 11 (early FDC depletion), 39 and 41 (late FDC depletion), and 100 and 102 (very late FDC depletion) after immunization with Qbeta particles.

BrdU labeling

For BrdU labeling, either BrdU (Sigma-Aldrich) was administered as a 0.8 mg/ml solution in the drinking water (light protected and changed every second day), or 1 mg of BrdU was injected in saline solution i.p.

ELISAs

ELISAs were performed, as described (13). Titers represent log2 dilutions of 40-fold prediluted sera at half-maximal OD.

ELISPOT assay

Qbeta-specific Ab-forming cell (AFC) frequencies were determined, as described (14). Briefly, 24-well plates were coated with 10 µg/ml Qbeta. Spleen or bone marrow cells were added in MEM containing 2% FCS and incubated for 5 h at 37°C. Cells were washed off and plates were incubated successively with goat anti-mouse IgG (EY Laboratories) and alkaline phosphatase-conjugated donkey anti-goat IgG Abs (Jackson ImmunoResearch Laboratories) before development of alkaline phosphatase color reactions.

Flow cytometry

In all cases, Fc receptors were blocked with anti-mouse CD16/32 (2.4G2). Abs were purchased from BD Biosciences, unless otherwise specified.

Detection of B cells expressing Qbeta-specific surface Ig was performed by incubation with Qbeta, followed by a polyclonal rabbit anti-Qbeta serum (produced by RCC) and Cy5-conjugated donkey anti-rabbit IgG serum (Jackson ImmunoResearch Laboratories), as previously described (13). Isotype-switched B cells were detected in two ways. 1) cells were stained with the following FITC-conjugated Abs: anti-IgD (11-26c.2a); goat anti-IgM serum (Jackson ImmunoResearch Laboratories); anti-CD4 (GK1.5); anti-CD8 (53-6.7); anti-CD11b (M1/70); anti-Gr-1 (RB6-8C5); and PerCP-Cy5.5-conjugated anti-CD19 (1D3) Abs. Biotinylated peanut agglutinin (PNA; Vector Laboratories) and streptavidin PE were used to assess PNA binding. Dead cells were excluded by staining with 0.005 µg/ml YO-PRO-1 (Molecular Probes). 2) Cells were stained with the following CyChrome-conjugated Abs: anti-IgD (11-26c.2a); anti-CD4 (GK1.5); anti-CD8 (53-6.7); anti-CD11b (M1/70); anti-Gr-1 (RB6-8C5); and PE-conjugated anti-CD19 (ID3). In this case, BrdU incorporation was detected by intracellular staining using a FITC-conjugated anti-BrdU Ab (B44) after cell permeabilization, as described below.

To detect intracellular Qbeta-specific Ig-positive bone marrow cells, surface Qbeta-specific Ig was blocked with excess unlabeled Qbeta VLP. Cells were permeabilized with 2x FACS lysing solution (BD Biosciences; 349202) containing 0.06% (v/v) Tween 20. Intracellular Qbeta-specific Ig was detected by staining with Qbeta particles labeled with Alexa 647 (Molecular Probes), prepared according to the manufacturer’s instructions. Simultaneous detection of BrdU incorporation was performed using a FITC-conjugated anti-BrdU Ab (B44), as described below.

Detection of incorporated BrdU

BrdU incorporation was measured by the method of Tough and Sprent (17). Briefly, surface-labeled and/or intracellular Ig-labeled spleen and bone marrow cells were fixed in ice-cold 95% (v/v) ethanol for 30 min, followed by permeabilization in 1% (w/v) paraformaldehyde containing 0.01% (v/v) Tween 20. Cells were treated with 50 Kunitz U/ml bovine pancreatic DNase I and stained subsequently with FITC-labeled anti-BrdU (B44; BD Biosciences).

Immunohistochemistry

Freshly removed organs were snap frozen in liquid nitrogen. Tissue sections of 5- to 7-µm thickness were cut in a cryostat and fixed with acetone. Qbeta Ag was detected by incubating sections with rabbit anti-Qbeta serum (RCC), followed by biotinylated sheep anti-rabbit Igs (The Binding Site) and alkaline phosphatase-labeled streptavidin (Roche). Alternatively, Alexa 488-labeled secondary Abs were used for detections. FDCs were visualized using the rat FDC-M1 (BD Biosciences) Abs, followed by biotinylated rabbit anti-rat Abs (Jackson ImmunoResearch Laboratories) and alkaline phosphatase-labeled streptavidin (Roche). IgD-expressing B cells were stained with sheep anti-mouse IgD Abs (The Binding Site) and HRP-labeled rabbit anti-sheep Abs (Jackson ImmunoResearch Laboratories). Alkaline phosphatase was visualized using the Vector Blue substrate (Vector Laboratories) and HRP with the substrate diaminobenzidine. GC B cells were stained with PNA biotin, followed by Alexa 546-labeled streptavidin (red). Qbeta-specific B cells were stained with Alexa 488-labeled Qbeta (green).

Statistical analysis

Levels of statistical significance between means were determined using Student’s t test. Average life spans of cell populations were calculated from multiple experiments. Results are indicated as average ± SEM. Only time points after day 45 were used for the analysis.


   Results
Top
 Abstract
 Introduction
 Materials and Methods
 Results
Discussion
 Disclosures
 References
 
Characterization of the B cell response induced by immunization with VLPs derived from Qbeta

We have shown previously that a single injection of VLPs derived from the bacteriophage Qbeta elicits strong and long-lasting IgG responses, which are strictly dependent on T cell help (13) and are mainly of the IgG2a isotype (our unpublished observation). As described before, mice immunized i.v. with 10 µg of Qbeta mounted a strong VLP-specific IgG response that peaked ~3–4 wk later and declined thereafter (Fig. 1A). The decay of the response at later time points occurred with a t1/2 of 80 days (±7 days). Immunization resulted in high numbers of VLP-specific IgG-secreting AFCs in the spleen and bone marrow (Fig. 1B). After the peak of the response, numbers of AFCs declined over time both in spleen and bone marrow. To quantify isotype-switched Qbeta-specific B cells, we used a previously described flow cytometry-based detection system (Fig. 1C) (13, 14); the specificity of the staining was controlled using VLPs from bacteriophage AP205 (data not shown). As observed for AFCs, numbers of Qbeta-specific isotype-switched B cells in the spleen gradually decayed with time (Fig. 1D). Immunization with Qbeta induced efficient GC reactions (Fig. 1, E and F): up to 80% of all Qbeta-specific isotype-switched B cells were PNAhigh 3 wk after immunization, and PNAhigh cells were maintained at low levels at least up to day 100 (Fig. 1E) (13). Histological analysis revealed large GCs containing Qbeta Ag (Fig. 1F), which colocalized with FDCs (data not shown), as well as with Qbeta-specific B cells (Fig. 1G). Moreover, large numbers of plasmablasts were observed outside GCs (Fig. 1G). Because Qbeta particles cannot replicate in the host and administration in adjuvant is not required, they are only expected to persist as immune complexes on the surface of FDCs.


Figure 1
View larger version (35K):
[in this window]
[in a new window]

  		FIGURE 1. Kinetics of Qbeta-specific B cell response. Mice were immunized i.v. with 10 µg of Qbeta, and the B cell response was followed over time. A, Kinetics of the IgG immune response directed against Qbeta Ag. Anti-Qbeta IgG titers are given as serum dilutions reaching half-maximal absorbance at 450 nm. B, Number of Qbeta-specific IgG-producing AFCs was enumerated by ELISPOT in the spleen and bone marrow in response to Qbeta immunization. C, Representative staining showing Qbeta-binding isotype-switched B cells in spleen of immunized and naive mice. D, The number of Qbeta-binding CD19+IgM–IgD– cells in the spleen was assessed by flow cytometry. The calculated limit of detection was 2 x 104 cells. E, The percentage of GC B cells within the pool of Qbeta-specific GC B cells was assessed by staining Qbeta-binding CD19+IgM–IgD– B cells additionally using PNA. F and G, Histological sections were stained 12 days after s.c. immunization for PNA (red), identifying GC, Qbeta-specific B cells (blue), as well as persisting Qbeta Ag (green). F, An overlay for PNA and persisting Ags is shown. G, An overlay for Qbeta-specific B cells and persisting Ags is shown. Note that Qbeta-specific B cells are found in either GCs or large aggregates as plasmablasts outside B cell follicles.

 
Rapid turnover of splenic Qbeta-specific B cells vs slow turnover of bone marrow plasma cells

To characterize the life span and turnover of splenic Qbeta-specific B cells and bone marrow plasma cells, BrdU-labeling experiments were performed. BrdU, which is a DNA base analog, is incorporated into the DNA of proliferating cells, marking them at a given time point. The life span of a cell as measured in such experiments is defined as the time it takes for a cell to either divide, which leads to loss of BrdU, or die. For an optimal estimation of life span, two types of experiments should be performed, as follows: 1) all cells are labeled initially, the treatment is stopped, and the number of BrdU+ cells followed over time; 2) cells are exposed to BrdU at later time points, and rate of accumulation of BrdU+ cells is measured. The combination of these two methods allowed us to distinguish between the loss of BrdU due to proliferation vs cell death. For this purpose, we immunized mice with Qbeta and administered BrdU into the drinking water during the first 20 days of the response (Fig. 2A). Qbeta-specific splenic B cells that had incorporated BrdU were clearly detectable by flow cytometry, even at late time points when present at low frequencies (Fig. 2, B and C). Numbers of Qbeta-specific plasma cells in the bone marrow were assessed similarly by flow cytometry, and were in the same range as those obtained by ELISPOT assay (Fig. 1B). The specificity of this intracellular staining has been demonstrated previously (14). As expected, virtually all VLP-specific splenic B cells and plasma cells in the bone marrow were BrdU+ on days 10–20 (Fig. 2, D and F). However, the frequency of BrdU+ Qbeta-specific B cells in the spleen rapidly declined by ~80% within a few days, but remained stable thereafter (Fig. 2D). Note, however, that the overall frequency of Qbeta-specific B cells also declined over time (t1/2 of the population is 20 ± 4 days) and, at later time points, the frequency of BrdU+ cells within the overall declining population of Qbeta-specific B cells remained constant (Fig. 2E). This indicates that the BrdU+ B cells were lost with the same kinetics as the total population of specific B cells and, consequently, that the life span of BrdU+ B cells was the same as the life span of the specific B cell population as a whole. In contrast to Qbeta-specific B cells in the spleen, the loss of BrdU+ bone marrow plasma cells specific for Qbeta occurred more slowly and the frequency of BrdU+ plasma cells declined over more than 1 mo (Fig. 2F). This indicates that bone marrow plasma cells were generated at a lower rate than splenic B cells. At later time points, the overall number of specific BrdU+ plasma cells declined slowly, but the frequency of BrdU+ plasma cells remained constant, indicating that the life span of BrdU+ plasma cells was the same as the life span of the population (Fig. 2G). The decay of plasma cells was biphasic, the later phase being characterized by a t1/2 of 80 ± 21 days, roughly corresponding to the decay of the IgG titers (Fig. 1A).


Figure 2
View larger version (31K):
[in this window]
[in a new window]

  		FIGURE 2. Rapid turnover of splenic B cells vs slow turnover of bone marrow plasma cells: BrdU labeling from day 0 until day 20. A, Mice were immunized i.v. with 10 µg of Qbeta, and BrdU was administered in the drinking water for the first 20 days until the peak of the response. Mice were sacrificed subsequently at the indicated time points, and B cells were assessed for incorporation of BrdU. B, Representative staining showing labeling of BM Qbeta-specific B cells in immunized and naive mice. C, Representative staining showing BrdU labeling of splenic Qbeta-specific B cells. Splenic Qbeta-specific CD19+IgM–IgD– B cells were stained for BrdU. As a control, spleen cells from immunized mice, but not pulsed with BrdU were stained. D, The percentage of Qbeta-specific B cells that were BrdU+ is shown. E, The frequency of total Qbeta-specific B cells and the frequency of BrdU+ Qbeta-specific cells are shown. Note that a fraction of the spleens was used for histology, and absolute numbers of Qbeta-specific B cells therefore cannot be given. The detection limit was assessed as being 0.04% for total Qbeta-specific B cells and 7 x 10–3% for BrdU+ Qbeta-specific B cells. F and G, Qbeta-specific plasma cells of the bone marrow were stained for BrdU. F, The percentage of Qbeta-specific plasma cells that were BrdU+ is shown. G, The number of total Qbeta-specific plasma cells and the number of BrdU+ Qbeta-specific plasma cells are shown. The detection limit was 7 x 102 for total Qbeta-specific plasma cells and 100 for BrdU+ Qbeta-specific plasma cells.

 
These results were complemented by a second experiment, in which BrdU labeling was performed from day 20 and BrdU incorporation followed over 100 days (Fig. 3A). In the spleen, 80% of the specific B cells incorporated BrdU within a few days, whereas 20% remained unlabeled for the rest of the experiment (Fig. 3B), most probably corresponding to the population of B cells that did not lose the BrdU label in Fig. 2D. BrdU labeling in the pool of bone marrow plasma cells was less pronounced and only reached ~40% of the cells at the end of the experiment. Thus, a substantial population of nondividing plasma cells was already present in the bone marrow at day 20 after immunization (Fig. 3C).


Figure 3
View larger version (16K):
[in this window]
[in a new window]

  		FIGURE 3. Rapid turnover of splenic B cells vs slow turnover of bone marrow plasma cells: BrdU labeling from day 20 until day 110. A, Mice were immunized i.v. with 10 µg of Qbeta, and BrdU was administered in the drinking water from day 20 until day 110. Mice were sacrificed subsequently at the indicated time points, and B cells and plasma cells were assessed for incorporation of BrdU. B, Splenic Qbeta-specific CD19+IgM–IgD– B cells were stained for BrdU. The percentage of Qbeta-specific B cells that were BrdU+ is shown. C, Qbeta-specific plasma cells of the bone marrow were stained for BrdU. The percentage of Qbeta-specific plasma cells that were BrdU+ is shown.

 
A sizeable population of B cells shows high turnover late after immunization

To determine the proportion of B cells that were actively dividing at different time points after immunization, BrdU was added to the drinking water for 10 days preceding analysis (Fig. 4A). In the spleen, a surprisingly large fraction of B cells was still actively dividing at later time points; indeed, ~30% of Qbeta-specific B cells proliferated between days 90 and 100 after immunization (Fig. 4, B and C). This observation was consistent with the ongoing GC reaction at these late time points. In contrast, Qbeta-specific BrdU+ bone marrow plasma cells were found in large numbers only early after immunization, and by day 100 only a small fraction of plasma cells was positive for BrdU (Fig. 4, D and E). Thus, there is ongoing proliferation in the splenic B cell population, whereas there is almost no turnover in the bone marrow plasma cell population. These data indicate that the two populations are rather independent, and that, therefore, few proliferating splenic B cells enter the bone marrow plasma cell pool at later time points.


Figure 4
View larger version (29K):
[in this window]
[in a new window]

  		FIGURE 4. Presence of a proliferating population of B cells up to 100 days after immunization. A, Mice were immunized i.v. with 10 µg of Qbeta. BrdU was administered in the drinking water for periods of 10 days, and mice were sacrificed at the indicated time points. B cells and plasma cells were assessed for incorporation of BrdU. B and C, Splenic Qbeta-specific CD19+IgM–IgD– B cells were stained for BrdU. B, The percentage of Qbeta-specific B cells that were BrdU+ is shown. C, The number of total Qbeta-specific B cells and the number of BrdU+ Qbeta-specific cells are shown. Calculated detection limits were 5 x 104 for total Qbeta-specific B cells and 5000 for BrdU+ Qbeta-specific B cells. D and E, Qbeta-specific plasma cells of the bone marrow were stained for BrdU. D, The percentage of Qbeta-specific plasma cells that were BrdU+ is shown. E, The number of total Qbeta-specific plasma cells and the number of BrdU+ Qbeta-specific plasma cells are shown. The limit of detection was 7 x 102 for total Qbeta-specific plasma cells and 100 for BrdU+ Qbeta-specific plasma cells.

 
Depletion of FDCs and associated Ag by treatment with LTbetaR-Ig

In a next set of experiments, we tested the importance of Ag in driving B cell responses and memory by depleting FDCs at various time points after immunization using a LTbetaR-Ig fusion molecule. It has been shown previously that LTbeta is continuously required for survival of FDCs and that these cells rapidly die if the action of LTbeta is blocked by LTbetaR-Ig (15, 16). To evaluate the efficacy of this treatment in depleting FDCs and the associated Ag, mice were immunized with Qbeta-derived VLPs, and LTbetaR-Ig fusion protein was injected on days 9 and 11 after immunization. Disappearance of FDCs and trapped Qbeta particles was assessed by immunohistochemistry 1 wk after LTbetaR-Ig treatment. As shown in Fig. 5A, injection of LTbetaR-Ig led to the loss of FDCs, as revealed by an absence of staining for the FDC-specific marker FDC-M1. Consistent with the disappearance of FDCs, no Qbeta Ag could be detected in mice treated with LTbetaR-Ig, in contrast to the readily visible deposits present in untreated control mice (Fig. 5B). The activity of the LTbetaR-Ig fusion protein was further confirmed by the reduction in frequencies of marginal zone B cells observed in treated mice (data not shown), which is in agreement with the reported alteration of the marginal zone organization caused by inhibition of LTbeta signaling (16, 18). Thus, a substantial depletion of FDC-associated Ag was achieved by treatment with LTbetaR-Ig.


Figure 5
View larger version (69K):
[in this window]
[in a new window]

  		FIGURE 5. Depletion of FDCs and associated persisting Ag by LTbetaR-Ig treatment. Mice were immunized i.v. with 30 µg of Qbeta. LTbetaR-Ig injections were performed on days 9 and 11 after immunization. A, Immunohistochemical staining of spleen sections with FDC-specific Abs FDC-M1 (blue) and anti-IgD (brown) 1 wk after treatment with LTbetaR-Ig (right panels) and in control untreated mice. B, Histological analysis of Qbeta Ag (stained in blue) on spleen sections from LTbetaR-Ig-treated mice and control untreated mice. Note that injection of control IgG preparations did not affect Ag persistence (data not shown). C, Quantification of GC reactions by histological analysis. Spleen sections were stained 22, 56, and 122 days after immunization for PNA and Qbeta-specific B cells in untreated and LTbetaR-treated mice. On the left side, an overlay for PNA (red) and Qbeta-specific B cells (green) is shown for a spleen 122 days after immunization.

 
LTbeta inhibition and disappearance of FDCs have been shown to prevent the formation of GC in the spleen (16). We therefore expected that treatment with LTbetaR-Ig fusion protein would lead to perturbed GC reactions also when administered after their formation at later stages of the response, thus emphasizing their output at several stages after immunization. To confirm the longevity of GCs induced by immunization with Qbeta particles, we assessed the number of VLP-specific PNA+ GCs by histology (Fig. 5C). GC numbers were highest 3 wk after immunization and declined thereafter. Few GCs could still be observed >100 days after immunization, indicating that GC reactions were still ongoing, as observed by flow cytometry. No Qbeta-specific GCs were detected in spleens of LTbetaR-Ig-treated mice, indicating that they had been disrupted by the treatment (Fig. 5C).

Role of FDCs in maintaining the early B cell responses

We studied the effect of depleting FDCs and the related role of Ag persisting on this cell population at three different stages of the B cell response, as follows: early (~day 10), late (~day 40), and very late (~day 100) after immunization. For these analyses, the frequencies of Qbeta-specific isotype-switched B cells and GC B cells in the spleen, numbers of Qbeta-specific IgG AFC in the bone marrow, and anti-Qbeta serum IgG titers were determined. In the first set of experiments, FDCs and retained Qbeta Ag were depleted by injections of LTbetaR-Ig on days 9 and 11 after immunization, i.e., shortly before GC reactions reach their peak, and the impact on the B cell response was assessed 10 and 45 days after the last LTbetaR-Ig treatment (corresponding to days 21 and 56 after immunization). As shown in Fig. 6A, isotype-switched B cells specific for Qbeta were 10-fold reduced in spleens of treated animals compared with controls 10 days after depletion of FDCs and the associated Ag. Qbeta-specific GC B cells were also drastically (13-fold) reduced, and they accounted for most of the loss of specific cells, because 80–90% of Qbeta-binding B cells were PNAhigh at this time point (Fig. 6A). Ten days after depletion, frequencies of anti-Qbeta IgG AFCs in the spleen (data not shown), but not those in the bone marrow, were reduced (Fig. 6A). This is consistent with the very short life span of splenic AFCs at this early time point compared with the slower turnover of bone marrow plasma cells (Fig. 1B). However, no short-term effect of the treatment on specific serum Ab titers was observed. The analysis of mice 45 days after FDC depletion highlighted a major impact of an early LTbetaR-Ig treatment on the long-term Ab response induced by Qbeta. At this later analysis time point, the role of FDCs and the associated Ag in establishing the slowly turning over bone marrow plasma cell pool and memory Ab titers becomes evident. Indeed, treated mice displayed a 3-fold reduction in anti-Qbeta serum Ab levels (Fig. 6A), which was the result of a corresponding reduction of Qbeta-specific AFCs in both the spleen (data not shown) and bone marrow (Fig. 6A). Similarly to what was observed at the earlier time point of analysis, frequencies of Qbeta-specific isotype-switched B cells were also lower in spleens of mice that had been depleted of FDC-associated Ag (Fig. 6A). Thus, in the early stages of B cell responses, FDCs and the Ag retained on their processes play a crucial role in the maintenance of GCs and in establishing long-term B cell and Ab memory.


Figure 6
View larger version (36K):
[in this window]
[in a new window]

  		FIGURE 6. Distinct effects of FDC depletion on GC formation vs plasma cell induction. Mice were immunized with Qbeta particles and treated subsequently with LTbetaR-Ig at day 10 (A), day 40 (B), or day 100 (C) postimmunization. The frequency of isotype-switched Qbeta-specific B cells and of Qbeta-specific PNAhigh GC B cells as well as the frequency of Qbeta-specific AFCs in the bone marrow and Qbeta-specific IgG Abs in the serum are shown. Note that no reduction in Qbeta-specific total or GC B cells was observed in mice injected with control Ig protein (data not shown). All data represent the mean ± SEM; mean values statistically different in treated mice are indicated by asterisks (*, p < 0.05; **, p < 0.01; ***, p < 0.001).

 
Role of FDCs in maintaining the late B cell responses

We next addressed the importance of Ag retained on FDCs at later time points. For this purpose, mice were treated with LTbetaR-Ig on days 39 and 41 after immunization with Qbeta; the effect of the treatment was analyzed subsequently 21, 50, and 106 days after the last injection of the fusion protein (corresponding to days 60, 89, and 145 after immunization). Depletion of FDCs and the associated Ag resulted in a short-term reduction of isotype-switched Qbeta-specific B cells (Fig. 6B). As expected from the fact that GC reactions were still ongoing, the largest impact was seen on the frequency of PNAhigh GC B cells. However, generation and/or maintenance of memory B cells were not significantly affected by this late injection of the LTbetaR-Ig, as demonstrated by comparable frequencies of Qbeta-binding B cells present in spleens of treated and control mice 145 days after immunization (Fig. 6B). The same was true for the number of plasma cells in the bone marrow and long-term Ab titers, which were not affected by the treatment. This suggests that most Qbeta-specific memory B cells and plasma cells had been generated within the first month after immunization. At later time points, the GC reaction may still be ongoing; however, the net output for the pool of memory B and plasma cells was negligible after day 40, a result consistent with the BrdU-labeling experiments shown above.

In accordance with these results, depletion of persisting Ag very late after immunization (by LTbetaR-Ig treatment on days 100 and 102) had no noticeable effect on the frequency of Qbeta-specific memory B cells or on the number of bone marrow plasma cells secreting anti-Qbeta Abs (Fig. 6C). The number of PNAhigh GC B cells was slightly, but not significantly reduced by the treatment, indicating that the late presence of PNAhigh cells was largely Ag independent. Thus, the maintenance of Qbeta-specific long-term B cell effector populations, bone marrow plasma cells, and memory B cells did not require the presence of the Ag persisting on FDCs.

Role of persisting Ag in driving B cell proliferation

The impact of FDC depletion on B cell proliferation was assessed next. Accordingly, mice were immunized with Qbeta, and FDCs were depleted 10 days later by LTbetaR-Ig treatment. Ten days after FDC depletion, BrdU was injected, and the frequencies of Qbeta-specific BrdU+ B cells and plasma cells were determined 24 h later (Fig. 7A). This early depletion of FDCs and associated Ag had a strong impact on the frequency of splenic Qbeta-specific BrdU+ B cells (Fig. 7A), as well as on PNAhigh B cells (data not shown). Thus, FDC-associated Ag was required for proliferation of B cells in the spleen early after immunization. As expected, the treatment had less short-term influence on the frequency of Qbeta-specific BrdU+ plasma cells accumulating in the bone marrow (Fig. 7B).


Figure 7
View larger version (29K):
[in this window]
[in a new window]

  		FIGURE 7. FDC depletion blocks B cell proliferation. Mice were immunized with Qbeta particles and treated with LTbetaR-Ig 10 days (A and B) or 100 days (C) postimmunization. A and B, Ten days after FDC depletion, BrdU was injected, and the analysis was performed 24 h later (i.e., on day 11 after depletion). The percentage of BrdU+ Qbeta-specific splenic B cells (A) and bone marrow plasma cells (B) was determined by flow cytometry. C, BrdU was added to the drinking water either 12 or 46 days after depletion for a period of 10 days. The percentage of Qbeta-specific splenic B cells that were BrdU+ was determined 22 and 56 days after depletion. Values are expressed as the mean ± SEM, with significant differences between means indicated by asterisks (*, p < 0.05; **, p < 0.01).

 
Surprisingly, when FDCs were depleted 100 days after immunization, proliferation of Qbeta-specific B cells in the spleen was hardly affected (Fig. 7C), indicating that the low turnover seen at this late time point occurs Ag independently (note that the fraction of Ag-dependent PNAhigh GC B cells is low at this late time point; Fig. 6C). Potential mechanisms for this proliferation, including the role of nonspecific stimulation by TLR ligands or cytokines, remain to be established.


   Discussion
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
Disclosures
 References
 
In this study, we determined the relation between maintenance of Ab titers, plasma cell life span, and turnover of B cells. In addition, the role of persisting Ag in driving B cell proliferation, GC reactions, and plasma cell differentiation was analyzed.

The life span of plasma cells in the bone marrow has been the subject of intensive discussion. Although some authors suggested the plasma cells to be short-lived, others concluded that they may live for up to several years (8, 10, 11). Most experimental systems used to date exhibit some technical limitations, rendering the assessment of plasma cell life span, under physiological conditions in the absence of exogenous Ag, difficult. On the one hand, mice have been immunized with proteins in depot-forming adjuvants, leading to persistence of Ag in an unphysiological manner, different from the natural Ag persistence on FDCs. Live lymphocytic choriomeningitis virus was also used to measure the life span of plasma cells (1). However, lymphocytic choriomeningitis virus tends to remain in the host as infectious virus (19). If high amounts of Ag persist in a depot or as infectious virus, B cell responses may remain ongoing through continuous antigenic stimulation, potentially leading to an underestimation of the plasma cell life span. In contrast, low level persistence of Ag may stimulate differentiation of memory B cells into plasma cells in the absence of proliferation, leading to an overestimation of the plasma cell life span in BrdU-labeling experiments.

To resolve this issue, we immunized mice with noninfectious VLPs derived from bacteriophage Qbeta (20) in the absence of adjuvant. These VLPs are rapidly transported into B cell follicles and efficiently retained on FDCs (13, 14), but are not expected to persist in any other form in the host. This efficient trapping of VLPs on FDCs and their ability to activate Th cell and induce T cell-dependent isotype switching lead to the generation of long-lasting GC reactions and strong long-lived IgG responses. Recent findings have revealed that the presence of contaminating TLR ligands in preparations of pneumococcal polysaccharides or hepatitis B nucleocapsids is required for induction of IgG Ab responses by these Ags (21, 22). Although endotoxin levels are low in our VLP preparations (1–10 EU LPS/µg capsid), we cannot exclude that contaminating bacterial LPS might enhance the immunogenicity of the VLPs and contribute to the activation of B cells. In addition, the VLPs contain bacterial RNA, a potential ligand for TLR3 and TLR7. We are currently studying the role of this RNA in driving B cell responses in detail. Nevertheless, the VLP preparations fail to induce isoytpe switching in the absence of T cells (13, 23) or CD40-CD40L interactions (our unpublished observations), indicating that Th cells, more than TLR-mediated costimulatory signals, are critical for the induction of IgG to Qbeta. The requirement for T cell help for the generation of IgG responses strongly suggests that GC reactions induced by immunization with VLPs are T cell dependent and hence support affinity maturation. This notion is supported by reduced GCs in absence of costimulation as well as the presence of somatic hypermutations in Ab sequences from VLP-specific B cells (our unpublished data).

Using Qbeta particles and an Ag-specific B cell detection system, we measured turnover and life span of B cells and plasma cells. As expected, we observed a relatively high turnover of splenic B cells in contrast to a lower turnover of bone marrow plasma cells. Moreover, two populations of splenic B cells were identified, a short-lived population with a turnover of a few days (5 days; all cells of the early phase of Fig. 4C and a subpopulation in the later phase) and a population of long-lived splenic B cells with a life span of ~43 days (late phase of Fig. 2E). In contrast, the life span of bone marrow plasma cells was found to be longer, in the range of 3 mo (late phase of Fig. 2G). The observation that in the late phase of the response the number of plasma cells declines exponentially indicates that at any given point in time, the cells have the same probability to die. Hence, plasma cells do not become old, and their expected future life span is independent of their age. The simplest explanation for such behavior is a competition model, whereby cells compete for niches within the bone marrow. Importantly, the calculated plasma cell life span of 80 days closely paralleled the decline of memory IgG titers, which also exhibited a t1/2 of ~80 days. Thus, the life span of plasma cells dictates the lifetime of the Ab response. These data also show that plasma cells may live for several months, but eventually decline as a population with a given t1/2 of ~80 days.

Ag may persist on FDCs for months or even years, and has been postulated to be important for driving B cell proliferation, plasma cell differentiation, and Ab production (4, 5, 24). FDCs are particularly important for the GC reaction, because the native Ags retained on their surface via Fc and complement receptors maintain the specificity of the GC and allow for selection of high-affinity B cell clones, leading to affinity maturation. Recently, the importance of Ag associated with FDCs has been questioned, and the idea has been put forward that FDCs maintain the GC reaction nonspecifically (25, 26), because Ag-specific GCs could be observed in the absence of detectable persisting Ag. Two explanations may account for this surprising observation. On the one hand, the detection methods used may simply not be sensitive enough or, in contrast, Ag leaking from the adjuvant depot may have allowed to artificially keep the GC reaction ongoing. Thus, the bulk of the data still indicates that the key role of FDCs is to expose native Ag to GC B cells (4, 5, 7, 27). By depleting FDCs and associated Ag, we found that B cell proliferation at early time points after immunization was strongly FDC dependent; in addition, the GC reaction remained FDC dependent, for most of the observed time span. B cell turnover at very late time points (>day 100) occurred slowly, and only ~10% of splenic B cells incorporated BrdU within a 10-day labeling period. Surprisingly, this late B cell proliferation occurred in the absence of FDCs and retained Ag, indicating that the population of memory B cells may be maintained by slow proliferation by cytokines, as has been observed for IL-15-dependent proliferation of CD8+ memory T cells (28, 29, 30). Alternatively, memory B cell proliferation may be due to environmental exposure to TLR ligands (9).

Despite a long-lived GC reaction, B cell proliferation, and plasma cell production, maintenance of IgG Ab titers was not dependent on these active B cell responses after day 30. Indeed, plasma cell numbers and Ab titers were independent of persisting Ag and FDCs at later time points. This observation may be explained by overall low numbers of plasma cells produced by the GC at later time points. In addition, it is possible that the generation of precursors for long-lived plasma cells is restricted to the first 30 days, and that the late GC reaction fails to generate these cells. The late GC reaction may therefore be more important to maintain a flexible, hypermutated B cell repertoire in case of re-emergence of the infection rather than for keeping high Ab levels.

Taken together, our results demonstrate that Ag deposits on FDCs are required in the early phase of B cell responses to Qbeta, when GCs afford a significant output of cells destined to differentiate into memory B cells and long-term Ab-secreting plasma cells. At later time points after immunization, although some B cells continue to proliferate in GCs, their overall contribution to the pool of memory B cells and bone marrow plasma cells is limited. These experiments indicate that Ab production late after immunization with Qbeta could be maintained in the absence of Ag retained on FDCs, and suggest a major role for long-lived plasma cells residing in the bone marrow in preserving long-term elevated serum Abs. However, the presented results establish that the pool of bone marrow plasma cells does not survive for the lifetime of a mouse, but declines with a t1/2 of ~3 mo.


   Disclosures
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 Disclosures
References
 
The authors have no financial conflict of interest.


   Footnotes
 
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

1 D.G. and S.W.M. contributed equally to this work. Back

2 Address correspondence and reprint requests to Dr. Martin F. Bachmann, Cytos Biotechnology AG, Wagistrasse 25, CH-8952 Zurich-Schlieren, Switzerland. E-mail address: martin.bachmann{at}cytos.com Back

3 Abbreviations used in this paper: FDC, follicular dendritic cell; AFC, Ab-forming cell; GC, germinal center; PNA, peanut agglutinin; VLP, virus-like particle; LT, lymphotoxin; LTbetaR, LT beta receptor. Back

Received for publication April 26, 2006. Accepted for publication October 4, 2006.


   References
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 Disclosures
 References
 

  1. Slifka, M. K., R. Antia, J. K. Whitmire, R. Ahmed. 1998. Humoral immunity due to long-lived plasma cells. Immunity 8: 363-372. [Medline]
  2. Manz, R. A., A. Thiel, A. Radbruch. 1997. Lifetime of plasma cells in the bone marrow. Nature 388: 133-134. [Medline]
  3. Crotty, S., P. Felgner, H. Davies, J. Glidewell, L. Villarreal, R. Ahmed. 2003. Cutting edge: long-term B cell memory in humans after smallpox vaccination. J. Immunol. 171: 4969-4973. [Abstract/Free Full Text]
  4. Williams, G. M., G. J. Nossal. 1966. Ontogeny of the immune response. I. The development of the follicular antigen-trapping mechanism. J. Exp. Med. 124: 47-56. [Abstract]
  5. Mandel, T. E., R. P. Phipps, A. Abbot, J. G. Tew. 1980. The follicular dendritic cell: long term antigen retention during immunity. Immunol. Rev. 53: 29-59. [Medline]
  6. Bachmann, M. F., T. M. Kundig, H. Hengartner, R. M. Zinkernagel. 1994. Regulation of IgG antibody titers by the amount persisting of immune-complexed antigen. Eur. J. Immunol. 24: 2567-2570. [Medline]
  7. Bachmann, M. F., B. Odermatt, H. Hengartner, R. M. Zinkernagel. 1996. Induction of long-lived germinal centers associated with persisting antigen after viral infection. J. Exp. Med. 183: 2259-2269. [Abstract/Free Full Text]
  8. Zinkernagel, R. M.. 2003. On natural and artificial vaccinations. Annu. Rev. Immunol. 21: 515-546. [Medline]
  9. Bernasconi, N. L., E. Traggiai, A. Lanzavecchia. 2002. Maintenance of serological memory by polyclonal activation of human memory B cells. Science 298: 2199-2202. [Abstract/Free Full Text]
  10. McHeyzer-Williams, M. G., R. Ahmed. 1999. B cell memory and the long-lived plasma cell. Curr. Opin. Immunol. 11: 172-179. [Medline]
  11. Manz, R. A., S. Arce, G. Cassese, A. E. Hauser, F. Hiepe, A. Radbruch. 2002. Humoral immunity and long-lived plasma cells. Curr. Opin. Immunol. 14: 517-521. [Medline]
  12. Cielens, I., V. Ose, I. Petrovskis, A. Strelnikova, R. Renhofa, T. Kozlovska, P. Pumpens. 2000. Mutilation of RNA phage Qbeta virus-like particles: from icosahedrons to rods. FEBS Lett. 482: 261-264. [Medline]
  13. Gatto, D., C. Ruedl, B. Odermatt, M. F. Bachmann. 2004. Rapid response of marginal zone B cells to viral particles. J. Immunol. 173: 4308-4316. [Abstract/Free Full Text]
  14. Gatto, D., T. Pfister, A. Jegerlehner, S. W. Martin, M. Kopf, M. F. Bachmann. 2005. Complement receptors regulate differentiation of bone marrow plasma cell precursors expressing transcription factors Blimp-1 and XBP-1. J. Exp. Med. 201: 993-1005. [Abstract/Free Full Text]
  15. Mackay, F., J. L. Browning. 1998. Turning off follicular dendritic cells. Nature 395: 26-27. [Medline]
  16. Mackay, F., G. R. Majeau, P. Lawton, P. S. Hochman, J. L. Browning. 1997. Lymphotoxin but not tumor necrosis factor functions to maintain splenic architecture and humoral responsiveness in adult mice. Eur. J. Immunol. 27: 2033-2042. [Medline]
  17. Tough, D. F., J. Sprent. 1994. Turnover of naive- and memory-phenotype T cells. J. Exp. Med. 179: 1127-1135. [Abstract/Free Full Text]
  18. Lu, T. T., J. G. Cyster. 2002. Integrin-mediated long-term B cell retention in the splenic marginal zone. Science 297: 409-412. [Abstract/Free Full Text]
  19. Ciurea, A., P. Klenerman, L. Hunziker, E. Horvath, B. Odermatt, A. F. Ochsenbein, H. Hengartner, R. M. Zinkernagel. 1999. Persistence of lymphocytic choriomeningitis virus at very low levels in immune mice. Proc. Natl. Acad. Sci. USA 96: 11964-11969. [Abstract/Free Full Text]
  20. Kozlovska, T. M., I. Cielens, D. Dreilinna, A. Dislers, V. Baumanis, V. Ose, P. Pumpens. 1993. Recombinant RNA phage Qbeta capsid particles synthesized and self-assembled in Escherichia coli. Gene 137: 133-137. [Medline]
  21. Sen, G., A. Q. Khan, Q. Chen, C. M. Snapper. 2005. In vivo humoral immune responses to isolated pneumococcal polysaccharides are dependent on the presence of associated TLR ligands. J. Immunol. 175: 3084-3091. [Abstract/Free Full Text]
  22. Vanlandschoot, P., F. Van Houtte, P. Ulrichts, J. Tavernier, G. Leroux-Roels. 2005. Immunostimulatory potential of hepatitis B nucleocapsid preparations: lipopolysaccharide contamination should not be overlooked. J. Gen. Virol. 86: 323-331. [Abstract/Free Full Text]
  23. Jegerlehner, A., T. Storni, G. Lipowsky, M. Schmid, P. Pumpens, M. F. Bachmann. 2002. Regulation of IgG antibody responses by epitope density and CD21-mediated costimulation. Eur. J. Immunol. 32: 3305-3314. [Medline]
  24. Tew, J. G., J. Wu, D. Qin, S. Helm, G. F. Burton, A. K. Szakal. 1997. Follicular dendritic cells and presentation of antigen and costimulatory signals to B cells. Immunol. Rev. 156: 39-52. [Medline]
  25. Haberman, A. M., M. J. Shlomchik. 2003. Reassessing the function of immune-complex retention by follicular dendritic cells. Nat. Rev. Immunol. 3: 757-764. [Medline]
  26. Hannum, L. G., A. M. Haberman, S. M. Anderson, M. J. Shlomchik. 2000. Germinal center initiation, variable region hypermutation, and mutant B cell selection without detectable immune complexes on follicular dendritic cells. J. Exp. Med. 192: 931-942. [Abstract/Free Full Text]
  27. Tew, J. G., T. E. Mandel, P. L. Rice. 1980. Immune elimination and immune retention: the relationship between antigen retained in the foot and the elicitation of footpad swelling. Immunology 40: 425-433. [Medline]
  28. Wherry, E. J., T. C. Becker, D. Boone, M. K. Kaja, A. Ma, R. Ahmed. 2002. Homeostatic proliferation but not the generation of virus specific memory CD8 T cells is impaired in the absence of IL-15 or IL-15R{alpha}. Adv. Exp. Med. Biol. 512: 165-175. [Medline]
  29. Berard, M., K. Brandt, S. Bulfone-Paus, D. F. Tough. 2003. IL-15 promotes the survival of naive and memory phenotype CD8+ T cells. J. Immunol. 170: 5018-5026. [Abstract/Free Full Text]
  30. Weng, N. P., K. Liu, M. Catalfamo, Y. Li, P. A. Henkart. 2002. IL-15 is a growth factor and an activator of CD8 memory T cells. Ann. NY Acad. Sci. 975: 46-56. [Medline]



This article has been cited by other articles:


Home page
 J. Immunol.Home page
P. Dosenovic, B. Chakrabarti, M. Soldemo, I. Douagi, M. N. E. Forsell, Y. Li, A. Phogat, S. Paulie, J. Hoxie, R. T. Wyatt, et al.
Selective Expansion of HIV-1 Envelope Glycoprotein-Specific B Cell Subsets Recognizing Distinct Structural Elements Following Immunization
J. Immunol., September 1, 2009; 183(5): 3373 - 3382.
[Abstract] [Full Text] [PDF]

Home page
 Proc. Natl. Acad. Sci. USAHome page
A. Ahuja, S. M. Anderson, A. Khalil, and M. J. Shlomchik
Maintenance of the plasma cell pool is independent of memory B cells
PNAS, March 25, 2008; 105(12): 4802 - 4807.
[Abstract] [Full Text] [PDF]

Home page
 CVIHome page
E. A. Clutterbuck, S. Oh, M. Hamaluba, S. Westcar, P. C. L. Beverley, and A. J. Pollard
Serotype-Specific and Age-Dependent Generation of Pneumococcal Polysaccharide-Specific Memory B-Cell and Antibody Responses to Immunization with a Pneumococcal Conjugate Vaccine
Clin. Vaccine Immunol., February 1, 2008; 15(2): 182 - 193.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow Request Permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Gatto, D.
Right arrow Articles by Bachmann, M. F.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Gatto, D.
Right arrow Articles by Bachmann, M. F.
Right arrowPubmed/NCBI databases
*Substance via MeSH

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS