In vivo, multimodal imaging of B cell distribution and response to antibody immunotherapy in mice

Abstract

Background: B cell depletion immunotherapy has been successfully employed to treat non-Hodgkin's lymphoma. In recent years, increasing attention has been directed towards also using B-cell depletion therapy as a treatment option in autoimmune disorders. However, it appears that the further development of these approaches will depend on a methodology to determine the relation of B-cell depletion to clinical response and how individual patients should be dosed. Thus far, patients have generally been followed by quantification of peripheral blood B cells, but it is not apparent that this measurement accurately reflects systemic B cell dynamics.

Methodology/principal findings: Cellular imaging of the targeted population in vivo may provide significant insight towards effective therapy and a greater understanding of underlying disease mechanics. Superparamagnetic iron oxide (SPIO) nanoparticles in concert with near infrared (NIR) fluorescent dyes were used to label and track primary C57BL/6 B cells. Following antibody mediated B cell depletion (anti-CD79), NIR-only labeled cells were expeditiously cleared from the circulation and spleen. Interestingly, B cells labeled with both SPIO and NIR were not depleted in the spleen.

Conclusions/significance: Whole body fluorescent tracking of B cells enabled noninvasive, longitudinal imaging of both the distribution and subsequent depletion of B lymphocytes in the spleen. Quantification of depletion revealed a greater than 40% decrease in splenic fluorescent signal-to-background ratio in antibody treated versus control mice. These data suggest that in vivo imaging can be used to follow B cell dynamics, but that the labeling method will need to be carefully chosen. SPIO labeling for tracking purposes, generally thought to be benign, appears to interfere with B cell functions and requires further examination.

Figure 1. Labeling of B cells with fluorescent and MR tracers and evaluation of cellular response.

A, SPIO nanoparticles function as MR contrast agents and consist of dextran-coated iron oxide. The dextran has been aminated and labeled with the dye Alexa 680. B, In addition to SPIO, GFP-expressing B cells are also labeled with the membrane intercalating near infrared dye CellVue NIR815 (NIR815). C, Loading of B cells with SPIO (Alexa 680) and NIR815 was confirmed by fluorescence microscopy (40×). D, SPIO loading did not cause any detectable change in the expression of CD40, CD86, CD80, or MHC II. PMA treated B cells were used as a positive control. E, B cells that were loaded with SPIO could still be activated upon the addition of LPS. F, The contrast labeled cells (20×106) were tail vein injected into C57BL/6 mice on day 0. B cell trafficking and distribution was monitored by MR and optical imaging techniques at the indicated time points. Treatment of either PBS or anti-CD79 was administered following the imaging session on day 1.

Figure 2. Longitudinal MR imaging and quantification of contrast-labeled B cells prior to and following B cell depletion therapy.

A, Representative axial T2*-weighted images of mice either pre-injection (day 0) or on the indicated days following injection of contrast labeled B cells. Hypointensity of the spleen was noted in animals of the groups given SPIO-loaded B cells (block ended arrow: groups i and ii). Isointense signal was observed on all pre-contrast images and in animals injected with NIR815-labeled B cells (groups iii and iv). Either PBS or anti-CD79 was administered to each group after imaging on day 1. B, Relative signal intensity (rSI; calculated as the ratio of signal in the spleen to paraspinal muscle) is plotted, normalized to pre-contrast images. For SPIO-labeled B cell groups (i and ii, ▪) a rapid and pronounced decrease in rSI was observed. There was also only limited difference between the normalized rSI of PBS (i, solid line) and anti-CD79 (ii, dashed line) administered groups following treatment. There was no significant change in spleen signal for groups devoid of SPIO, after B cell injection (iii and iv, ○). A gradual decrease in normalized rSI was seen in the anti-CD79 treated group (iv, dashed line). C, Although SPIO-labeled B cells were not depleted following the administration of anti-CD79 antibodies as detected by MR, the number of peripheral B cells were reduced compared to PBS-treated controls. A similar depletion profile was seen for the NIR-only labeled B cell groups.

Figure 3. In vivo fluorescent imaging of contrast-labeled B cells prior to and following B cell depletion therapy.

A, Representative whole animal fluorescence images of B cells loaded with SPIO and NIR815 (groups i and ii) or just NIR815 (groups iii and iv) following injection into C57BL/6 mice. Prior to injection (day 0), no signal was evident in the NIR channel. Signal accumulated within the spleen by 24 h after the cell injection. Immediately following imaging on day 1, animals were injected with either anti-CD79 antibodies or PBS. Anti-CD79 treatment led to a rapid (by day 3) loss is signal in mice injected with B cells labeled with NIR815 only. B, The abundance of B cells, labeled with SPIO and NIR815, in the spleen was quantified by measuring the spleen-to-muscle signal–to-background ratio (SBR). Quantification of fluorescence revealed only a gradual loss in SBR following treatment with PBS (group I, solid line) and anti-CD79 antibodies (group ii, dashed line). No statistical significance was observed between the two groups (p<0.05). C, NIR815-labeled B cells (i.e. no SPIO) were rapidly depleted following administration of anti-CD79 antibodies (group iv, dashed line) compared with PBS-treated controls (group iii, solid line). Statistical significance for individual time points is indicated with an asterisk (p<0.05).

Figure 4. Fluorescent images of contrast-labeled B cells in excised organs.

A, Representative fluorescent images of the excised heart (H), lung (L), spleen (S), and liver (L) from animals injected with contrast labeled B cells. The B cells were either labeled with SPIO and NIR815 (groups i and ii) or just NIR815 (groups iii and iv), as indicated. Animals were either treated with PBS (groups i and iii) or anti-CD79 antibodies (ii and iv) following imaging on day 1. One animal per group was sacrificed at each time point, immediately following MR and in vivo fluorescent imaging. B and C, The mean fluorescence intensity (MFI) of the spleen, normalized to day 1 values for each group, is plotted for the length of the experiment. A gradual loss of signal from the spleen is seen in all groups, save the rapid decrease following treatment of NIR815-only labeled cells (iv).

Figure 5. Fluorescent images of spleens following histological sectioning.

(A–D) Representative fluorescent images of an excised and sectioned spleen, obtained from a mouse that was not subject to B cell transfer. (A) Aside from autofluorescence, there is little detectable fluorescent signal in the (A) GFP, (B) Alexa-680 (i.e. SPIO), and (C) NIR815 channels. (D) A composite image shows no significant co-localization between the fluo rescent images. (E–F) Representative images of a spleen that was excised and sectioned 7 days following the transfer of SPIO- and NIR815-labeled B cells into C57BL/6 mice. (E) Despite the high level of autofluorescence, distinct punctate areas of fluorescence could still be discerned. Presumably, these fluorescent signals signify the presence of GFP-positive B cells. Similar patterns of punctate fluorescent signals were also observed in the (F) SPIO and (G) NIR815 channels. (H) The composite image shows that there is considerable overlap between the fluorescent signals in all three channels (arrows). All images were acquired using a LUC PLAN FLN 40× objective (NA 0.6).