Splenic red pulp macrophages are intrinsically superparamagnetic and contaminate magnetic cell isolates

Abstract

A main function of splenic red pulp macrophages is the degradation of damaged or aged erythrocytes. Here we show that these macrophages accumulate ferrimagnetic iron oxides that render them intrinsically superparamagnetic. Consequently, these cells routinely contaminate splenic cell isolates obtained with the use of MCS, a technique that has been widely used in immunological research for decades. These contaminations can profoundly alter experimental results. In mice deficient for the transcription factor SpiC, which lack red pulp macrophages, liver Kupffer cells take over the task of erythrocyte degradation and become superparamagnetic. We describe a simple additional magnetic separation step that avoids this problem and substantially improves purity of magnetic cell isolates from the spleen.

Figure 1. RPM contaminate MCSs.

(A) Cells isolated by MCS using CD11c-specific nanoparticles from spleens of Spi-C-deficient and –competent mice, stained for F4/80 and CD11c. (B) Expression of surface markers on RPM identified as shown in (a). Isotype controls given as gray background and autofluorescence (AF) in the 488 nm channel by analyzing unstained cells. (C) RPM contaminations in B and T cells isolated with CD19- or CD3ε-specific mircobeads. (D) Expression of CD19 on B-Cells and RPM (left), as well as expression of CD3ε on CD4 + T cells and RPM (right) in the 650 nm channel. RPM were identified as shown in (a), B cells as B220⁺ CD11c⁻, CD4⁺ T cells as CD4⁺ CD11c⁻. Isotype controls given as gray background. (E) RPM enrichment in splenocytes passed over MCS columns (input) without using paramagnetic nanoparticles, either applying a magnetic field (+Magnet) or not (–Magnet). (F) Ferric iron content of F4/80-negative splenocytes and RPM detected with Prussian blue staining. (G) Colorimetric determination of the iron content of RPM. Depicted is the total iron amount isolated from 1 × 10⁶ RPM according to the manufactures protocol. Results are shown for one representative of two to three individual experiments using 2–4 mice per group. Error bars, s.d. (n = 2–4 mice); *p < 0.05; ***p < 0.001.

Figure 2. RPM show superparamagnetic properties comparable to those of commercially available magnetic beads.

(A) SQUID analysis of magnetic properties of RPM, F4/80⁻splenocytes, lymph node cells and paramagnetic nanoparticles as a positive control. Depicted is the magnetic moment at room temperature as a function of the external magnetic field applied. Characteristic magnetization curves for superparamagnetic (RPM, Magnetic Nanoparticles) and diamagnetic behavior (lymph node cells, F4/80-negative Splenocytes) were recorded. (B) Ferromagnetic response of RPM (left graph) and magnetic nanoparticles (right graph) at 10 K measured in a SQUID magnetometer. (C) Superparamagnetic response of RPM (left graph) and magnetic nanoparticles (right graph) at 300 K measured in a SQUID magnetometer. (D) Temperature dependence of low-field susceptibility of RPM was measured in a magnetic field of 200 Oe after the sample was cooled in zero magnetic field (ZFC, blue) and in an external magnetic field of 200 Oe (FC, red). (E) Magnetic moment of one superparamagnetic (likely Fe Oxide) nanoparticle in the RPM or one magnetic nanoparticles obtained from a fit of superparamagnetic M(H, 300K) dependencies. (F) Enrichment of liver Kupffer cells in SpiC-deficient and—competent mice after passage over MCS columns (input) without using paramagnetic nanoparticles, either applying a magnetic field (+Magnet) or not (–Magnet). Results are shown for one representative of two to three individual experiments using 2–4 mice per group. Error bars, s.d. (n = 2–4 mice); *p < 0.05; ***p < 0.001.

Figure 3. Removing RPM by MCS.

(A) RPM contaminations in CD11c⁺ cell isolates obtained without RPM-depletion or after RPM-depletion on the indicated column. (B) Relative yield of splenic DCs after RPM-depletion using the negative selection column compared to isolation without depletion. (C) RPM contaminations in CD11c⁺ cell isolates obtained without RPM-depletion or after RPM-depletion using the indicated protocol. (D) Relative yield of splenic DCs after RPM-depletion using the two antibody-based depletion protocols compared to isolation without depletion. (D) RPM contaminations in B or T cell isolates with or without RPM-depletion using the F4/80-bead protocol. (E) CD11c⁺ cell isolates obtained without (left tube) or with RPM depletion using the 3-antibody protocol (right tube). Results are shown for one representative of two to three individual experiments using 2–4 mice per group. Error bars, s.d. (n = 2–4 mice); *p < 0.05; ***p < 0.001.

Figure 4. Functional consequences of RPM contaminations.

(A) IL-6 concentrations in supernatants from B cell (left graph), DC (middle graph) or T cell (right graph) cultures obtained by MCS from spleens or the mesenteric lymph nodes with or without RPM-depletion using the 3 antibody-based protocol. (B) mRNA levels of MR determined by RT-PCR. Cell type, source tissue, SpiC-expression and RPM-depletion as indicated. (C) mRNA-expression of GITRL (left graph) or PPARγ (right graph) in CD11c⁺ cell isolates with or without RPM-depletion using the 3 antibody-based protocol. (D) mRNA-expression of GITRL (left graph) or PPARγ (right graph) in sorted RPM, CD11c⁺ CD8⁺ DC or CD11c⁺ CD8⁻ DCs. (E, F) γδ T cells stained for the RPM-Marker MR (E, black line) or CD163 (F, black line) or the corresponding isotype (grey background). Left histograms: marker expression without excluding RPM, right histograms: marker expression when the analysis gate is free of autofluorescent RPM. The average (mean) values ± s.e.m. are shown. Results are shown for one representative of two to three individual experiments using 2–4 mice per group. Error bars, s.d. (n = 2–4 mice); *p < 0.05; ***p < 0.001.