Factors of the bone marrow microniche that support human plasma cell survival and immunoglobulin secretion

Abstract

Human antibody-secreting cells (ASC) in peripheral blood are found after vaccination or infection but rapidly apoptose unless they migrate to the bone marrow (BM). Yet, elements of the BM microenvironment required to sustain long-lived plasma cells (LLPC) remain elusive. Here, we identify BM factors that maintain human ASC > 50 days in vitro. The critical components of the cell-free in vitro BM mimic consist of products from primary BM mesenchymal stromal cells (MSC), a proliferation-inducing ligand (APRIL), and hypoxic conditions. Comparative analysis of protein-protein interactions between BM-MSC proteomics with differential RNA transcriptomics of blood ASC and BM LLPC identify two major survival factors, fibronectin and YWHAZ. The MSC secretome proteins and hypoxic conditions play a role in LLPC survival utilizing mechanisms that downregulate mTORC1 signaling and upregulate hypoxia signatures. In summary, we identify elements of the BM survival niche critical for maturation of blood ASC to BM LLPC.

Fig. 1

BM-MSC support in vitro survival of blood ASC. a Short-term survival of blood ASC in irradiated MSC (iMSC) co-cultures. FACS purified blood ASC (1000/well) were cultured in RPMI with 10% FBS (R10) (No MSC) or co-cultured with 50,000 BM-MSC (MSC) for 7 days (p < 0.001; ANOVA). Representative images of Elispot wells are shown. b Long-term survival of blood ASC and iMSC co-cultures from two different adult blood samples after vaccination. Five hundred to 1500 ASC with 25,000 or 30,000 iMSC were co-cultured in each well and IgG Elispots were performed on designated days. c, d Pro-survival support of BM-MSC is independent of cell–cell contact. Same number of blood ASC were cultured in R10 (Media), co-cultured with 50,000 iMSC (iMSC), in transwells with ASC and 50,000 iMSC in separate chambers (TW), or iMSC secretome (Secretome). Representative Elispot wells are shown in d. e Long-term survival of blood ASC in MSC secretome. One thousand ASC were cultured in MSC secretome or with media alone (p < 0.03; linear regression modeling using media (secretome) as a covariate). In a–c, e, Elispot assays were performed on the indicated day and the frequency (%) of IgG-secreting ASC were calculated based on the maximal Elispots on days 1, 2, or 3. Each figure is representative of >3 different experiments

Fig. 2

MSC secretome restores the secretory function of sorted ASC. a Flow cytometric analysis of intracellular Ki-67 expression of blood ASC. b Effects of antibody staining and sorting on ASC survival and function. IgG Elispot assays were performed on the same number of PBMC that were untouched (No Stain or Sort), only stained with antibodies and not FAC sorted (Stain), only FAC sorted and not stained with antibodies (Sort), or both stained and FAC sorted (Stain & Sort). Maximal number of IgG Elispots occurred in the No Stain or Sort wells and normalized to 100% (*p < 0.002; ANOVA). c BrdU incorporation of MSC and ASC. BrdU was added to ASC cultured in the secretome or secretome alone. Triplicate cultures of dividing noniMSC alone with and without BrdU served as positive and negative controls. BrdU was also added to conditions of secretome alone or secretome with ASC (*p < 0.004; ANOVA). This figure is representative of three experiments. d No BM-derived ASC contamination in iMSC co-cultures or MSC secretome. IgG Elispot assays were performed on iMSC (iMSC) or MSC secretome (Secretome). Blood ASC in MSC secretome cultures (ASC + Secretome) served as a positive control. Representative images of Elispot wells are shown below

Fig. 3

APRIL together with the MSC secretome enhances blood ASC survival. a, b Blood ASC survival is enhanced by the MSC secretome with exogenous APRIL. First two rows: sorted blood ASC were cultured in R10 alone or in media with exogenous APRIL for 7 days. Second two rows: blood ASC were cultured in MSC secretome alone or the secretome with APRIL. Blood ASC immediately from FAC sorting are shown (day 0). Representative images of Elispot wells are shown. Graphic plot of % survival of IgG ASC in media alone (R10) (open circles), APRIL alone (open squares), secretome alone (green circles), or secretome + APRIL (red triangles) (p < 10^-6; liner regression modeling between secretome alone and secretome + APRIL). c Exogenous APRIL together with the MSC secretome enhances ASC survival in short-term cultures. Blood ASC from four different subjects after vaccination were cultured in media alone (open circles), secretome alone (green circles), or secretome + APRIL (red triangles). Percentage of IgG Elipsots normalized to maximal frequency on days 1–3. Shown are p-values between secretome alone & secretome + APRIL. d Exogenous APRIL with the MSC secretome enhances ASC survival in long-term cultures. ASC cultured with the MSC secretome (green circles) or MSC secretome + APRIL (red triangles). e Treatment of MSC secretome with anti-IL-6 antibodies (anti-IL-6) diminishes blood ASC survival. Blood ASC were cultured in MSC secretome or MSC secretome treated with anti-IL-6 for 1, 3, 7 days and IgG Elispots were performed (p < 10⁻⁵; linear regression modeling between secretome vs secretome + anti-IL-6). The frequency of isotype controls was similar to MSC secretome. Representative of three experiments

Fig. 4

Integrated bioinformatics of MSC secretome proteomics with transcriptomics of blood ASC and BM LLPC. a Flow diagram of the integrated bioinformatic analysis. b Proteomics of distinct MSC secretome fractions. Three fractions strongly support blood ASC survival (blue symbols): iMSC secretome (blue filled square; left and right panels), noniMSC secretome (blue open triangles; left panel), and supernatant fractionated by ultracentrifugation of iMSC secretome (blue open circles; left panel); and three fractions with decreased ASC survival (black symbols): conventional media (R10) (black open circles; left panel), supernatant fractionated by overnight ultracentrifugation of noniMSC secretome (black open triangles; left panel), and secretome from blood (not BM) adherent cells (black open triangles; right panel). c A heat map of the 2558 DEG between 17 blood ASC (PB ASC) obtained from seven healthy subjects and BM LLPC (LLPC) obtained from four adult subjects. d HIPPIE analysis of 91 MSC protein revealed 4429 potential protein partners (PPI). Overlap of the 4429 PPI with the 2558 DEG uncovered 556 overlapping gene/protein targets and led to 20 statistically significant GSEA hallmark pathways. e Of the aforementioned 91 MSC secretome proteins, FN-1 and YWHAZ had the highest number (118 and 119, respectively) of potential interacting partners. Of these potential partners, 31 were shared between both FN-1 and YWHAZ. f From the 20 GSEA pathways, FN-1 and YWHAZ were found to be involved in these 10 potential GSEA hallmark pathways

Fig. 5

Validation of ASC survival. Treatment of MSC secretome with antibodies targeting a FN-1 or b YWHAZ diminishes blood ASC survival. Blood ASC were cultured in MSC secretome alone (white bar) or MSC secretome treated with antibodies targeting FN-1 (a; black bar) (p < 10^-3; linear regression modeling using media (secretome) as a covariate) or YWHAZ (b; black bar) (p < 10^-5; linear regression modeling using media (secretome) as a covariate), or appropriate isotype controls (gray bar). Graph represents average of at least three experiments. Sensitivity and resistance to Rapamycin in c blood ASC and d BM LLPC. Blood ASC from three healthy subjects or BM LLPC from three adults were sorted and cultured in untreated MSC secretome (black bar) or MSC secretome treated with rapamycin (gray bar). The frequency of vehicle controls (DMSO) was comparable to that of untreated MSC secretome. p = 0.001; ANOVA) (c); NS, non-significance (d)

Fig. 6

Survival of blood ASC is enhanced under hypoxic conditions with the MSC secretome and APRIL. a Circulating ASC post-tetanus vaccination from one healthy adult were cultured under the following conditions in normoxia: media alone (black open squares), secretome alone (open red squares), or secretome + APRIL (red closed squares) and under hypoxia conditions: media alone (black open circles), secretome alone (blue circles), or secretome + APRIL (blue closed circles). Percentage of IgG Elispots were normalized to the maximal frequency on day 3. Data are representative of four different blood ASC from four adults after vaccination (p-values shown in the table below). b Representative images of Elispot wells from the experiments in a are shown. Number of ASC per well: 1333 except on wells from days 21–56 for MSC secretome + APRIL in normoxia or hypoxia (800) to show images in a countable range