Complement in Human Pre-implantation Embryos: Attack and Defense

Abstract

It is essential for early human life that mucosal immunological responses to developing embryos are tightly regulated. An imbalance of the complement system is a common feature of pregnancy complications. We hereby present the first full analysis of the expression and deposition of complement molecules in human pre-implantation embryos. Thus, far, immunological imbalance has been considered in stages of pregnancy following implantation. We here show that complement activation against developing human embryos takes place already at the pre-implantation stage. Using confocal microscopy, we observed deposition of activation products on healthy developing embryos, which highlights the need for strict complement regulation. We show that embryos express complement membrane inhibitors and bind soluble regulators. These findings show that mucosal complement targets human embryos, and indicate potential adverse pregnancy outcomes, if regulation of activation fails. In addition, single-cell RNA sequencing revealed cellular expression of complement activators. This shows that the embryonic cells themselves have the capacity to express and activate C3 and C5. The specific local embryonic expression of complement components, regulators, and deposition of activation products on the surface of embryos suggests that complement has immunoregulatory functions and furthermore may impact cellular homeostasis and differentiation at the earliest stages of life.

Keywords: complement; development; embryo; mucosal immunology; pre-implantation; reproductive immunology.

Figure 1

Expression of complement genes in developing embryos. Single-cell RNA sequencing was applied on oocytes (O), zygotes (Z), 4-cell stage, and 8-cell stage embryos to identify the expression of complement related genes during early development. Included are reads tagged to the 5′UTR (Coding 5′-UTR) or proximal upstream region (Coding upstream), as well as known coding sequence (Coding CDs). The numerical values indicating normalized expression levels are determined relative to concentration of spike-in RNAs comparing two developmental stages at a time, in three individual libraries (O vs. Z, O vs. 4-cell stage, and 4-cell vs. 8-cell stages). Expression levels are averaged over n = 6 to 29 embryonic cells, as indicated. White denotes no expression observed for a given TFE in a particular library. While the data reveal the presence of non-degraded mRNA from the displayed genes, statistical analysis did not reveal significant variations in gene expression from one developmental stage to another. For detailed functional description of identified genes see below and (1, 2, 27). ADIPOR1/2, adiponectin receptor ½; C4BBP, C4 binding protein chain B; C1QBP, gC1qR/C1q globular domain-binding protein; CALR, calreticulin receptor; CFI, complement Factor I; CLU, Clusterin; CR2, complement receptor 2; C5AR1, C5a Receptor 1; CFB, Complement factor B; CFD, Complement factor D; ITGB2, Integrin beta chain-2 (part of complement receptor 3 and 4); MASP, MBL associated serine protease.

Figure 2

Complement targets developing embryos. Human cleavage stage in vitro fertilization (IVF) embryos were thawed in Vitrolife G-TL serum-free media. The embryos were then incubated with specific anti-complement antibodies and analyzed by confocal microscopy. Analysis of embryonic binding of complement activation products revealed binding of the classical pathway initiator C1q (A), and deposition of cascade activation components C3c/C3b/iC3b (B), and C3d (C). Finally, activation of the terminal pathway is evidenced by deposition of C5 (D). (A) Left panel: Single plane, overlay of C1q (green) and DAPI (blue). Middle panels top to bottom: DAPI, C1q, and BF. Right panels: Magnification of overlay (orange insert), and below the cross-sectional distribution of fluorescence intensity. This shows C1q is specifically found on the cell surface. (B–D) Left panels: 3D rendering, overlay of protein stain (green), DAPI (blue), and F-actin (magenta). Right panels top to bottom: DAPI, protein stain, and brightfield (BF). Scale bars: 50 μm, insert: 10 μm. For each staining, n = 3 to 4 + 1 to 3 (2PN + 3PN embryos).

Figure 3

Embryonic expression of surface-tethered complement inhibitors. Human cleavage stage IVF embryos were thawed in Vitrolife G-TL serum-free media. The embryos were incubated with anti-complement antibodies and analyzed by confocal microscopy. The analysis revealed a clear staining for both CD55 and CD59, particularly at cellular junctions. In contrast, no positive signal was observed for CD46. (A) CD46 (B) CD55 (C) CD59. Left panels: Single plane, overlay of protein stain (green) and DAPI (blue). Second column panels top to bottom: DAPI, protein, and BF. Third column panels: Magnification of overlay (orange insert), and below the cross-sectional distribution of fluorescence intensity. Right panels (B,C): 3D rendering, overlay of protein stain and DAPI. Scale bars: 50 μm, insert: 10 μm. For each staining, n = 3 + 7 (2PN + 3PN embryos).

Figure 4

Embryonically bound soluble complement regulators. Human cleavage stage IVF embryos were thawed in Vitrolife G-TL serum-free media. The embryos were incubated with anti-complement antibodies and analyzed by confocal microscopy. Embryonic binding of the soluble complement regulators C4bp and factor H are displayed. (A) C4bp is recruited to the embryonic surface and show strong staining on the cell membrane. No binding is observed to the ZP. (B) Factor H stains both the blastomere surface as well as the ZP. Left panels: single planes, overlay of protein stain (green) and DAPI (blue). Right panels top to bottom: DAPI (blue), protein stain (green), and BF. Scale bars: 50 μm. For each staining, n = 3 + 5 (2PN + 3PN embryos).

Figure 5

Functional overview of the embryonic complement system. Indicated are the canonical functional roles of membrane-expressed complement regulators (squares), soluble complement components (circles), and their cleaved activated membrane-deposited forms (demi-circles). Finally, embryonically expressed complement receptors are shown (pentagons). All molecules depicted were found to be expressed or bound by the embryos in this study (exceptions: C4, mannose binding lectin (MBL), ficolins (FCNs), and some MAC-components). The classical and the lectin pathways are initiated by target-binding of pattern recognition molecules such as C1q, or MBL and FCNs, respectively. Utilizing their associated proteases C1r/s or MASPs, they activate C4 and C2, which subsequently activate C3. Alternative pathway activation of C3 occurs when factor D cleaves C3-associated factor B, which generates a novel C3-cleaving enzyme; C3bBb. Cleavage of C3 by either pathway leads to generation of soluble C3a and surface-deposited C3b, which amplifies alternative pathway C3 activation, and subsequently activates C5 to C5a and C5b. Finally, C5b initiates the assembly of the pore-forming MAC (1, 36). To avoid excessive immunological targeting of self, human cells express or recruit inhibitors of complement activation. Membrane regulators may function; (1) by disrupting the enzymes cleaving C3 and C5 (CR1, CD55), (2) as co-factors for factor I-mediated degradation of C3b and C4b (CR1, CD46), or (3) by inhibiting MAC-formation directly (CD59). Soluble regulators such as factor H and C4bp inhibit activation by mechanisms 1 and 2, while clusterin work through mechanism 3. Inactivation of C3b, leads to generation of iC3b, C3dg, and finally C3d. While C3b and iC3b function as opsonins for increased phagocytosis by antigen presenting cells (through CR3 and CR4), C3d has important biological functions as an important internal adjuvant aiding antigen uptake by dendritic cells and inducing efficient antibody responses in B cells (through CR2) (1, 36, 37). While CR2, CR3 and CR4 expression is mainly described on immune cells, our study found embryonic expression of these receptors, along with the signaling receptors for C1q; calreticulin (CALR) and C1qbp.