A Critical Role for Murine Complement Regulator Crry in Fetomaternal Tolerance

See allHide authors and affiliations

Science  21 Jan 2000:
Vol. 287, Issue 5452, pp. 498-501
DOI: 10.1126/science.287.5452.498


Complement is a component of natural immunity. Its regulation is needed to protect tissues from inflammation, but mice with a disrupted gene for the complement regulator decay accelerating factor were normal. Mice that were deficient in another murine complement regulator, Crry, were generated to investigate its role in vivo. Survival of Crry −/− embryos was compromised because of complement deposition and concomitant placenta inflammation. Complement activation at the fetomaternal interface caused the fetal loss because breeding to C3 −/− mice rescuedCrry −/− mice from lethality. Thus, the regulation of complement is critical in fetal control of maternal processes that mediate tissue damage.

Activation of complement promotes natural immunity by inducing chemotaxis of inflammatory cells, enhancing phagocytosis by neutrophils and monocytes, facilitating immune complex clearance, and mediating cell lysis by the membrane attack complex (1). Complement can also bind and attack self tissues, especially in areas of active inflammation. In vitro studies have shown that cells are protected from the deleterious effects of complement by proteins that regulate complement activation (2).

Three membrane-bound proteins regulate the activation of the third and fourth components of complement (C3 and C4) on the surface of murine and human cells (3). Decay accelerating factor (DAF) inactivates the C3 convertase enzymes that activate C3. Membrane cofactor protein (MCP) serves as a cofactor for factor I–mediated degradation of activated C3 and C4. Crry, present only in rodents (4–6), regulates the deposition of activated C3 and C4 on the surface of autologous cells in vitro by exhibiting MCP- and DAF-like activities, although its relative contribution to complement regulation as compared to mouse MCP and DAF has not been elucidated.

Although decreased expression of complement regulatory molecules has been found in different inflammatory disorders (7), their specific contribution to pathogenesis is largely unknown. To investigate the role of complement regulation in vivo, we generated mice deficient in Crry by inserting a neomycin resistance gene that disrupted exon 5 of the mouse Crry gene in embryonal stem (ES) cells (8, 9). Three targeted ES cell clones with the expected homologous recombination were identified by Southern blotting (10). Two independently isolated cell clones were used to generate chimeric mice that subsequently transmitted the mutant allele to their progeny. Heterozygous germ line mutants appeared healthy and fertile.

Heterozygous animals were intercrossed to generate Crry null mice. However, no Crry −/− mice could be recovered from a total of 245 births, indicating that Crry deficiency resulted in embryonic lethality (Table 1). To determine the stage of lethality, we collected and genotyped embryos at various stages of development. At 9.5 days post coitus (dpc) or earlier, embryos with the expected frequency of the homozygous mutation were detected (∼25%). In contrast, the percentage of homozygous mutants declined progressively thereafter. In addition, most Crry −/− embryos at 9.5 dpc had signs of developmental arrest, such as the smaller deciduae resembling those of earlier stages (Fig. 1, A and B). Decidua dissection revealed developmentally arrested, and sometimes deceased, embryos (Fig. 1, D through J). To confirm that the targeted mutation was a null allele, we studied protein expression from primary fibroblasts prepared from 13.5-dpc Crry-deficient embryos. In contrast to the wild-type control, staining with a Crry-specific antibody revealed that theseCrry −/− fibroblasts did not express detectable Crry protein as determined by flow cytometry (8, 11). The insertional mutation therefore behaves as a null allele. These results suggest that Crry plays a crucial role during early embryonic development.

Figure 1

Developmental arrest ofCrry −/− embryos.Crry +/+ (left panels) andCrry −/− (right panels) embryos. (Aand B) Whole deciduae from 9.5-dpc embryos, ×1 magnification. (C and D) Decidua dissection to expose the yolk sac containing a 9.5-dpc embryo, ×2 magnification. (E and F) A 9.5-dpc embryo, ×2 magnification. (G and H) Decidua dissection to expose the yolk sac containing a 10.5-dpc embryo, ×1 magnification. (I andJ) A 10.5-dpc embryo, ×2 magnification. d, maternal decidua; ys, yolk sac [dashed areas in (D) and (H)]; em, embryos.

Table 1

Genotype analysis of littermates fromCrry +/− matings. Numbers in parentheses represent the percentage of the total.

View this table:

To determine the role of Crry on this developmental defect, we first analyzed its expression pattern in wild-type early embryos. Immunohistochemical detection of Crry in cryosectioned embryos indicated that Crry is highly expressed in trophoblasts as early as 7.5 dpc, with little expression in the embryo proper (12). In addition, Crry is also expressed in the maternally derived decidual tissues (Fig. 2A). This expression pattern persists in later stages of embryonic development (examined up to 16 dpc) (13). As expected, there is no Crry expression inCrry −/− trophoblast and embryos (Fig. 2B).

Figure 2

Spontaneous complement activation in Crry-deficient embryos (×10 magnification). Staining of (A)Crry +/+ and (B)Crry −/− 7.5-dpc embryos with a rabbit antibody to mouse Crry (α-Crry) or (C)Crry +/+ and (D)Crry −/− embryos with an antibody to mouse C3 (α-C3). d, maternal decidua; ec, ectoplacental cone; ep, embryo proper; tr, trophoectoderm.

Given that Crry has been implicated as a negative regulator of complement activation and because we demonstrated that it is normally expressed in embryonic tissue, we hypothesized that the developingCrry −/− embryos died from their inability to suppress spontaneous complement activation and tissue inflammation in the areas around the decidua and trophoectoderm. To test this hypothesis, we compared the state of C3 activation onCrry +/+ and Crry −/−embryos by staining with an antibody to mouse C3. In principle, native C3 is only present in soluble form, whereas activated C3 binds to the cell surface. In contrast to wild-type embryos (Fig. 2C),Crry −/− embryos had surface-deposited C3 in the trophoectoderm and the ectoplacental cone. Thus, the lack of Crry was associated with abnormal activation and deposition of complement (Fig. 2D).

To test if the spontaneous activation of C3 was the major mechanism by which embryonic lethality is observed in theCrry −/− mice, we examined the effect of this mutation in a C3-deficient background (14). To this end, we generated compound mutant mice that wereCrry +/− and C3 −/− and subsequently intercrossed them to generate mutants that were either Crry-sufficient or Crry-deficient in the C3-deficient background. Genotype analysis revealed that 27% (11 out of 41) of the resulting 3-week-old pups were Crry-deficient mutants, in contrast to the absence of Crry −/− newborns derived from the crossing of C3 +/+ Crry +/− mice (Table 1), indicating that the Crry −/−embryonic lethality results from the activation of complement (15).

One potential consequence of complement activation is the establishment of an inflammatory reaction in the target tissue due to the recruitment and activation of granulocytes (2). To investigate if the absence of Crry initiated a similar reaction, we examined histological sections of 7.5- and 8.5-dpc embryos. In contrast to wild-type controls, Crry −/− embryos were extensively invaded by polymorphonuclear inflammatory granulocytes in areas around the ectoplacental cone and the surrounding trophoectoderm (Fig. 3). Thus, failure to control complement activation leads to an inflammatory response in the Crry −/− fetuses and, eventually, to embryonic demise.

Figure 3

Infiltration of polymorphonuclear cells in the extraembryonic tissues of the Crry-deficient embryos. (A)Crry +/+ embryo and (B)Crry –/– embryo at ×10 magnification. (C) A ×40 magnification of the boxed area in (B). (D) A ×100 magnification of the boxed area in (C). Arrows denote polymorphonuclear cells. d, maternal decidua; ep, embryo proper.

It has been suggested that complement may be important in the reproductive system and in pregnancy (16). Our data indicate that complement regulation is indeed important in fetoplacental survival, maintenance of normal pregnancy, and adequate reproductive function by maintaining a form of fetomaternal tolerance against immunological mechanisms of tissue damage related to natural immunity.

Our studies are also relevant to the possible involvement of these molecules in pathologic pregnancies, both in animals and in humans, in which complement is believed to be involved in the disease pathogenesis (17). Given that mouse Crry and human DAF and MCP control C3 activation by the same biochemical mechanisms in vitro (4–6), we provide insight into the roles of the corresponding functional molecules in vivo that was not appreciated by in vitro analysis or by examination of structural orthologs. Mouse DAF is not expressed in early embryos and trophoblasts (13), and mouse MCP is exclusively expressed in the testes (18), thus leaving Crry as the critical regulator of complement activation during early murine embryonic development. These observations explain the difference in phenotype of the Crry-deficient embryos and mutant mice lacking DAF, in which embryonic development is not affected (19). In contrast, DAF and MCP are heavily expressed in human placentas (16), and there is no direct human counterpart to Crry. Thus, human DAF or MCP should play a similar role as mouse Crry during early embryonic development by controlling effector components of natural immunity, in the form of complement regulation, to protect fetomaternal tissues from tissue inflammation and destruction.

  • * These authors contributed equally to this work.

  • To whom correspondence should be addressed. E-mail: hmolina{at}


View Abstract

Stay Connected to Science

Navigate This Article