Introduction

Complement is a key system in innate immunity. The proteins that make up the complement system include serine proteases, control molecules and surface receptors. Interest in the function of these proteins is high because, while complement is critical for immune defense, it is becoming increasingly clear that in many cases complement is also associated with autoimmune disease [1].

There have been myriad studies on gene regulation in many biological systems, yet in-depth study of complement gene regulation has lagged. As molecular techniques were developed that facilitated the cloning of promoter-enhancer areas, researchers delved into the inner workings of these sequences, resulting in many advances in the field. In the early 1990’s there were a number of studies published that focused on the control of complement gene expression, but later in the decade these types of studies were less common. This was the result of research directives issued at the national and international levels that mandated disease-specific spending, thus there were fewer research monies available for studies in basic molecular regulation of gene expression. Furthermore, this type of research was disparagingly dubbed “promoter bashing” and the utility of these studies was called into question. Two recent advances have rekindled interest in the transcriptional regulation of many different genes and complement genes in particular.

The human genome project has greatly facilitated the discovery of disease-associated gene variants. As part of this research and discovery effort, genetic variations as defined by single nucleotide polymorphisms (SNPs) were examined in a number of different human populations. A modest goal of the identification of 100,000 SNPs was proposed in 1998 by the DOE and NIH genome programs, and in 1999 The SNP Consortium proposed the discovery of 300,000 SNPs in 2 years [2]. These goals were greatly exceeded and today there have been more than 11 million SNPs identified in the human genome. This information has been used in screening studies to uncover gene polymorphisms that predispose individuals to disease, an effort that has been well worth the price [3].

There have been an increasing number of genetic diseases in which complement has been implicated [4–6]. In some recent cases, polymorphisms in the promoter region of specific genes including complement regulatory genes, have been associated with disease. In age-related macular degeneration (AMD), there are promoter polymorphisms in the genes for VEGF [7], in the recently described HTRA1 [8, 9] as well as in complement Factor H (CFH) [10] that correlate with disease. In atypical hemolytic uremic syndrome (aHUS), there are promoter mutations in the genes for CFH [11] and membrane cofactor protein (MCP) [12] that indicate disease predisposition. These studies show that SNPs in the promoter-enhancer region can be important in disease and thus it is imperative that some attention be focused on these regions.

This review will focus on the genetic regulation of complement receptor gene expression, with emphasis on the protective complement receptors and the anaphylatoxin receptors. Although there are many other genes in the complement family that are also critical for immune responses, space does not permit their inclusion in this report.

CR1 (CD35)

When discussing the CR1 and CR2 genes, it is important to first clarify the differences between the mouse and human variants. The human CR1 gene codes for a 220 kDa protein (the most common allelic variant) with 30 short consensus repeats (SCRs; repeating 60–70 amino acid motifs characteristic of complement regulatory genes) [13]. This protein has both decay-accelerating activity (breakdown of formed convertase molecules) as well as cofactor activity (assisting in Factor I mediated cleavage of C3b), and these activities are critical for regulation of complement activation in the local environment and also for binding to nearby complement-opsonized ligands. The cells that express CR1 include most hematopoietic types except NK cells, platelets and most T cells [13]. When a genetic screen was performed to identify the murine homolog to CR1, the gene that was discovered was much smaller, had a ubiquitous expression pattern and was called complement-related receptor-y (Crry) [14–16]. The murine CR2 gene (now commonly referred to as Cr2) was found to code for two different transcripts, with one approximately the same size as human CR2 (15 SCR containing), while the other more closely resembling the size of an allelic variant of human CR1 (with 21 SCR domains) and contained analogous amino-terminal domains [17, 18]. Further studies showed that the human CR2 gene contains CR1-like sequences found in the mouse gene, but does not produce the larger form [19]. Because of these differences in genes, the human CR1 gene and the mouse Crry gene will be considered separately, while the mouse and human Cr2/CR2 genes will be combined.

The basic genetic organization of the CR1 gene was first described in 1993, and although this report described sequences corresponding to the presumptive transcriptional control region, there were no functional assays [20]. More detailed analyses showed that there was a modest negative regulatory element between 700 and 500 bp upstream of the transcription start site and that the majority of the transcriptional activity in the promoter was located in the first 140 bp [21]. Studies in the erythroleukemia cell line, K562 showed that the region responsible for Ara-C-induced CR1 expression was located between −80 and −40 [22]. A number of studies have shown that the product of the TEL/AML1 fusion gene negatively regulates AML1 induced CR1 expression, likely through the AML1 binding site at −58 to −53 [23, 24].

There is a lack of detailed studies on CR1 transcriptional regulation, with the bulk of studies concentrated on the promoter-proximal region of the gene. As has been shown for many other genes (including CR2) the area in the first intron can be critical for gene regulation, yet there have been no published studies to date examining this region. Also, studies in primary cells are always preferable to those in transformed cells, since the transformation process can significantly alter the transcriptional activity of genes being studied [25].

CR2/Cr2 (CD21)

As detailed above, the human and murine genes in the CR2/Cr2 region are very similar, except that the mouse locus encodes two different proteins; one with analogous function to human CR2 and another that is similar to CR1 [14, 18]. These different gene products are created by alternative splicing from a common precursor transcript, however there is little evidence that there is differential expression of the splice forms in different cells or tissues. The CR2 protein binds to the C3b breakdown products, iC3b, C3d and C3dg [26]. In addition, it is also a receptor for the Epstein-Barr virus [27] and CD23 [28]. CR2 is important as part of the B cell co-receptor complex and is thus important in both adaptive immune responses and the natural antibody repertoire (reviewed in [26]). The transcriptional regulation of the CR2/Cr2 gene has been extensively studied by a number of investigators.

Initial studies identified the transcriptional start sites in the human [29, 30] and mouse genes [31]. The sequence of the human gene revealed several potential regulatory elements including AP-1, Sp1, MHC class II enhancer, and Ig enhancer E motifs [29, 30]. Curiously, constructs that included regions 5′ of the transcription start site did not confer cell-specific expression, such that activity of the reporter was more active in cells that did not express the CR2 message (liver HepG2 cells) when compared to transformed B cell lines that express the gene [30]. Functional analyses of the identified promoter-proximal transcription factor binding sites indicated that there were two critical positive regulatory elements; the Sp1 site and a site at −47 that bound to the USF1 protein (and unknown proteins) [32, 33]. These sites were also critical for induction of gene expression by PMA + cAMP [34]. Additional CREB/AP-1 sequences located approximately 900 bp upstream of the transcription start site were also important for CR2 induction [34]. Other sequences in this promoter proximal region included a general repressor and a cell-specific repressor [32, 33]. Studies in the mouse showed that the Oct-2 transcription factor was at least partially responsible for Cr2 transcription, although these experiments did not examine cell-specific regulation [31]. There were clues that the site responsible for cell specificity of CR2/Cr2 transcription was localized to sequences in the first intron.

Studies were expanded to examine the role of sequences in the first intron for cell-specific CR2/Cr2 transcription and a complex picture emerged. Once again, sequences upstream of the transcription initiation site did not direct cell-specific expression, however further analyses demonstrated that there were specific control regions in the first intron [35]. Dissection of this region showed that there were two intronic repressor sites and a cell-specific enhancer, and when an approximately 9 kb sequence from the first intron was included in the construct, expression was restricted only to cells that typically express the Cr2 gene. DNase hypersensitivity studies revealed the presence of a site in the first intron of the human gene [36], as well as in the mouse [37]. Reporter gene experiments demonstrated the specificity of the intronic region in directing CR2 transcription in a cell-specific and developmental-specific manner. The authors also used the upstream region of the CR2 promoter, with or without sequences from the intron to investigate tissue-specific regulation in vivo. These studies showed that sequences upstream of the transcription start site would only direct tissue specific expression similar to that of the normal CR2 gene when connected to the intron sequences [36]. The authors of both these reports concluded that this CR2/Cr2 intronic region contains silencer activity that is critical for gene expression.

The mechanism used to facilitate gene silencing using regions in the intron was further investigated. Since previous reports had suggested that the intron region of CR2/Cr2 would bend to contact sequences in the promoter proximal region, it was reasonable to hypothesize that chromatin structure may play a role in this process. Indeed, addition of histone deacetylase inhibitors led to an induction of Cr2 expression [38]. Experiments using DNase I sensitivity indicated that histone acetylation was a critical factor in determining the binding status of proteins in the intron [37]. The most detailed analysis of the Cr2 intron region demonstrated that there was no single site that specifically controlled gene expression; rather there exist multiple regulatory modules that act in concert to control Cr2 expression [39]. These regulatory elements included binding sequences for Yin Yang 1 (YY1), Oct1, and NFAT-4. The proposed mechanism to account for cell-specific regulation was that YY1 and Oct1 recruited histone deacetylases to the intron region, while the NFAT-4 worked to displace histone deacetylases or recruited histone acetylases that allowed the formation of a functional transcription complex. Due to all these studies, our understanding of the transcriptional control of the CR2/Cr2 gene is more complete than for any other complement receptor gene.

Crry

The Crry protein is unique to rodents. The Crry gene was first identified as the murine genetic homolog to the human CR1 gene, although it has only five SCR motifs [14, 16]. In mice and rats, this protein is ubiquitously expressed [16, 40] and has both decay acceleration activity and cofactor activity [41, 42]. Thus Crry has been proposed to function in a manner similar to both DAF and MCP in rodents. Surprisingly, Crry knockout mice were not born at the expected frequency and it was found that the protein was required for protection of the fetus from complement attack [43]. The only way in which Crry-deficient mice could be born was if the mother of the pups was C3 deficient [44]. The protection conferred was specific to the alternative pathway and not the classical pathway [43, 44]. Other studies have shown that Crry is critical for the protection of erythrocytes from alternative pathway mediated clearance [45, 46], but other studies have demonstrated that DAF is also critical for protection of red blood cells from complement [47, 48].

There has been only one study on the molecular regulation of the Crry gene. The Crry gene has an abnormally long 5′ untranslated region 277 bp relative to the most 5′ translation start site) [49]. This long sequence contains a very strong hairpin structure [50], but although the function of this sequence in gene regulation is unknown, reporter constructs that lack even a small amount of this sequence rapidly lose reporter activity [49]. In fibroblasts, approximately 65% of the transcriptional activity was localized to an enhancer at position −1380. The sequence identified had identity with the core motif used by the Ets family of transcriptional regulators, but the specific identity of this protein was not established [49]. The ubiquitous expression pattern of the Crry gene indicated that the key enhancer would also be a ubiquitous protein, and indeed, evidence was presented that this Ets family member was present in diverse tissues and cell lines [49].

In summary, there has been minimal investigation into the regulation of the Crry gene. There are several reasons for the lack of further studies. First, as this is a rodent-specific gene, there is doubt that studies on this gene will be informative in human disease. Second, this gene is ubiquitously expressed, and even during acute bacterial infection when the Cr2 gene is significantly down regulated, expression of the Crry protein remains constant [31]. In the CNS, however, Crry protein expression goes down during demyelination in the cuprizone model [Submitted]. In addition, Crry expression is decreased during vaccinia virus infection of a liver cell line [51] and the gene is also down regulated in the liver during alcohol treatment [52]. These studies indicate that in many important health situations this key complement regulator has altered expression. As will be demonstrated in the next section, the lessons learned from Crry might translate into studies for the human MCP and DAF genes. Thus additional experimentation in Crry gene regulation is a worthy line of research.

Membrane cofactor protein (MCP, CD46)

Like Crry, MCP is a ubiquitously expressed regulator of both the alternative and classical pathways of complement activation [53]. As its name implies, MCP acts a cofactor for Factor I-mediated cleavage of C3b and C4b, rendering those proteins unable to catalyze further complement activation through formation of the convertase complexes. Thus, it is a critical protein for protection of homologous tissue from complement activation. The MCP protein has been called a “pathogen magnet”, since it serves as a receptor for a number of bacteria and viruses including Measles virus, Adenovirus, Herpes virus, Neisseria species and Streptococcus pyogenes [54]. New roles for MCP are emerging in the field including T regulatory cell generation and novel functions in reproduction including placental protection and control of the acrosomal reaction in spermatozoa [55].

Early work on the genetic control of MCP gene expression began with the cloning of the 5′ region of the gene. The MCP promoter region lacks a prototypical TATA box in the usual position and instead has three different Sp1 binding sites immediately upstream of the transcription start site [56]. The majority of transcriptional activity was localized within 384 bp of the transcription start site, but this report did not further investigate the role of individual regulatory elements. The lack of further work on the regulation of the MCP gene has made it difficult to draw conclusions about disease associations in aHUS. A pair of mutations in the upstream transcriptional regulatory region of MCP correlate strongly with aHUS [12]. Reporter constructs consistently showed that the identified SNP led to a 25% decrease in promoter activity, a change that is not inconsequential when one considers the fact that many other mutations in MCP that predispose to aHUS occur in the heterozygous state, thus only affecting one allele [12]. These data clearly show the importance of the transcription control region of MCP in disease.

The expression of the murine MCP gene is very different than in humans. While the human protein is ubiquitously expressed, in the mouse it is predominantly expressed in the testis [57]. Like the CR2/Cr2 genes, reporter gene constructs consisting of only murine 5′ promoter sequences did not exhibit cell-specific regulation [58]. This repressor activity was not localized to the first intron, but rather was found between 600 and 800 bp downstream of the stop codon for the gene. The putative silencer binding activity was lowest in testicular germ cells, suggesting that low expression of this negative regulator allowed expression in testis [58]. Although the mouse MCP gene is expressed at very low levels in somatic tissues, its expression can be induced by viral infection [59]. Murine cytomegalovirus infection strongly induced MCP expression in fibroblasts and the region responsible was localized near the −750 bp position. Induction of the gene rendered the cells more resistant to complement mediated lysis, suggesting a mechanism for protection of viral infected cells from host complement [59].

In summary, these studies suggest that MCP gene regulation is a critical feature for both autoimmunity (as in aHUS) and infection. Given that many pathogens use MCP as a receptor for infection, changes in transcription could be important in pathogenesis. This is not to say that transcription is the only way in which MCP expression is controlled. There is evidence that sequences in the 3′ untranslated region control protein expression as well [60, 61].

Decay accelerating factor (DAF, CD55)

DAF is another protein that is critical for protection of autologous surfaces from complement attack [62]. This protein functions by causing the dissociation of convertase components and like MCP and DAF, it is also a ligand for pathogens including enteroviruses and uropathogenic E. coli [63]. There is an emerging literature on the role of this protein in malignancies, suggesting that expression of DAF has several functions in tumorigenesis including rescue from apoptosis, induction of neoangiogenesis and increased invasiveness, in addition to protection of the tumor from complement [64]. Surprisingly, recent studies have demonstrated a role for DAF in negative regulation of T cell immunity [65, 66].

The DAF protein does not have a transmembrane domain; rather it is anchored to the cell surface via glycosylphosphatidylinositol (GPI) linkage. Because of this, patients with somatic stem cell defects in GPI attachment lack surface expression of DAF, as well as CD59 (see below). This leads to a disease called paroxysmal nocturnal hemoglobinuria (PNH) in which red blood cells are exquisitely susceptible to complement lysis [67].

Early studies on the genetic sequences of the DAF promoter indicated that, like other ubiquitously expressed genes, this gene does not have a prototypical TATA box [68]. The majority of basal control was dependent upon sequences from −200 to −140 in the promoter regions, but this report offered no further characterization [68]. Deletion studies with reporter genes suggested that there is a strong negative regulatory element between 275 and 720 bp upstream of the transcription start site. Another study did not find any evidence of this negative regulatory element in DAF, although this result may be due to differences in the cell types used for transfection [69]. Thomas and Lublin identified two separate enhancer regions, but did not further define these sequences [69]. Despite the key role of the DAF protein in human disease, these reports are thus far the only studies on regulation of the human gene.

The murine DAF gene enhancer-promoter was recently characterized. Like the human gene, the mouse gene lacks TATA or CCAAT elements [70]. Three Sp1 sites were identified and characterized as being critical for both basal and LPS-induced transcriptional activation of the gene, and the two sites with the strongest homology to the human promoter sequence were the most critical for expression. A negative regulatory element was localized between 1104 bp and 610 bp upstream of the transcriptions start site, but was not further characterized [70]. In addition, a CREB site (also with homology to the human sequence) was important for gene expression. These studies demonstrate that there are several common regulatory elements that contribute to DAF gene expression.

Given the importance of DAF in the protection of cells from homologous lysis, further studies on DAF regulation are certainly in order. Studies in regions other than the 5′ region (such as in the first intron) would be critical. Emerging research shows that DAF is important in more biological processes than previously believed, such as tumor biology, T regulatory cell activity and in reproduction. As the molecular regulation of DAF expression is better understood, pharmacological targets might be identified.

CD59 (membrane inhibitor of reactive lysis)

Unlike the other protective complement receptors discussed above, CD59 is not part of the RCA cluster and does not possess typical SCR motifs. Instead, CD59 is a member of the Ly6 protein superfamily [71]. The protective action of CD59 is due to its ability to block insertion of C9 molecules into the growing membrane attack complex (MAC), thus preventing lysis of the host cell [72]. A number of new studies have demonstrated that CD59 has functions beyond those previously described including contributing to T cell and B cell signaling, control of NK cell activity and roles in paracrine and autocrine death of T cells [72]. Like Crry, DAF and MCP, CD59 has also been suggested to have a role in reproduction [73]. These studies suggest that understanding CD59 gene regulation may be important for understanding the diverse roles of the protein.

The first study examining the molecular control of CD59 gene regulation showed that the majority of basal transcriptional activity localized to a region between −35 and −70 [74]. This region was also critical for PMA-induced transcription. Sequence analysis showed that there were two Sp1 elements that were hypothesized to be responsible for this activity, although site-directed mutants were not presented [74]. Another study demonstrated that high levels of expression in certain cell lines could not be explained using only sequences from the region 5′ of the transcription start site [75]. Further analyses showed that specific enhancer activity could be localized to the first intron. Sequence analysis indicated binding sites for several ubiquitous transcription factors in this region, thus it was difficult to explain the cell-specific regulation observed based only upon these sequences [75]. This report showed that the regulation of the CD59 gene was more complex than first realized. A recent study showed that the CD59 promoter has putative p53 binding elements [76]. These sites were shown to bind p53 in vitro and knockdown of p53 levels using siRNA led to a six-fold decrease in CD59 expression. Futhermore, CD59 levels were correlated with p53 levels in cells that survived chemotherapy treatment [76].

These studies show that there are a number of key regulatory elements in the CD59 gene that may be exploited in order to modify gene expression. As more research clarifies the diverse roles of CD59, there will be other opportunities to investigate how gene regulation contributes to the biology of this protein.

Anaphylatoxin receptors

Unlike the protective receptors described above, the anaphylatoxin receptors directly contribute to the inflammatory properties of complement activation. When the C3 and C5 proteins are cleaved by their respective convertases, the small 70–80 amino acid subunits are freed and diffuse throughout the local area, where they can then bind to their respective receptors, C3aR and C5aR. The C3aR and C5aR proteins are seven transmembrane spanning receptors that are coupled to G-proteins for signaling. There is also a newly described C5a binding protein called C5L2 that does not associate with G-proteins and has thus been deemed a scavenger receptor for C5a; however there is no available information on the transcriptional regulation of this gene [77, 78]. Knockout studies have shown that the C3aR and C5aR genes are critical for a number of biological responses, thus further study on the molecular regulation of these genes was in order.

C3a receptor (C3aR)

Historically, C5aR has been more often studied than C3aR, mainly because the C5a protein has very high chemotactic activity relative to C3a. However, recent data have shown the importance of C3a–C3aR interactions in a number of biological responses. Studies in knockout mice have shown that C3aR was protective in asthma [79, 80], demonstrating that C3a directly contributes to this disease and further studies have also strongly suggested a role for C3a and C3aR in lung biology [81–83]. C3aR has also been shown to have a role in sepsis, although in this instance genetic deletion of C3aR enhanced endotoxin-mediated toxicity, suggesting an anti-inflammatory role [84]. Recent data have demonstrated a novel role for C3a in liver regeneration [85, 86]. Interestingly, experiments in C3aR knockout mice have demonstrated marked alteration in Th2 effector functions, possibly through modulation of antigen presenting cell effector functions [87, 88]. Collectively, these and other studies have shown that the C3aR protein is very important in a number of biological systems, yet until recently, very little was known about the molecular regulation of the C3aR gene.

Nearly all G-protein coupled receptors have a simple genetic structure with the promoter-enhancer region separated by a single intron from a single coding exon [89]. The first published report on C3aR gene regulation suggested a very compact regulatory element in macrophages and indicated that both the mouse and human gene lacked TATA boxes [90]. In an area conserved between the mouse and human in the region from −70 to −35, three regulatory sequences were found, including an AP-1 element, an Ets site, and a GATA sequence. A region of the 5′ promoter conferred full cell-type transcriptional specificity and the single intron did not have any detectable modulating activity [90]. The most critical elements of the promoter were the AP-1 and Ets sequences, which acted in a synergistic manner. Studies in primary astrocytes indicated that there was cell-specific usage of these elements, as the Ets element was more active than in macrophages [91]. This study also highlighted the importance of using primary cells to examine transcriptional activity.

The regulation of the human C3aR gene closely mirrors that of the mouse gene. A report conclusively showed that the human gene did not have a functional TATA sequence [92]. The human gene was also found to have functional AP-1 and Ets elements and this study demonstrated direct activity of the Ets-1 protein on transcription. Expression of dominant-negative Ets-1 in cells induced with dibutyryl-cAMP led to a failure to up regulate C3aR expression [92]. Although the authors showed synergistic activity of AP-1 and Ets, evidence suggested that Ets was acting indirectly. These studies demonstrate that C3aR gene regulation is very similar in mice and humans.

The recent publications on C3aR gene regulation are only the start of these analyses. There are a number of other cell types that express C3aR that have not received any detailed study. As demonstrated in astrocytes, it is preferable to do reporter gene assays in primary cells and recent advances in transfection technology are opening up the field for studies in these types of cells.

C5a receptor (C5aR, CD88)

A wide variety of cells express the receptor for C5a. It has been recognized for nearly 30 years that the predominant cell types that respond to C5a are myeloid cells; mainly macrophages, eosinophils, and neutrophils [93]. However, there are a number of other cells that express receptors for the C5a protein including some T cells, some B cells, fibroblasts, epithelial cells, endothelial cells, smooth muscle, mast cells, and dendritic cells [94]. The binding of C5a–C5aR on cells can result in a range of pleiotropic effects. One key function of C5a is increased vascular permeability through binding of C5a to endothelial cells and altering the expression of adhesion ligands for immune cells [95, 96]. A well-studied and critical function of C5a involves the chemoattraction of immune effector cells. C5a induces the chemotaxis of nearly all myeloid cell types including mast cells, granulocytes (especially neutrophils and eosinophils), monocytes, and macrophages [93]. In mast cells, macrophages, and eosinophils, the binding of C5a can result in the release of internal granule contents (reviewed in [97]). Binding of C5a to smooth muscle cell receptors can induce contraction [98]. Knockout mice have demonstrated critical roles for C5aR in arthritis [99, 100], liver regeneration [85, 101], and the reverse-arthus reaction [102]. Despite an extensive literature on the function of C5aR, very little was known about C5aR gene regulation until recently.

The first study on the regulation of C5aR was described when the gene was first cloned [103]. This study showed the basic genetic structure of the gene and delineated several potential transcription factor-binding sites (including a degenerate TATA element), but did not further evaluate their contribution to gene expression. Our group recently published more detailed studies. The 5′ region of the enhancer-promoter was shown to have cell-specific regulatory activity [104]. Transcriptional activity in both macrophages and endothelial cells was critically dependent upon a prototypical CCAAT box, but other sites in this region had little activity. The CCAAT box was also shown to be critical for LPS-induced C5aR activation [104]. Surprisingly, in both primary and transformed astrocytes, there was much less dependence on the CCAAT box site. Data suggested that there were other areas of the promoter-enhancer that had negative regulatory activity, but these sites have not been further defined [104].

Neither of these studies examined sequences in the single intron for their contribution to C5aR gene regulation. Since C5aR has such a ubiquitous expression pattern, there are a large number of other cell types in which C5aR expression could be studied. New research has shown that C5aR can contribute to T regulatory cell responses [105] and also plays a role in Toll-like receptor biology [106], thus it is important that we understand C5aR gene regulation in order to assess the role of C5a in diverse biological responses.

Conclusions

It is evident that many of the protective receptors (Crry, MCP, DAF, and CD59) have basic transcriptional machinery that directs expression to a wide variety of cells. A number of these genes lack TATA box elements and instead rely upon Sp1 sites that are important for both basal and induced gene expression. There are many recent studies demonstrating the importance of these proteins in diverse biological systems, indicating that it is critical that we gain a better understanding of gene regulation so that we might be able to modulate expression pharmacologically in a cell-specific manner. Similarly, the anaphylatoxin receptors are being recognized as having more functions than has been previously recognized. The C3aR gene has a very concise regulatory region that has at least two interacting sites, while the C5aR gene has what appears to be a basal regulatory arrangement that directs expression to a wide variety of cells. Only further study will indicate whether other regions of these genes are also critical in other cells, tissues or during gene induction mediated by stimulatory signals such as cytokines.