Molecular Biology and Evolution 19:718-727 (2002)
© 2002 Society for Molecular Biology and Evolution
Functional Diversification During Evolution of the Murine 
1-Proteinase Inhibitor Family: Role of the Hypervariable Reactive Center Loop
*Department of Biological Sciences, University of South Carolina;
Department of Cell and Molecular Biology, Roswell Park Cancer Institute
 |
Abstract |
|---|
 TOP Abstract IntroductionMaterials and MethodsResultsDiscussionAcknowledgementsReferences |
|---|
1-Proteinase inhibitor (1-PI) is a member of the serpin superfamily of serine proteinase inhibitors that are involved in the regulation of a number of proteolytic processes. 1-PI, like most serpins, functions by covalent binding to, and inhibition of, target proteinases. The interaction between 1-PI and its target is directed by the so-called reactive center loop (RCL), an 
20 residue domain that extends out from the body of the 1-PI polypeptide and determines the inhibitor's specificity. Mice express at least seven closely related 1-PI isoforms, encoded by a family of genes clustered at the Spi1 locus on chromosome 12. The amino acid sequence of the RCL region is hypervariable among 1-PIs, a phenomenon that has been attributed to high rates of evolution driven by positive Darwinian selection. This suggests that the various isoforms are functionally diverse. To test this notion, we have compared the proteinase specificities of individual 1-PIs from each of the two mouse species. As predicted from the positive Darwinian selection hypothesis, the various 1-PIs differ in their ability to form covalent complexes with serine proteinases, such as elastase, trypsin, chymotrypsin, and cathepsin G. In addition, they differ in their binding ability to proteinases in crude snake venoms. Importantly, the RCL region of the 1-PI polypeptide is the primary determinant of isoform-specific differences in proteinase recognition, indicating that hypervariability within this region drives the functional diversification of 1-PIs during evolution. The possible physiological benefits of 1-PI diversity are discussed. |
Introduction |
|---|
|
TOPAbstractIntroductionMaterials and MethodsResultsDiscussionAcknowledgementsReferences |
|---|
The serpins (serine proteinase inhibitors) comprise a superfamily of proteins that are found in a wide variety of animal and plant taxa and function in the control of serine proteinases. In mammals, serpins play roles in a diverse range of proteolytic processes, including blood clotting, fibrinolysis, inflammation, complement activation, and turnover of extracellular matrix (Potempa, Korzus, and Travis 1994
; Wright 1996
). Extensive chemical and physical studies have provided a rather comprehensive picture of the mechanisms by which serpins recognize and inhibit their target proteinases. A small domain within the serpin molecule, termed the reactive center loop (RCL), extends out from the body of the polypeptide and directs binding of the inhibitor to the target proteinase. The active site serine of the proteinase initiates cleavage of the peptide bond between the so-called P1 and P1' residues of the RCL,2
establishing an acyl bond linkage between the P1 carboxyl group of the serpin and the serine hydroxyl of the proteinase (Lawrence et al. 1995
; Olson et al. 1995
; Wilczynska et al. 1995
). After cleavage of the P1-P1' bond, the complex undergoes a conformational change that results in the insertion of the RCL, which remains covalently bound to the proteinase, into a ß-sheet within the body of the serpin polypeptide (Björk, Nordling, and Olson 1993
; Hopkins and Stone 1995
; Stratikos and Gettins 1997
; Huntington, Read, and Carrell 2000
). The energy for this conformational transition, which leads to prolonged inhibition of proteinase activity, is provided by the relaxation of the stressed configuration of the RCL. In general, the RCL can be divided into two functional regions: the hinge region at P14 to P5, whose amino acid sequence is relatively well conserved and controls conformational flexibility of the RCL domain, and the P5 to P1 region, which is hypervariable and governs proteinase specificity (Stein and Carrell 1995
; Wright 1996
; Zhou, Carrell, and Huntington 2001
). A number of studies have shown that the P1 amino acid is a major determinant of the target specificity of serpins. However, other amino acids within the RCL are also important for target recognition, as well as for the conformational changes that occur in association with cleavage of the P1-P1' bond (Potempa, Korzus, and Travis 1994; Wright 1996
).
1-Proteinase inhibitor (1-PI; also known as 1-antitrypsin) is a 50-to 55-kDa glycosylated plasma protein that is synthesized primarily in the liver and represents one of the more abundant serpins in the mammalian bloodstream. The central function of 1-PI in humans is the inhibition of neutrophil elastase (Potempa, Korzus, and Travis 1991; Wright 1996
). The importance of this inhibitory function is underscored by the existence of the Z allele of the 1-PI gene, which encodes a mutant protein that undergoes polymerization within the endoplasmic reticulum of liver cells and is therefore not secreted. This results in a severe deficiency of the inhibitor in the bloodstream, excessive degradation of elastin fibers in the lung, and eventually, pulmonary emphysema (Crystal 1989
; Le et al. 1992
; Lomas et al. 1992, 1993
).
The number of 1-PI genes differs among mammalian species. Primates, sheep, bovine, rat, and some mouse species contain a single gene, whereas guinea pigs, rabbits, gerbils, and other mouse species contain as many as five (Borriello and Krauter 1991
; Rheaume et al. 1994
; Goodwin, Baumann, and Berger 1996
; Goodwin, Barbour, and Berger 1997
). Phylogenetic considerations suggest that the small gene families in the latter species were formed by amplification events that took place at separate times during evolution. The 1-PIs in mice probably arose just before the rat-mouse split (i.e., about 15–25 MYA) and underwent changes in number during subsequent evolution of the Mus genus. This led to the emergence of variation in the size of the 1-PI family within and among murine species (Borriello and Krauter 1991
; Rheaume et al. 1994
; Goodwin, Baumann, and Berger 1996
; Goodwin, Barbour, and Berger 1997
).
The amino acid sequence of the RCL of 1-PIs, like that of other serpins, is highly divergent within and among species, particularly in the region between P6 and P1'. A number of studies have indicated that RCL hypervariability has been driven by positive Darwinian selection, whereby the high rates of amino acid substitution have resulted in adaptive functional alterations (Carrell, Pemberton, and Boswell 1987
; Hill and Hastie 1987
; Laskowski et al. 1987
; Borriello and Krauter 1991
; Inglis and Hill 1991
; Rheaume et al. 1994
; Goodwin, Baumann, and Berger 1996
; Goodwin, Barbour, and Berger 1997
). This predicts that the biochemical function, i.e., the target specificities, of the various 1-PIs are distinct. Recent evidence indicates that only two of the five 1-PIs in mice bind and inhibit elastase (Paterson and Moore 1996
). However, the role of the RCL in such a isoform-specific function remains to be addressed.
In the current study, we demonstrate that individual mouse 1-PI isoforms in two mouse species differ with regard to their ability to bind serine proteinases, indicating that they have distinct biochemical functions. Furthermore, we show that the RCL is a primary determinant of such differences, which are likely to be a consequence of positive selection acting upon this region of the 1-PI polypeptide. We discuss several possible physiological roles of 1-PI diversity.
|
Materials and Methods |
|---|
|
TOPAbstractIntroductionMaterials and MethodsResultsDiscussionAcknowledgementsReferences |
|---|
Expression of
1-PI mRNAsAll mice were obtained from the Jackson Laboratory (Bar Harbor, Me.). Total liver RNA (10 µg per sample) was analyzed by Northern blotting using standard techniques. Oligonucleotide probes (Integrated DNA Technologies, Inc.) corresponded to the RCL regions of individual
1-PI isoforms, including DOM-2 (5'-CATAGGAACGGCTTCAAAGA-3'), DOM-3 (5'-AACGGCTAGTAAGACTGTAG-3'), DOM-5 (5'-GTTTTTTGTCTATGCCCCCTATCTT-3'), and DOM-7 (5'-AGCTGCTTGTAAGACTGTGG-3'). The probes were end-labeled with [
-32P]-ATP in the presence of T4 polynucleotide kinase.
Constructs for Expression of 1-PI Isoforms
Complementary DNAs corresponding to 1-PIs in the C57BL/6J and BALB/cJ strains of Mus musculus domesticus and in the wild-derived species Mus saxicola were isolated from appropriate liver cDNA libraries (Rheaume et al. 1994
). All cDNAs, which were approximately 1.3 kb in length, spanned the full length of the 1-PI mRNA molecule, from the 5'-untranslated region to the poly(A) tail. Each was isolated as an EcoRI fragment and was subcloned into expression vector pSVSportI (GIBCO BRL) or pcDNAI (Invitrogen).
Chimeric 1-PIs containing the RCL region of one isoform in the place of the homologous region of another were generated using the M. saxicola cDNAs. The 1.3-kb cDNAs were subcloned as EcoRI fragments into a derivative of pBS-SK(-) from which the BamHI and SmaI restriction sites were deleted by mutagenesis. The 242-bp BamHI-SmaI fragment of the donor cDNA, which corresponds to the RCL region between residues 328 and 408, was cloned into the recipient cDNA from which the homologous BamHI-SmaI fragment had been removed. The resulting chimeric cDNA was cut out of pBS-SK(-) with EcoRI and subcloned into pSVSportI or pcDNAI.
Transfection of COS-1 Cells
The 1-PI–expressing plasmids were introduced into COS-1 cells by transient transfection. Cells cultured in 15 cm dishes in DMEM containing 10% heat-inactivated fetal calf serum were transfected by the DEAE-dextran method (Lopata, Cleveland, and Sollner-Webb 1984
). After a 24-h recovery period, the cells were washed three times with phosphate-buffered saline and incubated for an additional 24 h in serum-free DMEM containing 4.5 mg/ml glucose; the high glucose concentrations were required to maintain glycosylation of 1-PI polypeptide chains (Baumann and Jahreis 1983
). Conditioned medium containing the secreted 1-PI was collected, and the cell debris was removed by brief centrifugation.
Purification of Recombinant 1-PIs
The conditioned medium was diluted in two volumes H2O, adjusted to a pH of 7.8, and applied to a 1-ml DEAE-Sepharose column (Pharmacia) equilibrated in 50 mM Tris-HCl, pH 7.5. The column was washed thoroughly with 50 mM Tris-HCl, pH 7.5, followed by a brief wash with 10 ml of buffer containing 100 mM NaCl; bound 1-PI was eluted with buffer containing 150 mM NaCl. Fractions containing the 1-PI were pooled and loaded onto a 0.5 x 30 cm Superose 200 column (Pharmacia) equilibrated with 50 mM Tris-HCl, pH 7.8, 150 mM NaCl. The inhibitor eluted in a single peak at the expected molecular weight range of approximately 50 kDa. Accurate quantitation of 1-PI concentrations in pooled peak fractions was done by one- and two-dimensional polyacrylamide gel electrophoresis, together with known amounts of the purified plasma-derived inhibitor; Coomassie blue–stained proteins were quantitated by densitometry, and the results were confirmed by Western blotting. We estimate that the 1-PI concentrations in the unprocessed conditioned medium of transfected cells range between 1 and 10 µg/ml.
The relative abundance of recombinant 1-PIs, along with their degrees of size and charge microheterogeneity, were assessed by incubating cells in serum-free medium containing 100 µCi/ml [35S]methionine (>1,000 Ci/mmol; Amersham) and analyzing 10 µl aliquots of the medium by two-dimensional PAGE (1.5 x 120 x 160 mm; Hoeffer); radiolabeled 1-PIs were visualized by fluorography.
Immunodetection of free and proteinase-bound 1-PI was carried out by Western blotting, using a polyclonal sheep antiserum raised against purified mouse 1-PI as probe; the antiserum was provided by Dr. J. Gauldie (McMaster University, Hamilton, Ontario, Canada). Antigen was detected by enhanced chemiluminescence (Amersham).
Determination of 1-PI Binding to Serine Proteinases
Recombinant 1-PI isoforms were assayed for their ability to covalently bind each of several serine proteinases. Elastase binding was assayed in reaction mixtures containing the inhibitor (1–10 ng), 8 µl binding buffer (0.1 M Tris-HCl, pH 8.0, 0.5 M NaCl, 20 mM CaCl2), and 1 µl human neutrophil elastase (1–80 ng; Elastin Products, Orensville, Mo.). Mixtures were incubated at 37°C for 5 min and stopped by the addition of 2 µl 15 mM Na2EDTA and 100 ng N-MeOSuc-Ala-Ala-Pro-Val-chloromethylketone (Sigma). An equal volume of SDS sample buffer was added, and the mixtures were loaded onto 7.5% SDS PAGE gels. Free- and proteinase-bound 1-PIs were visualized by Western blotting, as described previously.
Binding of 1-PI to other serine proteinases (chymotrypsin, trypsin, and cathepsin G) was assayed in a similar fashion, except that the binding buffer was 50 mM Tris-HCl, pH 7.8, 150 mM NaCl, and 20 mM CaCl2. Reaction mixtures contained 12.5–200 ng bovine pancreatic -chymotrypsin (Type VII, TLCK treated; Sigma), 25–400 ng bovine pancreatic trypsin (Type XIII, TPCK treated; Sigma), or 15–240 ng human leukocyte cathepsin G (Sigma). Mixtures were incubated for 5 min at 37°C, SDS sample buffer was added, and samples were subjected to SDS PAGE and Western blotting to detect free- and proteinase-bound 1-PIs.
1-PI Binding to Snake Venom Proteinases
Lyophilized, unfractionated venoms were obtained from several snake species. Venoms from the Russell's viper (Vipera russelli) and Gabon viper (Bitis gabonica) were purchased from Sigma, whereas those from the Malayan pit viper (Agkistrodon rhodostoma), Rhombic night adder (Causus rhombeatus), sand viper (Vipera ammodytes), and Barba amarilla (Bothrops atrox) were obtained from the Miami Serpentarium Laboratories and were provided to us by Dr. L. Kress (Roswell Park Cancer Institute, Buffalo, N.Y.). To detect covalent complexes between 1-PIs and proteinases in the venoms, the venom samples were dissolved in 150 mM NaCl, 0.001 N HCl to a protein concentration of 50 mg/ml and immediately diluted 10-fold in 50 mM Tris-HCl, pH 7.8, 150 mM NaCl, 20 mM CaCl2. Aliquots containing 0.1–80 µg venom protein were used to measure covalent 1-PI binding to proteinases.
|
Results |
|---|
|
TOPAbstractIntroductionMaterials and MethodsResultsDiscussionAcknowledgementsReferences |
|---|
Number and Expression of
1-PI IsoformsPrevious analyses have focused on the number and sequence of
1-PI mRNAs in inbred strains of mice (species M. m. domesticus; Borriello and Krauter 1991
; Rheaume et al. 1994
; Goodwin, Baumann, and Berger 1996
; Goodwin, Barbour, and Berger 1997
; K. Barbour et al., unpublished data). A total of seven 1-PI isoforms, encoded by genes in the so-called Spi1 cluster on chromosome 12, have been identified (Borriello and Krauter 1991
; Rheaume et al. 1994
; Goodwin, Baumann, and Berger 1996
; Goodwin, Barbour, and Berger 1997
; K. Barbour et al., unpublished data). Interestingly, not all mice express all seven isoforms. The Spi1 cluster is polymorphic among inbred strains, with various strains containing three, four, or five 1-PI–encoding genes (Goodwin, Barbour, and Berger 1997
; K. Barbour et al., unpublished data). Another murine species (M. saxicola) also contains a polymorphic 1-PI gene family (Rheaume et al. 1994
; Goodwin, Baumann, and Berger 1996
; Goodwin, Barbour, and Berger 1997
). For the purposes of the current paper, the isoforms in M. m. domesticus are denoted DOM-1, -2, -3, etc., whereas those in M. saxicola are denoted SAX-1, -2, -3, etc. No allelic relationships among 1-PI genes in the two species are implied by this nomenclature.
Northern blot analysis of mouse liver RNA has indicated that the levels of most 1-PI transcripts are relatively constant during postnatal development and adult life; however, two mRNAs, i.e., those encoding DOM-5 and DOM-7 in M. m. domesticus, are sexually dimorphic. Figure 1
shows that the concentrations of the DOM-5–specific and DOM-7–specific mRNAs are higher in males than in females. Expression of mRNAs encoding the other isoforms is not significantly different between the sexes (fig. 1
and data not shown). Both the DOM-5 and DOM-7 mRNAs are induced in females after treatment with testosterone (data not shown), indicating that 1-PI expression is modulated by androgens.
|
Serine Proteinase Binding to
1-PIsTo determine if the biochemical function(s) of
1-PI family members are distinct, we have compared individual inhibitors in their ability to covalently bind serine proteinases. Because the 1-PIs in mouse plasma exist as sets of physicochemically overlapping isoforms that are similar in amino acid sequence and that are highly glycosylated, it was impractical to isolate individual isoforms directly from serum. Thus, we obtained them from cultured cells into which expression vectors containing appropriate 1-PI cDNAs had been introduced. We chose mammalian cells, as opposed to Escherichia coli or yeast for this purpose because they carry out complex-type posttranslational glycosylation and therefore, should generate 1-PI molecules that closely mimic the structure of the native inhibitor.
Two-dimensional gel electrophoresis indicated that 1-PI is a major secreted protein of transfected COS-1 cells. Figure 2
shows representative gels containing proteins secreted from cells expressing SAX-1, -3, or -4. Each 1-PI exists as a characteristic series of proteins that exhibit microheterogeneity with regard to molecular weight and isoelectric point. This microheterogeneity spans the ranges expected for native murine 1-PI. These proteins were not observed in control, mock-transfected cells (fig. 2
). Structural studies involving glycosidase digestion (data not shown) have indicated that the secreted 1-PIs contain, like the native inhibitor, as many as four N-linked oligosaccharide units. Similar results were obtained using 1-PI cDNAs encoding DOM-1, -2, -3, -4, and -6 (data not shown).
|
We analyzed the proteinase specificities of DOM-1, -2, -3, -4, and -6, as well as SAX-1, -3, and -4. The amino acid sequences of the RCL regions of these isoforms are shown in figure 3 . Secreted
1-PIs were partially purified from the conditioned medium of transfected cells (see Materials and Methods) and examined for their ability to enter into a stable, covalent complex with different serine proteinases. Preparations of individual 1-PIs were incubated with increasing amounts of the relevant proteinase, and the mixtures were subjected to Western blot analysis using antibody to mouse 1-PI as probe. The reaction conditions for optimal binding of 1-PI to proteinase were determined in separate experiments. Results for the M. m. domesticus 1-PIs are shown in figure 4
. The free inhibitor is indicated on the figure, as is the intact proteinase-inhibitor complex. Bands at a molecular size intermediate between that of the free inhibitor and the complex are a consequence of self-digestion of proteinase in the complex and are commonly seen in these types of analyses (Suzuki et al. 1991
). Such self-digestion was more prevalent with chymotrypsin than with the other proteinases (fig. 4
). DOM-1 and -2 both formed stable complexes with trypsin, chymotrypsin, and elastase, the molecular sizes of which were similar to that expected for a 1:1 complex between inhibitor and proteinase. DOM-4 had a distinct proteinase binding profile, in that it readily reacted with chymotrypsin, formed a small amount of complex with trypsin, and failed to react with elastase. Finally, DOM-3 and -6 formed complexes with trypsin and chymotrypsin but barely reacted with elastase. These results indicate that 1-PI family members differ in their abilities to recognize and bind individual serine proteinases.
|
|
A similar analysis was carried out with three inhibitors from M. saxicola (SAX-1, -3, and -4). As depicted in figure 5 , SAX-1 formed a complex with elastase, chymotrypsin, and cathepsin G but only barely reacted with trypsin. SAX-3 reacted well with chymotrypsin but did not detectably bind to trypsin, elastase, or cathepsin G. Finally, SAX-4 formed a small amount of complex with trypsin and cathepsin G and little or no complex with the other two proteinases. Thus, as observed with the
1-PIs of M. m. domesticus, the M. saxicola inhibitors differ with regard to their ability to bind proteinases, indicating that the occurrence of functional alterations among members of 1-PI families is found in divergent species.
|
It is interesting to note that the patterns of proteinase recognition did not correlate completely with the identity of the P1 amino acid. For example, DOM-3, -4, and -6 all have tyrosine at the P1 site (fig. 3 ; Goodwin, Baumann, and Berger 1996
); yet, DOM-4 differs from DOM-3 and -6 in its proteinase specificity (fig. 4
). Likewise, SAX-1 and -3 both have methionine at the P1 residue (fig. 3
; Goodwin, Baumann, and Berger 1996
); yet they differ in proteinase recognition (fig. 5
). These observations indicate that P1 residue is not the sole determinant of target proteinase specificity.
Recognition of Serine Proteinases in Snake Venoms
The experiments described in the preceding section made use of commercially available serine proteinases derived from a variety of sources. These may or may not represent the physiologically relevant targets in the healthy, intact mouse. It is of interest to consider proteinases from a more natural context, such as those in snake venom. It has been postulated that RCL hypervariability within serpin molecules broadens the spectrum of proteinases recognized and inhibited by 1-PIs, providing protection against exogenous proteinases introduced by predators or pathogens (or both; Hill and Hastie 1987
). Accordingly, we predicted that various 1-PI isoforms should differ in recognition of proteinases in snake venom, which represents a complex mixture of serine proteinases, as well as other classes of proteolytic enzymes. Specific 1-PI isoforms may bind and inhibit some venom proteinases, while being degraded by others.
To test this prediction, we measured the ability of individual 1-PIs to covalently bind proteinases in crude venoms from each of several snake species. Aliquots of the venoms were incubated with SAX-1, -3, or -4, and the reaction mixtures were subjected to Western blot analysis. As seen in figure 6
, the SAX-1 isoform entered into a covalent complex with one or more proteinase(s) in each of the two Vipera venoms and formed a distinct complex with B. gabonica venom. SAX-1 formed no detectable complex with C. rhombeatus venom and was degraded without appreciable complex formation by venoms from A. rhodostoma and B. atrox (fig. 6
). SAX-3 formed a complex with proteinase(s) in the venom of V. ammodytes but not V. russelli or B. gabonica (fig. 6
). Finally, SAX-4 bound proteinase(s) in venom from V. russelli but not V. ammodytes or B. gabonica (fig. 6
). We conclude that there are striking differences among the three isoforms with regard to recognition of snake venom proteinases. The identities of the venom proteinases that interact with the various isoforms have not been determined. That 1-PIs can be an inhibitor for some venom proteinases and a substrate for others indicates that the total level of proteinase activity at a given site of venom introduction is determined by the balance between substrate and inhibitory activities of the serpin.
|
Role of the RCL in Proteinase Specificity of
1-PIsGiven the findings described previously, an important question relates to the contribution, if any, of the hypervariable RCL in differential proteinase binding by various
1-PI isoforms. We addressed this issue by exchanging RCL regions among M. saxicola 1-PIs. A 242-bp RCL-containing region of the SAX-4 cDNA was inserted in place of the corresponding region of SAX-1. The resulting construct encodes a chimeric inhibitor, denoted SAX-1-4, which contains residues 328–408 of SAX-4 in place of the homologous region of SAX-1 (the RCL is located at residues 368–391). SAX-1-4 was tested for serine proteinase binding and like SAX-4, was unable to form a covalent complex with elastase (fig. 7
). Thus, introduction of the RCL of SAX-4 into SAX-1 removed elastase binding, indicating that this region (or, more accurately, the region between residues 328 and 408) is responsible for the differential ability of SAX-1 and SAX-4 to covalently bind elastase.
|
A second chimeric construct, representing the reciprocal of the first, was also tested. The 242-bp RCL-containing fragment of SAX-1 cDNA was introduced in place of the corresponding region of SAX-4, generating the construct SAX-4-1, which encodes a SAX-4 inhibitor containing the RCL region of SAX-1. This chimeric inhibitor, like SAX-1, bound elastase (fig. 7 ). Thus, the RCL region of SAX-1 confers elastase recognition to SAX-4, again implicating the RCL as a major determinant of the differences between SAX-1 and SAX-4.
A similar set of experiments was done to assess the role of the RCL in the functional differences between SAX-1 and SAX-3. The 242-bp RCL-containing fragment of SAX-1 was inserted into SAX-3, generating SAX-3-1; conversely, the RCL region of SAX-3 was inserted into SAX-1, generating SAX-1-3. As shown in figure 7 , SAX-3-1, like SAX-1, bound elastase, whereas SAX-1-3, like SAX-3, was inactive. Thus, similar to what was observed in comparing SAX-1 and SAX-4, differences in elastase binding between SAX-1 and SAX-3 are caused by the RCL region.
The differential abilities of SAX-1 and SAX-4 to bind either trypsin or chymotrypsin were also analyzed. SAX-1-4, like SAX-4, formed a complex with trypsin, yet failed to react with chymotrypsin (fig. 7
); conversely, SAX-4-1, like SAX-1, complexed with chymotrypsin, yet exhibited no reaction with trypsin (fig. 7
). These results show that, as with elastase binding, the differential abilities of SAX-1 and SAX-4 to bind trypsin and chymotrypsin are caused by the RCL-containing regions of the inhibitors. One or more amino acid substitutions in this region of the 1-PI polypeptide confers functional differences to the various 1-PI isoforms.
We observed that SAX-3-1 retained the ability to form a complex with chymotrypsin, whereas SAX-1-3 did not (fig. 7 ). This was rather unexpected because both SAX-1 and SAX-3 are able to bind this proteinase (fig. 5 ). The inability of SAX-1-3 to bind chymotrypsin indicates that an intramolecular incompatibility exists between the RCL of SAX-3 and one or more other regions within SAX-1, suggesting that sequences outside the RCL-containing region between residues 368 and 391 influence the ability of the inhibitor to recognize chymotrypsin. Further studies will be necessary in order to understand the mechanism(s) underlying this observation.
|
Discussion |
|---|
|
TOPAbstractIntroductionMaterials and MethodsResultsDiscussionAcknowledgementsReferences |
|---|
The serpin superfamily of proteins serves a broad array of functions throughout the animal, plant, and microbial kingdoms. Since the original suggestion that the RCL of serpins is under positive Darwinian selection (Hill et al. 1984
; Hill and Hastie 1987
), there has been a great deal of interest in these proteins as models of adaptive molecular evolution. The work described in the present report clearly indicates that individual 1-PI isoforms have distinct biochemical activities, i.e., they differ with regard to target proteinase specificities. Importantly, such differences are primarily caused by the region of the 1-PI polypeptide containing the RCL. These results provide biochemical evidence in favor of the hypothesis, proposed on the basis of mRNA and protein sequence data, that positive Darwinian selection has driven sequence divergence within the RCL of murine 1-PIs (Hill et al. 1984
; Carrell, Pemberton, and Boswell 1987
; Hill and Hastie 1987
; Laskowski et al. 1987
; Borriello and Krauter 1991
; Inglis and Hill 1991
; Rheaume et al. 1994
; Goodwin, Baumann, and Berger 1996
; Goodwin, Barbour, and Berger 1997
). This hypothesis predicts that such divergence should mediate changes in the function of the 1-PI polypeptide, which is exactly what is concluded from our experiments. That naturally occurring alterations in the RCL are responsible for functional diversification of 1-PIs is not completely unexpected because it is well known that the RCL is a critical determinant of the target specificity of serpins (Stein and Carrell 1995
; Wright 1996
; Zhou, Carrell, and Huntington 2001
). However, the current work represents the first direct demonstration that positive Darwinian selection is associated with changes in 1-PI function during mammalian evolution.
Functional diversification of serpins has been observed in organisms other than mice. Five serpins in wheat grain, classified into two subfamilies, have been shown to have distinct RCL sequences and proteinases specificities (Østergaard et al. 2000
). In the tobacco hornworm (Manduca sexta), the serpin-1 gene encodes a family of serpin isoforms with distinct RCL regions (Jiang et al. 1996
). These isoforms, which arise as a consequence of alternative splicing of a single gene, exhibit diverse inhibitory profiles toward mammalian, fungal, and microbial proteinases (Jiang and Kanost 1997
). It is likely that the presence of functionally diverse serpins, generated by a variety of mechanisms, is important in a wide range of organisms.
It is to be emphasized that the analytical techniques used in the present report provide a qualitative, rather than a quantitative, comparison of the various 1-PI isoforms. Extensive kinetic studies of serpin binding to proteinases, which are beyond the scope of the present work, will be required to obtain a more quantitative perspective. Regardless, the current results clearly indicate that murine 1-PIs exhibit functional differences that derive from variation within the RCL, providing strong support for the role of positive Darwinian selection in 1-PI evolution.
The physiological relevance of altered target specificities brought on by evolutionarily derived amino acid substitutions within the RCL is not known. The primary in vivo function of 1-PI, i.e., protection of the extracellular matrix against excessive proteolytic degradation, particularly in the lung, derives from its ability to inhibit neutrophil elastase (Potempa, Korzus, and Travis 1994
; Wright 1996
). However, there is evidence to suggest that the serpin plays one or more roles that are independent of elastase inhibition (Nagai et al. 1992
). Furthermore, several 1-PI isoforms lack a methionine residue at the P1 site and do not inhibit elastase (Suzuki et al. 1991
; see figs. 4 and 5 ). It seems clear, therefore, that altered proteinase specificities conferred by the rapidly evolving RCL must have unrecognized functions.
We suggest three possible physiological advantages provided by RCL divergence.
(1) Defense against exogenous parasites, predators, or pathogens: As has been discussed by others (Hill and Hastie 1987
; Borriello and Krauter 1991
; Goodwin, Baumann, and Berger 1996
), RCL hypervariability may confer more efficient defense against exogenous proteinases introduced by parasites, predators, or pathogens. This resistance could occur through changes in the spectrum of target proteinases recognized by the inhibitor, a notion consistent with the experiments presented in the current report. Alternatively, reductions in the sensitivity of the RCL to cleavage and inactivation by exogenous proteinases that are not targets for the inhibitor may arise. An interesting example of this is the opossum, which is resistant to rattlesnake venom as a consequence of production of a novel protein called oprin, which is a potent inhibitor of venom metalloproteinases. Furthermore, although the RCLs of mouse and human 1-PIs are particularly susceptible to cleavage by several snake venom metalloproteinases, the RCL of opossum 1-PI is resistant to such cleavage (Cantanese and Kress 1993
). Production of oprin and resistance of 1-PI to cleavage together probably contribute to the animal's rather remarkable ability to tolerate snake venom (Cantanese and Kress 1992
).
(2) Regulation of inflammation: Several studies indicate that 1-PI, like a number of hepatic plasma proteins that are induced during the acute phase response, acts as an anti-inflammatory agent (Tilg et al. 1993
; Tilg, Dinarello, and Mier 1997
). In vitro, 1-PI has been shown to inhibit superoxide production by human neutrophils (Bucurenci et al. 1992
) and protects cultured lung epithelial cells from the effects of endotoxins (Tumen et al. 1988
). Administration of 1-PI in vivo decreases bleomycin-induced fibrosis in the hamster lung (Nagai et al. 1992
). It is possible that the anti-inflammatory activities of 1-PI may be enhanced or broadened (or both) through diversity in proteinase recognition generated by RCL hypervariability, thereby providing a physiological benefit to the organism.
(3) Protection against carcinogenesis: Proteolytic enzymes are well-known mediators of tumorigenesis because of their ability to degrade and modify components of the extracellular matrix with which tumor cells interact (DeClerck et al. 1997
; Werb 1997
). Maspin, a serpin produced in normal mammary gland, suppresses the growth and metastasis of breast carcinoma cells (Zou et al. 1994
; Sager et al. 1997
). Deficiencies in matrilysin, a matrix metalloproteinase that efficiently cleaves the RCL of 1-PI in vitro (Sires et al. 1994
; Zhang et al. 1994
), results in significant reductions of the tumor burden in mice predisposed to intestinal cancer (Wilson et al. 1997
). It is possible that 1-PI is an antitumorigenic agent whose function is to protect the organism against the multitude of carcinogenic substances (e.g., plant secondary metabolites) encountered within its environment. Variation within the RCL may broaden the spectrum of tumor growth-promoting proteinases that are recognized by the inhibitor, thereby increasing the breadth or efficiency of its anticarcinogenic activity. Our unpublished observations indicating that cleavage and inactivation of 1-PI accompanies tumor development in mice (K. Barbour, unpublished data) is consistent with the notion that the serpin inhibits tumorigenesis.
|
Acknowledgements |
|---|
|
TOPAbstractIntroductionMaterials and MethodsResultsDiscussionAcknowledgementsReferences |
|---|
This work was supported by a grant DK33886 from the National Institutes of Health.
|
Footnotes |
|---|
Claudia Kappen, Reviewing Editor
1 Present address: Department of Developmental Biology and Anatomy, School of Medicine, University of South Carolina 
Address for correspondence and reprints: Franklin G. Berger, Department of Biological Sciences, University of South Carolina, Columbia, South Carolina 29208. berger{at}sc.edu
.
2 Nomenclature for amino acids within the RCL is according to Schechter and Berger (1967).
|
References |
|---|
|
TOPAbstractIntroductionMaterials and MethodsResultsDiscussionAcknowledgementsReferences |
|---|
Baumann H., G. P. Jahreis, 1983 Glucose starvation leads in rat hepatoma cells to partially N-glycosylated glycoproteins including alpha 1-acid glycoproteins. Identification by endoglycolytic digestions in polyacrylamide gels J. Biol. Chem 258:3942-3949
Björk I., K. Nordling, S. T. Olson, 1993 Immunologic evidence for insertion of the reactive-bond loop of antithrombin into the A beta-sheet of the inhibitor during trapping of target proteinases Biochemistry 32:6501-6506[Medline]
Borriello F., K. S. Krauter, 1991 Multiple murine alpha 1-protease inhibitor genes show unusual evolutionary divergence Proc. Natl. Acad. Sci. USA 88:9417-9421
Bucurenci N., D. R. Blake, K. Chidwick, P. G. Winyard, 1992 Inhibition of neutrophil superoxide production by human plasma alpha 1-antitrypsin FEBS Lett 300:21-24[Web of Science][Medline]
Cantanese J. J., L. F. Kress, 1992 Isolation from opossum serum of a metalloproteinase inhibitor homologous to human alpha 1B-glycoprotein Biochemistry 31:410-417[Medline]
———. 1993 Opossum serum alpha 1-proteinase inhibitor: purification, linear sequence, and resistance to inactivation by rattlesnake venom metalloproteinases Biochemistry 32:509-515[Medline]
Carrell R. W., P. A. Pemberton, D. R. Boswell, 1987 The serpins: evolution and adaptation in a family of protease inhibitors Cold Spring Harbor Symp. Quant. Biol 52:527-535
Crystal R. G., 1989 The alpha 1-antitrypsin gene and its deficiency states Trends Genet 5:411-417[Web of Science][Medline]
DeClerck Y. A., S. Imren, A. M. P. Montgomery, B. M. Mueller, R. A. Reisfield, W. E. Laug, 1997 Proteases and protease inhibitors in tumor progression Adv. Exp. Med. Biol 425:89-97[Web of Science][Medline]
Goodwin R. L., K. W. Barbour, F. G. Berger, 1997 Expression of the alpha 1-proteinase inhibitor gene family during evolution of the genus Mus Mol. Biol. Evol 14:420-428[Abstract]
Goodwin R. L., H. Baumann, F. G. Berger, 1996 Patterns of divergence during evolution of alpha 1-proteinase inhibitors in mammals Mol. Biol. Evol 13:346-358[Abstract]
Hill R. E., N. D. Hastie, 1987 Accelerated evolution in the reactive centre regions of serine protease inhibitors Nature 326:96-99[Medline]
Hill R. E., P. H. Shaw, P. A. Boyd, H. Baumann, N. D. Hastie, 1984 Plasma protease inhibitors in mouse and man: divergence within the reactive centre regions Nature 311:175-177[Medline]
Hood D. B., J. A. Huntington, P. G. W. Gettins, 1994 Alpha 1-proteinase inhibitor variant T345R. Influence of P14 residue on substrate and inhibitory pathways Biochemistry 33:8538-8547[Medline]
Hopkins P. C. R., S. R. Stone, 1995 The contribution of the conserved hinge region residues of alpha1-antitrypsin to its reaction with elastase Biochemistry 34:15872-15879[Medline]
Huntington J. A., R. J. Read, R. W. Carrell, 2000 Structure of a serpin-protease complex shows inhibition by deformation Nature 407:923-926[Medline]
Inglis J. D., R. E. Hill, 1991 The murine Spi-2 proteinase inhibitor locus: a multigene family with a hypervariable reactive site domain EMBO. J 10:255-261[Web of Science][Medline]
Jiang H., M. R. Kanost, 1997 Characterization and functional analysis of 12 naturally occurring reactive site vairants of serpin-1 from Manduca sexta J. Biol. Chem 272:1082-1087
Jiang H., Y. Wang, Y. Huang, A. B. Mulnix, J. Kadel, K. Cole, M. R. Kanost, 1996 Organization of serpin gene-1 from Manduca sexta. Evolution of a family of alternative exons encoding the reactive center loop J. Biol. Chem 271:28017-28023
Laskowski M. Jr.,, I. Kato, W. J. Kohr, S. J. Park, M. Tashiro, H. E. Whatley, 1987 Positive Darwinian selection in evolution of protein inhibitors of serine proteinases Cold Spring Harbor Symp. Quant. Biol 52:545-553
Lawrence D. A., D. Ginsburg, D. E. Day, M. B. Berkenpas, I. M. Verhamme, J.-O. Kvassman, J. D. Shore, 1995 Serpin-protease complexes are trapped as stable acyl-enzyme intermediates J. Biol. Chem 270:25309-25312
Le A., G. A. Ferrell, D. S. Dishon, Q.-Q. A. Le, R. N. Sifers, 1992 Soluble aggregates of the human PiZ alpha 1-antitrypsin variant are degraded within the endoplasmic reticulum by a mechanism sensitive to inhibitors of protein synthesis J. Biol. Chem 267:1072-1080
Lomas D. A., D. L. Evans, J. T. Finch, R. W. Carrell, 1992 The mechanism of Z alpha 1-antitrypsin accumulation in the liver Nature 357:605-607[Medline]
Lomas D. A., D. L. Evans, S. R. Stone, W.-Sx. Chang, R. W. Carrell, 1993 Effect of the Z mutation on the physical and inhibitory properties of alpha 1-antitrypsin Biochemistry 32:500-508[Medline]
Lopata M. A., D. W. Cleveland, H. Sollner-webb, 1984 High level transient expression of a chloramphenicol acetyl transferase gene by DEAE-dextran mediated DNA transfection coupled with a dimethyl sulfoxide or glycerol shock treatment Nucleic Acids Res 12:5707-5717
Nagai A., K. Aoshiba, Y. Ishihara, H. Inano, K. Sakamoto, E. Yamaguchi, J. Kagawa, T. Takizawa, 1992 Administration of alpha 1-proteinase inhibitor ameliorates bleomycin-induced pulmonary fibrosis in hamsters Am. Rev. Respir. Dis 145:651-656[Web of Science][Medline]
Olson S. T., P. E. Bock, J. Kvassman, J. D. Shore, D. A. Lawrence, D. Ginsburg, I. Björk, 1995 Role of the catalytic serine in the interactions of serine proteinases with protein inhibitors of the serpin family. Contribution of a covalent interaction to the binding energy of serpin-proteinase complexes J. Biol. Chem 270:30007-30017
Østergaard H., S. K. Rasmussen, T. H. Roberts, J. Hejgaard, 2000 Inhibitory serpins from wheat grain with reactive centers resembling glutamine-rich repeats of prolamin storage proteins: cloning and characterization of five major molecular forms J. Biol. Chem 275:33272-33279
Paterson T., S. Moore, 1996 The expression and characterization of five recombinant murine alpha 1-protease inhibitor proteins Biochem. Biophys. Res. Commun 219:64-69[Web of Science][Medline]
Potempa J., E. Korzus, J. Travis, 1994 The serpin superfamily of proteinase inhibitors: structure, function, and regulation J. Biol. Chem 269:15957-15960
Rheaume C., R. L. Goodwin, J. J. Latimer, H. Baumann, F. G. Berger, 1994 Evolution of murine alpha 1-proteinase inhibitors: gene amplification and reactive center divergence J. Mol. Evol 38:121-131[Web of Science][Medline]
Sager R., S. Sheng, P. Pemberton, M. J. C. Hendrix, 1997 Maspin. A tumor suppressing serpin Adv. Exp. Biol. Med 425:77-88[Web of Science][Medline]
Schechter I., A. Berger, 1967 On the size of the active site in proteases. I. Papain Biochem. Biophys. Res. Commun 27:157-62[Web of Science][Medline]
Sires U. I., G. Murphy, V. M. Baragi, C. J. Fliszar, H. G. Welgus, R. M. Senior, 1994 Matrilysin is much more efficient than other matrix metalloproteinases in the proteolytic inactivation of alpha 1-antitrypsin Biochem. Biophys. Res. Commun 204:613-620[Web of Science][Medline]
Song H. K., K. N. Lee, K.-S. Kwon, M.-H. Yu, S. W. Suh, 1995 Crystal structure of an uncleaved alpha 1-antitrypsin reveals the conformation of its inhibitory reactive loop FEBS Lett 377:150-154[Web of Science][Medline]
Stein P. E., R. W. Carrell, 1995 What do dysfunctional serpins tell us about molecular mobility and disease? Struct. Biol 2:96-113
Stratikos E., P. G. W. Gettins, 1997 Major proteinase movement upon stable serpin-proteinase complex formation Proc. Natl. Acad. Sci. USA 94:453-458
Suzuki Y., K. Yoshida, E. Honda, H. Sinohara, 1991 Molecular cloning and sequence analysis of cDNAs coding for guinea pig alpha 1-antiproteinases S and F and contrapsin J. Biol. Chem 266:928-932
Tilg H., C. A. Dinarello, J. W. Mier, 1997 IL-6 and APPs: anti-inflammatory and immunosuppressive mediators Immunol. Today 18:428-432[Web of Science][Medline]
Tilg H., E. Vannier, G. Vachino, C. A. Dinarello, J. W. Mier, 1993 Antiinflammatory properties of hepatic acute phase proteins: preferential induction of interleukin 1 (IL-1) receptor antagonist over IL-1 beta synthesis by human peripheral blood mononuclear cells J. Exp. Med 178:1629-1636
Tumen J., B. Meyrick, L. Berry, K. L. Brigham, 1988 Antiproteinases protect cultured lung endothelial cells from endotoxin injury J. Appl. Physiol 65:835-843
Werb Z., 1997 ECM and cell surface proteolysis: regulating cellular ecology Cell 91:439-442[Web of Science][Medline]
Wilczynska M., M. Fa, P. I. Ohlsson, T. Ny, 1995 The inhibition mechanism of serpins. Evidence that the mobile reactive center loop is cleaved in the native protease-inhibitor complex J. Biol. Chem 270:29652-29655
Wilson C. L., K. J. Heppner, P. A. Labosky, B. L. Hogan, L. M. Matrisian, 1997 Intestinal tumorigenesis is suppressed in mice lacking the metalloproteinase matrilysin Proc. Natl. Acad. Sci. USA 94:1402-1407
Wright H. T., 1996 The structural puzzle of how serpin serine proteinase inhibitors work Bioessays 18:453-464[Web of Science][Medline]
Zhang Z., P. G. Winyard, K. Chidwick, G. Murphy, M. Wardell, R. W. Carrell, D. R. Blake, 1994 Proteolysis of human native and oxidised alpha 1-proteinase inhibitor by matrilysin and stromelysin Biochim. Biophys. Acta 1199:224-228[Medline]
Zhou A., R. W. Carrell, J. A. Huntington, 2001 The serpin inhibitory mechanism is critically dependent on the length of the reactive center loop J. Biol. Chem 276:27541-27547
Zou Z., A. Anisowicz, M. J. C. Hendrix, A. Thor, M. Neveu, S. Sheng, K. Rafidi, E. Seftor, R. Sager, 1994 Maspin, a serpin with tumor-suppressing activity in human mammary epithelial cells Science 263:526-529
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  M. B. Padua, A. A. Kowalski, M. Y. Canas, and P. J. Hansen The molecular phylogeny of uterine serpins and its relationship to evolution of placentation FASEB J, February 1, 2010; 24(2): 526 - 537. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. M. Sevilla, R. Nachat, K. R. Groot, J. F. Klement, J. Uitto, P. Djian, A. Maatta, and F. M. Watt Mice deficient in involucrin, envoplakin, and periplakin have a defective epidermal barrier J. Cell Biol., December 31, 2007; 179(7): 1599 - 1612. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. Dai, K. Viswanathan, Y. M. Sun, X. Li, L. Y. Liu, B. Togonu-Bickersteth, J. Richardson, C. Macaulay, P. Nash, P. Turner, et al. Identification of Myxomaviral Serpin Reactive Site Loop Sequences That Regulate Innate Immune Responses J. Biol. Chem., March 24, 2006; 281(12): 8041 - 8050. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. J. Horvath, J. A. Irving, J. Rossjohn, R. H. Law, S. P. Bottomley, N. S. Quinsey, R. N. Pike, P. B. Coughlin, and J. C. Whisstock The Murine Orthologue of Human Antichymotrypsin: A STRUCTURAL PARADIGM FOR CLADE A3 SERPINS J. Biol. Chem., December 30, 2005; 280(52): 43168 - 43178. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 I. G. Winkler, J. Hendy, P. Coughlin, A. Horvath, and J.-P. Levesque Serine protease inhibitors serpina1 and serpina3 are down-regulated in bone marrow during hematopoietic progenitor mobilization J. Exp. Med., April 4, 2005; 201(7): 1077 - 1088. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 W. Li, A. Y. Savinov, D. V. Rozanov, V. S. Golubkov, H. Hedayat, T. I. Postnova, N. V. Golubkova, Y. Linli, S. Krajewski, and A. Y. Strongin Matrix Metalloproteinase-26 Is Associated with Estrogen-Dependent Malignancies and Targets {alpha}1-Antitrypsin Serpin Cancer Res., December 1, 2004; 64(23): 8657 - 8665. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||