MBE Advance Access originally published online on November 20, 2006
Molecular Biology and Evolution 2007 24(2):505-512; doi:10.1093/molbev/msl177
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Research Articles |
Disulfide-Bond Reshuffling in the Evolution of an Ape Placental Ribonuclease
Department of Ecology and Evolutionary Biology, University of Michigan
E-mail: jianzhi{at}umich.edu.
 |
Abstract |
|---|
 TOP Abstract IntroductionMaterials and MethodsResultsDiscussionSupplementary MaterialAcknowledgementsReferences |
|---|
Disulfide bonds play important roles in the folding and stability of proteins and are evolutionarily conserved. A classic example is RNase A (also known as bovine pancreatic ribonuclease), which contains 4 conserved disulfide bonds among 8 cysteines. However, human RNase 8, a paralog of RNase A uniquely expressed in the placenta, has lost one of the conserved cysteines but gained another, when compared with RNase 8 of various monkeys and with RNase A. We here show that both the loss and gain of the cysteines in human RNase 8 occurred in the common ancestor of African great apes (humans, chimps, and gorillas) 7–13 MYA. Computational predictions suggest changes of disulfide bonding by these cysteine substitutions. Site-directed mutagenesis indicates that if the ribonucleolytic activity is essential for RNase 8's function, the gain of the cysteine must have preceded the loss. Human RNase 8 represents one of the first examples in which the presumable evolutionary change of a disulfide bond involves 1 loss and 1 gain of cysteine, instead of 2 losses or 2 gains. Our results provide the foundation for detailed analysis toward understanding the impact of disulfide-bond reshuffling on the structure, function, and evolution of proteins in general and human RNase 8 in particular.
Key Words: ribonuclease • primate • disulfide bond
|
Introduction |
|---|
|
TOPAbstractIntroductionMaterials and MethodsResultsDiscussionSupplementary MaterialAcknowledgementsReferences |
|---|
A disulfide bond is a covalent bond formed between the sulfur atoms of 2 nonadjacent cysteine residues that are either in the same peptide or in different peptides. Disulfide bonds are important to the folding and stability of some proteins, usually those secreted to the extracellular medium. Because of their structural importance, disulfide bonds tend to be evolutionarily conserved (Thornton 1981
). In fact, cysteine is the second most conserved amino acid (after tryptophan) in protein evolution (Jones et al. 1992), and cysteines that form disulfide bonds are much more conserved than those that do not form disulfide bonds (Thornton 1981). In the event that a disulfide bond is not evolutionarily conserved, both cysteines are replaced by other amino acids (Thornton 1981; Kreisberg et al. 1995); very rarely does 1 of the 2 free cysteines remain, probably because free cysteines are potentially reactive (Thornton 1981; Kreisberg et al. 1995). The importance of cysteines and disulfide bonds to protein folding and stability is exemplified by RNase A in virtually every biochemistry textbook, after the pioneering work of the Nobel Laureate Christian Anfinsen (1973). RNase A is also known as bovine pancreatic ribonuclease. It contains 8 conserved cysteines, which form 4 disulfide bonds. Recent studies revealed details of the contribution of disulfide bonds to the folding, stability, and catalytic activity of RNase A (Laity et al. 1993; Klink et al. 2000; Ruoppolo et al. 2000). RNase A and its paralogous proteins form a large protein superfamily (Beintema and Kleineidam 1998). Analysis of the human genome sequence reveals the presence of 13 genes potentially belonging to the RNase A superfamily (Cho et al. 2005). Among them, RNases 1–8 are referred to as canonical RNases because they all share certain conserved sequence motifs (fig. 1) and have ribonucleolytic activities (Zhang, Dyer, et al. 2002), whereas RNases 9–13 lack some of these motifs and are not known to have ribonucleolytic activities (Castella et al. 2004; Cho et al. 2005; Cho and Zhang 2006). The majority of the canonical RNases of mammals, like RNase A, have 8 conserved cysteines (fig. 1). A notable exception is RNase 5 (also known as angiogenin), which has only 6 cysteines, forming 3 disulfide bonds (Strydom et al. 1985) (fig. 1). Recently, it was reported that human RNase 8 lost the sixth (C6) of the 8 conserved cysteines but gained another (fig. 1), when compared with other canonical RNases (Zhang, Dyer, et al. 2002). The newly gained cysteine is located between the fourth and fifth conserved cysteines and is hereby referred to as CX. A phylogenetic survey showed that the RNase 8 proteins from Old World and New World monkeys have the 8 conserved cysteines, whereas chimpanzees and gorillas share with humans the deviating pattern (Zhang, Dyer, et al. 2002). This finding indicates that the alterations of the cysteines in RNase 8 took place after the divergence of hominoids from Old World monkeys about 25 MYA, but before the separation of humans and chimps from gorillas about 7 MYA (Glazko and Nei 2003). The physiological function of RNase 8 is still unclear, although it has a ribonucleolytic activity and exclusive expression in the placenta (Zhang, Dyer, et al. 2002). In this work, we attempt to achieve 3 goals. First, we obtain a more precise date of the loss and gain of cysteines in primate RNase 8 evolution by sampling additional species. Second, we use computational modeling to investigate how the cysteine substitutions affect the disulfide bonding in RNase 8. Third, we examine the impact of the cysteine changes on the ribonucleolytic activity of the enzyme. Taking all together, we propose evolutionary scenarios of disulfide-bond reshuffling in primate RNase 8.
|
|
Materials and Methods |
|---|
|
TOPAbstractIntroductionMaterials and MethodsResultsDiscussionSupplementary MaterialAcknowledgementsReferences |
|---|
Comparison of Primate RNase 8 Sequences
The RNase 8 gene has only 1 coding exon, which was amplified from the genomic DNA of an orangutan (Pongo pygmaeus) by polymerase chain reaction (PCR) with primers 337 (5'-CTCCTAAGAGAGATGGCACCGGCC) and 338 (5'-CAAAGAGCAAGCCAGTCTGGAAACCTA). The PCR products were cloned into pCR II TA cloning vector (Invitrogen, San Diego, CA) and sequenced from both directions by the dideoxy chain termination method with an automatic sequencer. The DNA sequences of the RNase 8 genes of humans (Homo sapiens), chimpanzee (Pan troglodytes), gorilla (Gorilla gorilla), rhesus monkey (Macaca mulatta), pig-tailed macaque (Macaca nemestrina), baboon (Papio hamadryas), African green monkey (Cercopithecus aethiops), talapoin monkey (Miopithecus talapoin), tamarin (Saguinus oedipus), and owl monkey (Aotus trivirgatus) were originally reported in Zhang, Dyer, et al. (2002)
and were retrieved from GenBank. The 11 primate RNase 8 gene sequences were aligned using Clustal (Thompson et al. 1994).
Recombinant RNase 8 and Its Enzymatic Activity
The mature peptide region of the human and owl monkey RNase 8 gene was subcloned into the bacterial expression vector pFLAG CTS (Kodak, New Haven, CT) and was verified by DNA sequencing. The vector adds the octapeptide DYKDDDDK (FLAG) to the carboxy terminus of the recombinant protein, which facilitates its purification and detection with M2 anti-FLAG monoclonal antibody (Sigma, St. Louis, MO) without altering ribonucleolytic activity (Rosenberg and Dyer 1995). Recombinant proteins were isolated, purified, and quantified as described (Rosenberg and Dyer 1995). The ribonuclease activity of the recombinant proteins against a standard yeast tRNA substrate was measured in 40 mM sodium phosphate buffer (pH = 7.4) at 25 °C. Purified RNase (0.1–1 pmol) was added into 0.8 ml of the aforementioned buffer with 1.42 nmol tRNA. The reaction was stopped by 0.5 ml of 20 mM lanthanum nitrate with 3% perchloric acid, and insoluble tRNA was removed by centrifugation. The amount of solubilized tRNA was determined by ultraviolet absorbance at 260 nm. The catalytic activity of the RNase was determined as the picamolar of RNA digested per second per picamolar of RNase (Rosenberg and Dyer 1995; Zhang, Dyer, et al. 2002). The average values from 3 experiments and the standard deviation of the mean are presented.
Site-Directed Mutagenesis
The QuikChange site-directed mutagenesis kit of Stratagene (La Jolla, CA) was used to generate designed mutations in the human RNase 8 gene construct following manufacturer's instructions. The mutations were confirmed by DNA sequencing. Three mutants were made. The first mutant, referred to as intermediate A, contained Gly81Cys, meaning that Gly was replaced by Cys at position 81 of the mature peptide of human RNase 8. The second mutant, referred to as intermediate B, contained Cys66His. The third mutant, referred to as node P, contained both Gly81Cys and Cys66His.
Computational Prediction of Disulfide Bonds
The amino acid sequence of the mature peptide of RNase 8 (or its mutant forms) is used to predict the disulfide bonds through the DiANNA 1.1 Web server at http://clavius.bc.edu/
clotelab/DiANNA/ and DISULFIND Web server at http://disulfind.dsi.unifi.it/.
|
Results |
|---|
|
TOPAbstractIntroductionMaterials and MethodsResultsDiscussionSupplementary MaterialAcknowledgementsReferences |
|---|
Time of the Cysteine Substitutions in RNase 8
To date the loss of C6 and gain of CX in primate RNase 8 and to determine which event occurred first, we sequenced the RNase 8 gene from the orangutan P. pygmaeus and compared it with the available RNase 8 sequences from various primates. The comparison showed the presence of C6 and absence of CX in the orangutan as well as in all Old World and New World monkeys, but the absence of C6 and presence of CX in all African great apes (humans, chimps, and gorillas) (fig. 2). Thus, both the loss of C6 and gain of CX took place in the common ancestor of African great apes since its separation from the orangutan (bolded branch in fig. 3). African great apes and orangutans diverged around 13 MYA, and gorillas diverged from the common ancestor of humans and chimps about 7 MYA (Glazko and Nei 2003
). Hence, the loss and gain of the cysteines occurred in a short window between 7 and 13 MYA. Because there are no extant species that diverged from African great apes within the above time window, it is not possible to further narrow the time window of the cysteine substitutions. Nor is it possible to use phylogenetic information to infer the temporal order of the loss and gain of the cysteines. It has been reported that RNase 8 was independently pseudogenized multiple times in primate evolution (Zhang, Dyer, et al. 2002). Indeed, figure 2 shows that in 6 of the 11 primates, RNase 8 is likely a pseudogene because of the lack of an open reading frame or substitutions at catalytically or structurally important positions. However, the pseudogenizations do not affect our phylogenetic analysis and conclusion, because the nucleotide substitutions at C6 and CX can be reliably inferred even in pseudogenes (fig. 2). Interestingly, the gain of CX apparently required 2 nucleotide substitutions, one from C to T at the first-codon position and the other from A to G at the second-codon position (fig. 3). Given the orangutan codon of CGT at the CX position, it is likely that the substitution at the second-codon position preceded that at the first position. In other words, 2 amino acid substitutions, from His to Arg (in the common ancestor of great apes) and then to Cys (in the common ancestor of African great apes), took place at the CX position (fig. 3). The loss of C6, however, was due to a change of T to A or G at the first-codon position, which occurred in the common ancestor of African great apes (fig. 3).
|
|
Disulfide-Bond Reshuffling in RNase 8 Evolution: DiANNA Analysis
How did the loss of C6 and gain of CX affect the disulfide bonds in RNase 8? We used the computational tool DiANNA (Ferre and Clote 2005b
) to predict the disulfide bonds in RNase 8 based on its primary sequence. DiANNA prediction involves 3 steps. First, PSIPRED is run to predict the protein's secondary structure and PSIBlast is run against the nonredundant SwissProt to generate a multiple sequence alignment of the input sequence and its relatives. The predicted secondary structure and the alignment are used to train a neural network. Second, cysteine oxidation state is predicted and each pair of cysteines is assigned a likelihood of forming a disulfide bond through the diresidue neural network. Finally, Rothberg's implementation of Gabow's maximum weighted matching algorithm is applied to diresidue neural network scores to produce the final connectivity prediction (Ferre and Clote 2005b). There are 3 species (orangutan, talapoin, and owl monkey, see fig. 2) that have functional RNase 8 genes whose products contain the 8 conserved cysteines (i.e., with C6 but without CX). As expected, the DiANNA-predicted disulfide bonds of RNase 8 for each of these 3 species are identical to those of RNase A, with the 4 disulfide bonds being C1–C6, C2–C7, C3–C8, and C4–C5. Hence, it can be safely inferred that the disulfide bonds in the common ancestor of great apes (node P in fig. 3, before the cysteine substitutions) are also identical to those of RNase A (fig. 4).
|
We next predicted the disulfide bonds in human and chimp RNase 8 as the gorilla RNase 8 is a pseudogene (see fig. 2). We found that the disulfide bonds are identical between humans and chimps. Presumably, the common ancestor of African great apes (node Q in fig. 3) also had these disulfide bonds (fig. 4). Compared with node P, 3 disulfide bonds (C2–C7, C3–C8, and C4–C5) remain unchanged for node Q. However, C1–C6 is replaced by C1–CX in node Q due to the loss of C6 and gain of CX.
Because the temporal order of the loss of C6 and gain of CX is unknown, there are 2 possible evolutionary scenarios. In the first scenario, the gain of CX preceded the loss of C6. We predicted the disulfide bonds in a human RNase 8 mutant with both CX and C6, representing the intermediate state in this evolutionary scenario (see intermediate state A in fig. 4). This intermediate state has the same disulfide bonds as those in human and chimp RNase 8, with the extra C6 being a free cysteine. In the second scenario, the loss of C6 preceded the gain of CX. We predicted the disulfide bonds in a human RNase 8 mutant with neither CX nor C6, representing the intermediate state in the second evolutionary scenario (see Intermediate state B in fig. 4). Not unexpectedly, the ancestral C1–C6 is now broken owing to the loss of C6. As a result, C1 becomes a free cysteine.
Disulfide-Bond Reshuffling in RNase 8 Evolution: DISULFIND Analysis
Because computational predictions of disulfide bonds are not always correct, we tried 2 other commonly used prediction algorithms, DiPro (Cheng et al. 2006) and DISULFIND (Vullo and Frasconi 2004). DiPro uses machine-learning methods to predict whether a given protein chain contains intrachain disulfide bonds and uses recursive neural networks to predict the bonding probabilities of each pair of cysteines in the chain. However, we found that DiPro (http://contact.ics.uci.edu/bridge.html) predicted only 3 disulfide bonds in typical members of the RNase superfamily such as RNase A and RNase 2, in which the presence of 4 disulfide bonds have been experimentally demonstrated. Thus, DiPro is unreliable, at least for the RNases. DISULFIND predicts disulfide bonds using recursive neural networks and evolutionary information. We found that it correctly predicted the 4 conserved disulfide bonds for Old World and New World monkey RNase 8 as well as that of node P (fig. 5). Its prediction for human and chimp RNase 8, however, was slightly different from the DiANNA prediction. That is, DISULFIND predicted C1–C5 and C4–CX instead of C1–CX and C4–C5, resulting in 2 altered disulfide bonds in human RNase 8, compared with the 4 conserved disulfide bonds in canonical RNases (fig. 5). DISULFIND and DiANNA also made different predictions for intermediate states A and B (fig. 5). DISULFIND predicted that intermediate A and node P have the same disulfide bonds, indicating that the addition of CX in intermediate A did not alter disulfide bonding. For intermediate B, DISULFIND predicted C1–C3 with C8 being unpaired, whereas DiANNA predicted C3–C8 with C1 being unpaired.
|
Impact of the Cysteine Substitutions on the Ribonucleolytic Activity of RNase 8
It is interesting to know whether the loss of C6 and gain of CX in RNase 8 have affected its catalytic activity. We cloned the human RNase 8 gene into a bacterial expression vector, expressed the gene, purified the recombinant protein, and examined its ribonucleolytic activity against yeast tRNAs in a standard assay. Using site-directed mutagenesis, we also constructed 3 mutant forms of human RNase 8 and examined their activities. The first mutant has both C6 and CX, representing the intermediate state A in the first evolutionary scenario described above (figs. 4 and 5). The second mutant has neither C6 nor CX, representing the intermediate state B in the second evolutionary scenario (figs. 4 and 5). The third mutant has C6 but not CX, representing the ancestral protein at node P (figs. 4 and 5). We found that the mutant protein representing node P and the intermediate protein A both have slightly lower activities than that of human RNase 8 (fig. 6). The activity of the intermediate protein B is over 300-fold lower than that of human RNase 8 (fig. 6). In fact, our RNase assay cannot reject the null hypothesis of no activity in protein B (P > 0.1, t-test). If we trust the DiANNA prediction of disulfide bonds (fig. 4), the above results suggest that the C1–CX bond may substitute the C1–C6 bond without substantively reducing the RNase activity. However, removing the C1–C6 bond without adding a substitute leads to an almost complete loss of the RNase activity. If the DISULFIND prediction is correct, our results suggest that replacing C1–C6 and C4–C5 by C1–C5 and C4–CX does not change the RNase activity much, but the loss of C3–C8 and C1–C6 dramatically reduces the enzyme activity.
|
We also examined the ribonucleolytic activity of owl monkey RNase 8 and found it to be 7.4 times that of human RNase 8 (fig. 6). Although the human and owl monkey RNase 8 proteins differ at the CX and C6 positions, these differences are not the cause of their divergence in the ribonucleolytic activity because the mutant RNase 8 at node P has a catalytic activity comparable to that of human RNase 8 despite having the same residues at the 2 positions as owl monkey RNase 8 has.
|
Discussion |
|---|
|
TOPAbstractIntroductionMaterials and MethodsResultsDiscussionSupplementary MaterialAcknowledgementsReferences |
|---|
Disulfide bonds in proteins are evolutionarily conserved. Rarely does an existing bond disappear or a new bond appear during evolution. In the event when an existing disulfide bond is lost, both cysteines are replaced by other amino acids. Similarly, when a new disulfide bond emerges in evolution, 2 new cysteine residues are required. Primate RNase 8 is probably the first reported case where the change of a disulfide bond involves the loss of 1 cysteine and the gain of another. Our phylogenetic analysis showed that both the loss of C6 and gain of CX occurred in a relatively short evolutionary time between 7 and 13 MYA in the common ancestor of African great apes since its separation from orangutans. If the gain of CX preceded the loss of C6, the intermediate state A has 9 cysteines. Our RNase assay showed that A has a catalytic activity comparable to that of human RNase 8. Alternatively, if the loss of C6 preceded the gain of CX, the intermediate state B has only 7 cysteines and has virtually no catalytic activity. Thus, if the ribonucleolytic activity is important to the physiological function of RNase 8, the second evolutionary scenario cannot be true and the first must be true. However, it is currently unknown whether the physiological function of RNase 8 requires the ribonucleolytic activity, although it seems unlikely that a specific catalytic activity in a natural protein would be physiologically irrelevant.
We used 2 computational methods to predict disulfide bonds in RNase 8. DiANNA predicted that A has 4 disulfide bonds that are identical to those in human RNase 8, whereas B has only 3 disulfide bonds, with C1 being a free cysteine (fig. 4). DISULFIND, however, predicted that A has the 4 conserved disulfide bonds as in canonical RNases, whereas B has 2 conserved and a new disulfide bond, with C8 being a free cysteine. Although it is difficult to judge which computational prediction is more likely to be correct, we note that both are consistent with our knowledge of the RNase A superfamily in mammals because all canonical RNases with 4 disulfide bonds have appreciable levels of ribonucleolytic activity (Zhang, Dyer, et al. 2002), whereas RNase 5, which has only 3 disulfide bonds (lacking C4–C5), has virtually no RNase activity (Shapiro et al. 1986). Outside mammals, however, there are known RNases with 3 disulfide bonds but with relatively high RNase activities (Nitto et al. 2006). Interestingly, the double mutant of human RNase 8 that represents the ancestral protein at node P has a comparable ribonucleolytic activity with human RNase 8, suggesting that the 2 cysteine substitutions together did not substantively change the catalytic activity. However, here we only used yeast tRNAs as the substrate in the RNase assay. It is possible that the cysteine substitutions may have changed the catalytic activity of RNase 8 in degrading other RNA substrates.
The intermediate state A has the extra CX, in addition to the 8 conserved cysteines. DISULFIND predicted that it has the same disulfide bonds as the ancestral protein of node P. However, DiANNA predicted the breakage of the conserved C1–C6 bond and the formation of the new C1–CX bond. It is unclear why C1 tends to form a disulfide bond with CX rather than its original partner C6. Although a final set of disulfide bonds are predicted in DiANNA, it also generates scores for all possible disulfide bonds. For protein A, the score for the C1–CX pair (0.0119) is only slightly higher than that for C1–C6 (0.0104). Hence, in reality, some A molecules may form the C1–CX bond, whereas others form the C1–C6 bond. In fact, if we assume that human RNase 8 and RNase A share a similar 3-dimensional structure, the C1–CX bond may be difficult to form because of a long distance between the 2 cysteines. We stress that computational prediction of disulfide bonds, although relatively well developed, still generates errors (Vullo and Frasconi 2004; Ferre and Clote 2005a; Chen et al. 2006). Thus, our predictions here should be experimentally verified in the future.
The bovine pancreatic ribonuclease (i.e., the famous RNase A) gene was duplicated twice in the common ancestor of ruminants (Beintema and Kleineidam 1998; Breukelman et al. 2001; Zhang 2003). Interestingly, 1 resultant duplicate, named seminal RNase, gained 2 cysteines between C1 and C2. Because the 2 new cysteines are adjacent to each other in the primary sequence, they do not form a disulfide bond. Rather, they form 2 intermolecular disulfide bonds between 2 seminal RNase peptides, thus creating a homodimer (D'Alessio et al. 1972). It is possible that the altered cysteines make RNase 8 a homodimer, as in seminal RNase. Because DiANNA and DISULFIND do not predict intermolecular disulfide bonds, we conducted a Western analysis of recombinant human RNase 8, owl monkey RNase 8, and the 3 RNase 8 mutants in nonreducing conditions. The estimated molecular weights indicated that all these RNase 8 proteins are monomers (supplementary fig. S1, Supplementary Material online).
Human RNase 8 is uniquely expressed in the placenta, although its physiological function is unclear (Zhang, Dyer, et al. 2002). The closest relative to RNase 8 is RNase 7, and the gene duplication that gave rise to this gene pair likely occurred shortly before the divergence of Old World and New World monkeys within primates (Zhang, Dyer, et al. 2002; Cho and Zhang 2006). RNase 7 is expressed in several tissues including the kidney, spleen, heart, skeletal muscle, and skin, but not in the placenta (Harder and Schroder 2002; Zhang et al. 2003). Interestingly, although human RNase 7 has a potent antibacterial activity against multiple bacteria (Harder and Schroder 2002; Zhang et al. 2003), human RNase 8 lacks such an activity (Zhang, Dyer, et al. 2002; but see Rudolph et al. 2006). This functional dissimilarity is unsurprising because members of the RNase A superfamily are known to have a diverse array of biochemical activities and functions ranging from digestion to host-defense and angiogenesis (Barnard 1969; Fett et al. 1985; Rosenberg and Dyer 1995; Domachowske et al. 1998; Harder and Schroder 2002; Zhang J, Zhang YP, et al. 2002; Hooper et al. 2003; Zhang et al. 2003; Zhang 2006). Previous studies of RNase 3 (also known as the eosinophil cationic protein) showed that the key molecular determinant of the antibacterial activity is not the ribonucleolytic activity (Rosenberg 1995), but positively charged amino acid residues (Young et al. 1986; Zhang et al. 1998; Carreras et al. 2003). In fact, all RNases known to have antibacterial activities have a large number of positively charged amino acids (Zhang et al. 2003). Thus, we believe that the disulfide-bond reshuffling is unlikely the reason why human RNase 8 lacks the antibacterial activity.
An interesting phenomenon about primate RNase 8 is its independent pseudogenizations in multiple lineages, although there are no signs of duplication of the RNase 8 gene in any of the primates examined here and elsewhere (Zhang, Dyer, et al. 2002). Of the 11 primates examined, only 5 (humans, chimp, orangutan, talapoin, and owl monkey) appear to have functional RNase 8. However, these 5 species belong to all major lineages of higher primates (hominoids, Old World monkeys, and New World monkeys). This phylogenetic distribution suggests that at present the function of RNase 8 is needed only in a limited number of species, which nevertheless are scattered throughout higher primates. The RNase 8 pseudogenes identified all have a relatively intact open reading frame, suggesting that even in these lineages RNase 8 had been functional till relatively recently. It is intriguing to consider potential alterations of RNase 8 function by disulfide-bond reshuffling. Because RNase 8 is a paralog of RNase A, the classic model for studying the role of disulfide bonds in protein folding and stability, our findings provide the foundation for future detailed structural and functional analyses that may reveal the effect of disulfide-bond reshuffling on protein evolution.
|
Supplementary Material |
|---|
|
TOPAbstractIntroductionMaterials and MethodsResultsDiscussionSupplementary MaterialAcknowledgementsReferences |
|---|
Supplementary figure S1 is available at Molecular Biology and Evolution online (http://www.mbe.oxfordjournals.org/).
|
Acknowledgements |
|---|
|
TOPAbstractIntroductionMaterials and MethodsResultsDiscussionSupplementary MaterialAcknowledgementsReferences |
|---|
I thank Jaap Beintema, Soochin Cho, Wendy Grus, and 3 anonymous reviewers for their valuable comments. This work was supported by a research grant from the National Institutes of Health.
|
Footnotes |
|---|
Andriana Briscoe, Associate Editor
|
References |
|---|
|
TOPAbstractIntroductionMaterials and MethodsResultsDiscussionSupplementary MaterialAcknowledgementsReferences |
|---|
Anfinsen CB. (1973) Principles that govern the folding of protein chains. Science 181:223–230.
Barnard EA. (1969) Biological function of pancreatic ribonuclease. Nature 221:340–344.[CrossRef][Medline]
Beintema JJ and Kleineidam RG. (1998) The ribonuclease A superfamily: general discussion. Cell Mol Life Sci 54:825–832.[CrossRef][Web of Science][Medline]
Breukelman HJ, Jekel PA, Dubois JY, Mulder PP, Warmels HW, Beintema JJ. (2001) Secretory ribonucleases in the primitive ruminant chevrotain (Tragulus javanicus). Eur J Biochem 268:3890–3897.[Web of Science][Medline]
Carreras E, Boix E, Rosenberg HF, Cuchillo CM, Nogues MV. (2003) Both aromatic and cationic residues contribute to the membrane-lytic and bactericidal activity of eosinophil cationic protein. Biochemistry 42:6636–6644.[CrossRef][Medline]
Castella S, Benedetti H, de Llorens R, Dacheux JL, Dacheux F. (2004) Train A, an RNase A-like protein without RNase activity, is secreted and reabsorbed by the same epididymal cells under testicular control. Biol Reprod 71:1677–1687.
Chen BJ, Tsai CH, Chan CH, Kao CY. (2006) Disulfide connectivity prediction with 70% accuracy using two-level models. Proteins 64:246–252.[CrossRef][Web of Science][Medline]
Cheng J, Saigo H, Baldi P. (2006) Large-scale prediction of disulphide bridges using kernel methods, two-dimensional recursive neural networks, and weighted graph matching. Proteins 62:617–629.[CrossRef][Web of Science][Medline]
Cho S, Beintema JJ, Zhang J. (2005) The ribonuclease A superfamily of mammals and birds: identifying new members and tracing evolutionary histories. Genomics 85:208–220.[CrossRef][Web of Science][Medline]
Cho S and Zhang J. (2006) Ancient expansion of the ribonuclease A superfamily revealed by genomic analysis of placental and marsupial mammals. Gene 373:116–125.[CrossRef][Web of Science][Medline]
D'Alessio G, Parente A, Guida C, Leone E. (1972) Dimeric structure of seminal ribonuclease. FEBS Lett 27:285–288.[CrossRef][Web of Science][Medline]
Domachowske JB, Dyer KD, Bonville CA, Rosenberg HF. (1998) Recombinant human eosinophil-derived neurotoxin/RNase 2 functions as an effective antiviral agent against respiratory syncytial virus. J Infect Dis 177:1458–1464.[Web of Science][Medline]
Ferre F and Clote P. (2005a) Disulfide connectivity prediction using secondary structure information and diresidue frequencies. Bioinformatics 21:2336–2346.
Ferre F and Clote P. (2005b) DiANNA: a web server for disulfide connectivity prediction. Nucleic Acids Res 33:W230–W232.
Fett JW, Strydom DJ, Lobb RR, Alderman EM, Bethune JL, Riordan JF, Vallee BL. (1985) Isolation and characterization of angiogenin, an angiogenic protein from human carcinoma cells. Biochemistry 24:5480–5486.[CrossRef][Medline]
Glazko GV and Nei M. (2003) Estimation of divergence times for major lineages of primate species. Mol Biol Evol 20:424–434.
Harder J and Schroder JM. (2002) RNase 7, a novel innate immune defense antimicrobial protein of healthy human skin. J Biol Chem 277:46779–46784.
Hooper LV, Stappenbeck TS, Hong CV, Gordon JI. (2003) Angiogenins: a new class of microbicidal proteins involved in innate immunity. Nat Immunol 4:269–273.[CrossRef][Web of Science][Medline]
Jones DT, Taylor WR, Thornton JM. (1992) The rapid generation of mutation data matrices from protein sequences. Comput Appl Biosci 8:275–282.
Klink TA, Woycechowsky KJ, Taylor KM, Raines RT. (2000) Contribution of disulfide bonds to the conformational stability and catalytic activity of ribonuclease A. Eur J Biochem 267:566–572.[Web of Science][Medline]
Kreisberg R, Buchner V, Arad D. (1995) Paired natural cysteine mutation mapping: aid to constraining models of protein tertiary structure. Protein Sci 4:2405–2410.[Web of Science][Medline]
Laity JH, Shimotakahara S, Scheraga HA. (1993) Expression of wild-type and mutant bovine pancreatic ribonuclease A in Escherichia coli. Proc Natl Acad Sci USA 90:615–619.
Nitto T, Dyer KD, Czapiga M, Rosenberg HF. (2006) Evolution and function of leukocyte RNase A ribonucleases of the avian species, Gallus gallus. J Biol Chem 281:25622–25634.
Rosenberg HF. (1995) Recombinant human eosinophil cationic protein. Ribonuclease activity is not essential for cytotoxicity. J Biol Chem 270:7876–7881.
Rosenberg HF and Dyer KD. (1995) Eosinophil cationic protein and eosinophil-derived neurotoxin. Evolution of novel function in a primate ribonuclease gene family. J Biol Chem 270:21539–21544.
Rudolph B, Podschun R, Sahly H, Schubert S, Schroder JM, Harder J. (2006) Identification of RNase 8 as a novel human antimicrobial protein. Antimicrob Agents Chemother 50:3194–3196.
Ruoppolo M, Vinci F, Klink TA, Raines RT, Marino G. (2000) Contribution of individual disulfide bonds to the oxidative folding of ribonuclease A. Biochemistry 39:12033–12042.[CrossRef][Medline]
Shapiro R, Riordan JF, Vallee BL. (1986) Characteristic ribonucleolytic activity of human angiogenin. Biochemistry 25:3527–3532.[CrossRef][Medline]
Strydom DJ, Fett JW, Lobb RR, Alderman EM, Bethune JL, Riordan JF, Vallee BL. (1985) Amino acid sequence of human tumor derived angiogenin. Biochemistry 24:5486–5494.[CrossRef][Medline]
Thompson JD, Higgins DG, Gibson TJ. (1994) CLUSTAL W: improving the sensitivity of progressive multiple sequence alignment through sequence weighting, position-specific gap penalties and weight matrix choice. Nucleic Acids Res 22:4673–4680.
Thornton JM. (1981) Disulphide bridges in globular proteins. J Mol Biol 151:261–287.[CrossRef][Web of Science][Medline]
Vullo A and Frasconi P. (2004) Disulfide connectivity prediction using recursive neural networks and evolutionary information. Bioinformatics 20:653–659.
Young JD, Peterson CG, Venge P, Cohn ZA. (1986) Mechanism of membrane damage mediated by human eosinophil cationic protein. Nature 321:613–616.[CrossRef][Medline]
Zhang J. (2003) Parallel functional changes in the digestive RNases of ruminants and colobines by divergent amino acid substitutions. Mol Biol Evol 20:1310–1317.
Zhang J. (2006) Parallel adaptive origins of digestive RNases in Asian and African leaf monkeys. Nat Genet 38:819–823.[CrossRef][Web of Science][Medline]
Zhang J, Dyer KD, Rosenberg HF. (2002) RNase 8, a novel RNase A superfamily ribonuclease expressed uniquely in placenta. Nucleic Acids Res 30:1169–1175.
Zhang J, Dyer KD, Rosenberg HF. (2003) Human RNase 7: a new cationic ribonuclease of the RNase A superfamily. Nucleic Acids Res 31:602–607.
Zhang J, Rosenberg HF, Nei M. (1998) Positive Darwinian selection after gene duplication in primate ribonuclease genes. Proc Natl Acad Sci USA 95:3708–3713.
Zhang J, Zhang YP, Rosenberg HF. (2002) Adaptive evolution of a duplicated pancreatic ribonuclease gene in a leaf-eating monkey. Nat Genet 30:411–415.[CrossRef][Web of Science][Medline]
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  S. Cho and J. Zhang Zebrafish Ribonucleases Are Bactericidal: Implications for the Origin of the Vertebrate RNase A Superfamily Mol. Biol. Evol., May 1, 2007; 24(5): 1259 - 1268. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||