Gene duplication and the evolution of group II chaperonins: implications for structure and function

Chaperonins are multisubunit protein-folding assemblies. They are composed of two distinct structural classes, which also have a characteristic phylogenetic distribution. Group I chaperonins (called GroEL/cpn60/hsp60) are present in Bacteria and eukaryotic organelles while group II chaperonins are found in Archaea (called the thermosome or TF55) and the cytoplasm of eukaryotes (called CCT or TriC). Gene duplication has been an important force in the evolution of group II chaperonins: Archaea possess one, two, or three homologous chaperonin subunit-encoding genes, and eight distinct CCT gene families (paralogs) have been described in eukaryotes. Phylogenetic analyses indicate that while the duplications in archaeal chaperonin genes have occurred numerous times independently in a lineage-specific fashion, the eight different CCT subunits found in eukaryotes are the products of duplications that occurred early and very likely only once in the evolution of the eukaryotic nuclear genome. Analyses of CCT sequences from diverse eukaryotic species reveal that each of the CCT subunits possesses a suite of invariant subunit-specific amino acid residues ("signatures"). When mapped onto the crystal structure of the archaeal chaperonin from Thermoplasma acidophilum, these signatures are located in the apical, intermediate, and equatorial domains. Regions that were found to be variable in length and/or amino acid sequence were localized primarily to the exterior of the molecule and, significantly, to the extreme tip of the apical domain (the "helical protrusion"). In light of recent biochemical and electron microscopic data describing specific CCT-substrate interactions, our results have implications for the evolution of subunit-specific functions in CCT.     

Evidence of independent gene duplications during the evolution of archaeal and eukaryotic family B DNA polymerases

Eukaryotes and archaea both possess multiple genes coding for family B DNA polymerases. In animals and fungi, three family B DNA polymerases, alpha, delta, and epsilon, are responsible for replication of nuclear DNA. We used a PCR-based approach to amplify and sequence phylogenetically conserved regions of these three DNA polymerases from Giardia intestinalis and Trichomonas vaginalis, representatives of early-diverging eukaryotic lineages. Phylogenetic analysis of eukaryotic and archaeal paralogs suggests that the gene duplications that gave rise to the three replicative paralogs occurred before the divergence of the earliest eukaryotic lineages, and that all eukaryotes are likely to possess these paralogs. One eukaryotic paralog, epsilon, consistently branches within archaeal sequences to the exclusion of other eukaryotic paralogs, suggesting that an epsilon-like family B DNA polymerase was ancestral to both archaea and eukaryotes. Because crenarchaeote and euryarchaeote paralogs do not form monophyletic groups in phylogenetic analysis, it is possible that archaeal family B paralogs themselves evolved by a series of gene duplications independent of the gene duplications that gave rise to eukaryotic paralogs.      

Recurrent paralogy in the evolution of archaeal chaperonins.

Chaperonins are multisubunit double-ring complexes that mediate the folding of nascent proteins [1] [2]. In bacteria, chaperonins are homo-oligomeric and are composed of seven-membered rings. Eukaryotic and most archaeal chaperonin rings are eight-membered and exhibit varying degrees of hetero-oligomerism [3] [4]. We have cloned and sequenced seven new genes encoding chaperonin subunits from the crenarchaeotes Sulfolobus solfataricus, S. acidocaldarius, S. shibatae and Desulfurococcus mobilis. Although some archaeal genomes possess a single chaperonin gene, most have two. We describe a third chaperonin-encoding gene (TF55-gamma) from two Sulfolobus species; phylogenetic analyses indicate that the gene duplication producing TF55-gamma occurred within crenarchaeal evolution. The presence of TF55-gamma in Sulfolobus correlates with their unique nine-membered chaperonin rings. Duplicate genes (paralogs) for chaperonins within archaeal genomes very often resemble each other more than they resemble chaperonin genes from other archaea. Our phylogenetic analyses suggest multiple independent gene duplications - at least seven among the archaea examined. The persistence of paralogous genes for chaperonin subunits in multiple archaeal lineages may involve a process of co-evolution, where chaperonin subunit heterogeneity changes independently of selection on function.     

Novel Chaperonins in a Prokaryote

Group II chaperonins belong to the Hsp60 family occurring in archaea and eukaryotes. The archaeal chaperonins build the thermosome, which is similar to the eukaryotic CCT (chaperonin-containing TCP-1). Eukaryotes have eight subunits, and up until now, it was thought that archaea had between one and three subunits, depending on the species. We now report two novel subunits, termed Hsp60-4 and Hsp60-5, in the archaeon Methanosarcina acetivorans, which also has Hsp60-1, Hsp60-2, and Hsp60-3 with orthologs in Methanosarcinae. Hsp60-4 and Hsp60-5 occur only in M. acetivorans, which makes this organism unique in that it has the highest number of chaperonin subunits ever described for an archaeon. Evolutionary analysis suggests that either Hsp60-4 or Hsp60-5 paralogs have arisen by gene duplication with vastly increased accepted substitution rates or that they represent ancestral types found only in this species.Reviewing Editor: Dr. W. Ford Doolittle     

Evolution of DNA Replication Protein Complexes in Eukaryotes and Archaea

Background

The replication of DNA in Archaea and eukaryotes requires several ancillary complexes, including proliferating cell nuclear antigen (PCNA), replication factor C (RFC), and the minichromosome maintenance (MCM) complex. Bacterial DNA replication utilizes comparable proteins, but these are distantly related phylogenetically to their archaeal and eukaryotic counterparts at best.

Methodology/Principal Findings

While the structures of each of the complexes do not differ significantly between the archaeal and eukaryotic versions thereof, the evolutionary dynamic in the two cases does. The number of subunits in each complex is constant across all taxa. However, they vary subtly with regard to composition. In some taxa the subunits are all identical in sequence, while in others some are homologous rather than identical. In the case of eukaryotes, there is no phylogenetic variation in the makeup of each complex—all appear to derive from a common eukaryotic ancestor. This is not the case in Archaea, where the relationship between the subunits within each complex varies taxon-to-taxon. We have performed a detailed phylogenetic analysis of these relationships in order to better understand the gene duplications and divergences that gave rise to the homologous subunits in Archaea.

Conclusion/Significance

This domain level difference in evolution suggests that different forces have driven the evolution of DNA replication proteins in each of these two domains. In addition, the phylogenies of all three gene families support the distinctiveness of the proposed archaeal phylum Thaumarchaeota.     

All three chaperonin genes in the archaeon Haloferax volcanii are individually dispensable

The Hsp60 or chaperonin class of molecular chaperones is divided into two phylogenetic groups: group I, found in bacteria, mitochondria and chloroplasts, and group II, found in eukaryotic cytosol and archaea. Group I chaperonins are generally essential in bacteria, although when multiple copies are found one or more of these are dispensable. Eukaryotes contain eight genes for group II chaperonins, all of which are essential, and it has been shown that these proteins assemble into double-ring complexes with eightfold symmetry where all proteins occupy specific positions in the ring. In archaea, there are one, two or three genes for the group II chaperonins, but whether they are essential for growth is unknown. Here we describe a detailed genetic, structural and biochemical analysis of these proteins in the halophilic archaeon, Haloferax volcanii. This organism contains three genes for group II chaperonins, and we show that all are individually dispensable but at least one must be present for growth. Two of the three possible double mutants can be constructed, but only one of the three genes is capable of fully complementing the stress-dependent phenotypes that these double mutants show. The chaperonin complexes are made up of hetero-oligomers with eightfold symmetry, and the properties of the different combinations of subunits derived from the mutants are distinct. We conclude that, although they are more homologous to eukaryotic than prokaryotic chaperonins, archaeal chaperonins have some redundancy of function.     

Positive selection and subfunctionalization of duplicated CCT chaperonin subunits

To reach a functional and energetically stable conformation, many proteins need molecular helpers called chaperonins. Among the group II chaperonins, CCT proteins provide crucial machinery for the stabilization and proper folding of several proteins in the cytosol of eukaryotic cells through interactions that are subunit-specific and geometry-dependent. CCT proteins are made up of eight different subunits, all with similar sequences, positioned in a precise arrangement. Each subunit has been proposed to have a specialized function during the binding and folding of the CCT protein substrate. Here, we demonstrate that functional divergence occurred after several CCT duplication events due to the fixation of amino acid substitutions by positive selection. Sites critical for ATP binding and substrate binding were found to have undergone positive selection and functional divergence predominantly in subunits that bind tubulin but not actin. Furthermore, we show clear functional divergence between CCT subunits that bind the C-terminal domains of actin and tubulin and those that bind the N-terminal domains. Phylogenetic analyses could not resolve the deep relationships between most subunits, except for the groups alpha/beta/eta and delta/epsilon, suggesting several almost simultaneous ancient duplication events. Together, the results support the idea that, in contrast to homo-oligomeric chaperonins such as GroEL, the high divergence level between CCT subunits is the result of positive selection after each duplication event to provide a specialized role for each CCT subunit in the different steps of protein folding.     

Origin of eukaryotes from within archaea,archaeal eukaryome and bursts of gene gain: eukaryogenesis just made easier?

The origin of eukaryotes is a fundamental, forbidding evolutionary puzzle. Comparative genomic analysis clearly shows that the last eukaryotic common ancestor (LECA) possessed most of the signature complex features of modern eukaryotic cells, in particular the mitochondria, the endomembrane system including the nucleus, an advanced cytoskeleton and the ubiquitin network. Numerous duplications of ancestral genes, e.g. DNA polymerases, RNA polymerases and proteasome subunits, also can be traced back to the LECA. Thus, the LECA was not a primitive organism and its emergence must have resulted from extensive evolution towards cellular complexity. However, the scenario of eukaryogenesis, and in particular the relationship between endosymbiosis and the origin of eukaryotes, is far from being clear. Four recent developments provide new clues to the likely routes of eukaryogenesis. First, evolutionary reconstructions suggest complex ancestors for most of the major groups of archaea, with the subsequent evolution dominated by gene loss. Second, homologues of signature eukaryotic proteins, such as actin and tubulin that form the core of the cytoskeleton or the ubiquitin system, have been detected in diverse archaea. The discovery of this ‘dispersed eukaryome’ implies that the archaeal ancestor of eukaryotes was a complex cell that might have been capable of a primitive form of phagocytosis and thus conducive to endosymbiont capture. Third, phylogenomic analyses converge on the origin of most eukaryotic genes of archaeal descent from within the archaeal evolutionary tree, specifically, the TACK superphylum. Fourth, evidence has been presented that the origin of the major archaeal phyla involved massive acquisition of bacterial genes. Taken together, these findings make the symbiogenetic scenario for the origin of eukaryotes considerably more plausible and the origin of the organizational complexity of eukaryotic cells more readily explainable than they appeared until recently.     

The archaeal molecular chaperone machine: peculiarities and paradoxes.

A major finding within the field of archaea and molecular chaperones has been the demonstration that, while some species have the stress (heat-shock) gene hsp70(dnaK), others do not. This gene encodes Hsp70(DnaK), an essential molecular chaperone in bacteria and eukaryotes. Due to the physiological importance and the high degree of conservation of this protein, its absence in archaeal organisms has raised intriguing questions pertaining to the evolution of the chaperone machine as a whole and that of its components in particular, namely, Hsp70(DnaK), Hsp40(DnaJ), and GrpE. Another archaeal paradox is that the proteins coded by these genes are very similar to bacterial homologs, as if the genes had been received via lateral transfer from bacteria, whereas the upstream flanking regions have no bacterial markers, but instead have typical archaeal promoters, which are like those of eukaryotes. Furthermore, the chaperonin system in all archaea studied to the present, including those that possess a bacterial-like chaperone machine, is similar to that of the eukaryotic-cell cytosol. Thus, two chaperoning systems that are designed to interact with a compatible partner, e.g., the bacterial chaperone machine physiologically interacts with the bacterial but not with the eucaryal chaperonins, coexist in archaeal cells in spite of their apparent functional incompatibility. It is difficult to understand how these hybrid characteristics of the archaeal chaperoning system became established and work, if one bears in mind the classical ideas learned from studying bacteria and eukaryotes. No doubt, archaea are intriguing organisms that offer an opportunity to find novel molecules and mechanisms that will, most likely, enhance our understanding of the stress response and the protein folding and refolding processes in the three phylogenetic domains.     

Navigation inside a protease: substrate selection and product exit in the tricorn protease from Thermoplasma acidophilum

Chaperonins are multi-subunit double-ring complexes that mediate the folding of nascent or denatured proteins. Gene duplication has been a potent force in the evolution of chaperonins in Archaea. Here we show that gene conversion has also been an important factor. We utilized a novel maximum likehood-based phylogenetic method for scanning DNA sequence alignments for regions of anomalous phylogenetic signal, such as those affected by gene conversion. Our results suggest that in crenarchaeotes, where an ancient gene duplication producing alpha and beta subunits took place in the common ancestor of the Pyrodictium, Aeropyrum, Pyrobaculum and Sulfolobus lineages, multiple independent gene conversions have occurred between the alpha and beta genes independently in each of these groups. Significantly, the conversions have repeatedly homogenized the region of the gene encoding the substrate-binding domain. This suggests that while the alpha and beta subunits in crenarchaeotes share only 50-60% overall amino acid sequence identity, they do not possess distinct roles in the binding of substrate. Cryptic gene conversion between distantly related paralogs may be more common than is currently appreciated, and could be a significant factor in slowing the functional differentiation of proteins encoded by duplicate genes long after their duplication.     

Testing the neutral fixation of hetero-oligomerism in the archaeal chaperonin CCT

The evolutionary transition from homo-oligomerism to hetero-oligomerism in multimeric proteins and its contribution to function innovation and organism complexity remain to be investigated. Here, we undertake the challenge of contributing to this theoretical ground by investigating the hetero-oligomerism in the molecular chaperonin cytosolic chaperonin containing tailless complex polypeptide 1 (CCT) from archaea. CCT is amenable to this study because, in contrast to eukaryotic CCTs where sub-functionalization after gene duplication has been taken to completion, archaeal CCTs present no evidence for subunit functional specialization. Our analyses yield additional information to previous reports on archaeal CCT paralogy by identifying new duplication events. Analyses of selective constraints show that amino acid sites from 1 subunit have fixed slightly deleterious mutations at inter-subunit interfaces after gene duplication. These mutations have been followed by compensatory mutations in nearby regions of the same subunit and in the interface contact regions of its paralogous subunit. The strong selective constraints in these regions after speciation support the evolutionary entrapment of CCTs as hetero-oligomers. In addition, our results unveil different evolutionary dynamics depending on the degree of CCT hetero-oligomerism. Archaeal CCT protein complexes comprising 3 distinct classes of subunits present 2 evolutionary processes. First, slightly deleterious and compensatory mutations were fixed neutrally at inter-subunit regions. Second, sub-functionalization may have occurred at substrate-binding and adenosine triphosphate-binding regions after the 2nd gene duplication event took place. CCTs with 2 distinct types of subunits did not present evidence of sub-functionalization. Our results provide the 1st in silico evidence for the neutral fixation of hetero-oligomerism in archaeal CCTs and provide information on the evolution of hetero-oligomerism toward sub-functionalization in archaeal CCTs.     

MtGimC, a novel archaeal chaperone related to the eukaryotic chaperonin cofactor GimC/prefoldin

Group II chaperonins in the eukaryotic and archaeal cytosol assist in protein folding independently of the GroES-like cofactors of eubacterial group I chaperonins. Recently, the eukaryotic chaperonin was shown to cooperate with the hetero-oligomeric protein complex GimC (prefoldin) in folding actin and tubulins. Here we report the characterization of the first archaeal homologue of GimC, from Methanobacterium thermoautotrophicum. MtGimC is a hexamer of 87 kDa, consisting of two alpha and four beta subunits of high alpha-helical content that are predicted to contain extended coiled coils and represent two evolutionarily conserved classes of Gim subunits. Reconstitution experiments with MtGimC suggest that two subunits of the alpha class (archaeal Gimalpha and eukaryotic Gim2 and 5) form a dimer onto which four subunits of the beta class (archaeal Gimbeta and eukaryotic Gim1, 3, 4 and 6) assemble. MtGimalpha and beta can form hetero-complexes with yeast Gim subunits and MtGimbeta partially complements yeast strains lacking Gim1 and 4. MtGimC is a molecular chaperone capable of stabilizing a range of non-native proteins and releasing them for subsequent chaperonin-assisted folding. In light of the absence of Hsp70 chaperones in many archaea, GimC may fulfil an ATP-independent, Hsp70-like function in archaeal de novo protein folding.     

Identification of the TRiC/CCT substrate binding sites uncovers the function of subunit diversity in eukaryotic chaperonins

The ring-shaped hetero-oligomeric chaperonin TRiC/CCT uses ATP to fold a diverse subset of eukaryotic proteins. To define the basis of TRiC/CCT substrate recognition, we mapped the chaperonin interactions with the VHL tumor suppressor. VHL has two well-defined TRiC binding determinants. Each determinant contacts a specific subset of chaperonin subunits, indicating that TRiC paralogs exhibit distinct but overlapping specificities. The substrate binding site in these subunits localizes to a helical region in the apical domains that is structurally equivalent to that of bacterial chaperonins. Transferring the distal portion of helix 11 between TRiC subunits suffices to transfer specificity for a given substrate motif. We conclude that the architecture of the substrate binding domain is evolutionarily conserved among eukaryotic and bacterial chaperonins. The unique combination of specificity and plasticity in TRiC substrate binding may diversify the range of motifs recognized by this chaperonin and contribute to its unique ability to fold eukaryotic proteins.     

The deep archaeal roots of eukaryotes

The set of conserved eukaryotic protein-coding genes includes distinct subsets one of which appears to be most closely related to and, by inference, derived from archaea, whereas another one appears to be of bacterial, possibly, endosymbiotic origin. The "archaeal" genes of eukaryotes, primarily, encode components of information-processing systems, whereas the "bacterial" genes are predominantly operational. The precise nature of the archaeo-eukaryotic relationship remains uncertain, and it has been variously argued that eukaryotic informational genes evolved from the homologous genes of Euryarchaeota or Crenarchaeota (the major branches of extant archaea) or that the origin of eukaryotes lies outside the known diversity of archaea. We describe a comprehensive set of 355 eukaryotic genes of apparent archaeal origin identified through ortholog detection and phylogenetic analysis. Phylogenetic hypothesis testing using constrained trees, combined with a systematic search for shared derived characters in the form of homologous inserts in conserved proteins, indicate that, for the majority of these genes, the preferred tree topology is one with the eukaryotic branch placed outside the extant diversity of archaea although small subsets of genes show crenarchaeal and euryarchaeal affinities. Thus, the archaeal genes in eukaryotes appear to descend from a distinct, ancient, and otherwise uncharacterized archaeal lineage that acquired some euryarchaeal and crenarchaeal genes via early horizontal gene transfer.     

Coexistence of group I and group II chaperonins in the archaeon Methanosarcina mazei

Two distantly related classes of cylindrical chaperonin complexes assist in the folding of newly synthesized and stress-denatured proteins in an ATP-dependent manner. Group I chaperonins are thought to be restricted to the cytosol of bacteria and to mitochondria and chloroplasts, whereas the group II chaperonins are found in the archaeal and eukaryotic cytosol. Here we show that members of the archaeal genus Methanosarcina co-express both the complete group I (GroEL/GroES) and group II (thermosome/prefoldin) chaperonin systems in their cytosol. These mesophilic archaea have acquired between 20 and 35% of their genes by lateral gene transfer from bacteria. In Methanosarcina mazei G?1, both chaperonins are similarly abundant and are moderately induced under heat stress. The M. mazei GroEL/GroES proteins have the structural features of their bacterial counterparts. The thermosome contains three paralogous subunits, alpha, beta, and gamma, which assemble preferentially at a molar ratio of 2:1:1. As shown in vitro, the assembly reaction is dependent on ATP/Mg2+ or ADP/Mg2+ and the regulatory role of the beta subunit. The co-existence of both chaperonin systems in the same cellular compartment suggests the Methanosarcina species as useful model systems in studying the differential substrate specificity of the group I and II chaperonins and in elucidating how newly synthesized proteins are sorted from the ribosome to the proper chaperonin for folding.     

Estimation of phylogenetic inconsistencies in the three domains of life

Discrepancies in phylogenetic trees of bacteria and archaea are often explained as lateral gene transfer events. However, such discrepancies may also be due to phylogenetic artifacts or orthology assignment problems. A first step that may help to resolve this dilemma is to estimate the extent of phylogenetic inconsistencies in trees of prokaryotes in comparison with those of higher eukaryotes, where no lateral gene transfer is expected. To test this, we used 21 proteomes each of eukaryotes (mainly opisthokonts), proteobacteria, and archaea that spanned equivalent levels of genetic divergence. In each domain of life, we defined a set of putative orthologous sequences using a phylogenetic-based orthology protocol and, as a reference topology, we used a tree constructed with concatenated genes of each domain. Our results show, for most of the tests performed, that the magnitude of topological inconsistencies with respect to the reference tree was very similar in the trees of proteobacteria and eukaryotes. When clade support was taken into account, prokaryotes showed some more inconsistencies, but then all values were very low. Discrepancies were only consistently higher in archaea but, as shown by simulation analysis, this is likely due to the particular tree of the archaeal species used here being more difficult to reconstruct, whereas the trees of proteobacteria and eukaryotes were of similar difficulty. Although these results are based on a relatively small number of genes, it seems that phylogenetic reconstruction problems, including orthology assignment problems, have a similar overall effect over prokaryotic and eukaryotic trees based on single genes. Consequently, lateral gene transfer between distant prokaryotic species may have been more rare than previously thought, which opens the way to obtain the tree of life of bacterial and archaeal species using genomic data and the concatenation of adequate genes, in the same way as it is usually done in eukaryotes.     

真核生物SMC基因家族中拷贝数目的长期稳定进化

Members of the Structural Maintenance of Chromosome (SMC) family have long been of interest to molecular and evolutionary biologists for their role in chromosome structural dynamics, particularly sister chromatid cohesion, condensation, and DNA repair. SMC and related proteins are found in all major groups of living organisms and share a common structure of conserved N and C globular domains separated from the conserved hinge domain by long coiled-coil regions. In eukaryotes there are six paralogous proteins that form three heterodimeric pairs, whereas in prokaryotes there is only one SMC protein that homodimerizes. From recently completed genome sequences, we have identified SMC genes from 34 eukaryotes that have not been described in previous reports. Our phylogenetic analysis of these and previously identified SMC genes supports an origin for the vertebrate meiotic SMC1 in the most recent common ancestor since the divergence from invertebrate animals. Additionally, we have identified duplicate copies due to segmental duplications for some of the SMC paralogs in plants and yeast, mainly SMC2 and SMC6, and detected evidence that duplicates of other paralogs were lost, suggesting differential evolution for these genes. Our analysis indicates that the SMC paralogs have been stably maintained at very low copy numbers, even after segmental (genome-wide) duplications. It is possible that such low copy numbers might be selected during eukaryotic evolution, although other possibilities are not ruled out.     

Kinesin-related genes from diplomonad,sponge, amphioxus,and cyclostomes: divergence pattern of kinesin family and evolution of giardial membrane-bounded organella

To understand the question of whether divergence of eukaryotic genes by gene duplications and domain shufflings proceeded gradually or intermittently during evolution, we have cloned and sequenced Giardia lamblia cDNAs encoding kinesins and kinesin-related proteins and have obtained 13 kinesin-related cDNAs, some of which are likely homologs of vertebrate kinesins involved in vesicle transfer to ER, Golgi, and plasma membrane. A phylogenetic tree of the kinesin family revealed that most gene duplications that gave rise to different kinesin subfamilies with distinct functions have been completed before the earliest divergence of extant eukaryotes. This suggests that the complex endomembrane system has arisen very early in eukaryotic evolution, and the diminutive ER and Golgi apparatus recognized in the giardial cells, together with the absence of mitochondria, might be characters acquired secondarily during the evolution of parasitism. To understand the divergence pattern of the kinesin family in the lineage leading to vertebrates, seven more Unc104-related cDNAs have been cloned from sponge, amphioxus, hagfish, and lamprey. The divergence pattern of the animal Unc104/KIF1 subfamily is characterized by two active periods in gene duplication interrupted by a considerably long period of silence, instead of proceeding gradually: animals underwent extensive gene duplications before the parazoan-eumetazoan split. In the early evolution of vertebrates around the cyclostome-gnathostome split, further gene duplications occurred, by which a variety of genes with similar structures over the entire regions were generated. This pattern of divergence is similar to those of animal genes involved in cell-cell communication and developmental control.     

Evolution of the proteasome components

 A phylogenetic analysis of proteasome subunits revealed two major families (α and β) which originated by an ancient gene duplication prior to the divergence of archaebacteria and eukaryotes. Numerous gene duplications have subsequently occurred in eukaryotes; at least nine of these duplications were shown to have occurred prior to the divergence of animals and fungi. In mammals, two genes encoding proteasome subunits (LMP2 and LMP7) are located in the major histocompatibility complex (MHC) region and play a specific role in generation of peptides for presentation by class I MHC molecules. Phylogenetic analysis of LMP7 and related sequences from mammals and lower vertebrates indicated that this locus arose by gene duplication prior to the divergence of jawed and jawless vertebrates; the time of this duplication was estimated to have been about 600 million years ago. The evolutionary history of the proteasome subunits provides support for a model of the evolution of new gene function postulating that, after gene duplication, the proteins encoded by daughter loci can adapt to specialized functions previously performed by the product of a single generalized ancestral locus. Received: 19 August 1996 / Revised: 24 December 1996