Introduction

For all kinds of eukaryotic cells, actin is an essential protein that provides the basis for the cytoskeleton and motility. A characteristic feature of actin is its high conservativeness; actin in most organisms commonly displays greater than 85% amino acid sequence homology to mammalian actin. Certain protozoa possess unconventional actins, but their homology to mammalian actin is still generally greater than 60%. Such high homology between different actins may have been preserved because of the structural requirement imposed on actin: it must interact with various kinds of actin-binding proteins. In addition to actin, several kinds of actin-related proteins (Arps), with only 30–40% amino acid sequence homology to actin, are present in many, possibly all, kinds of eukaryotes, forming an actin superfamily together with actin (Goodson and Hawse 2002; Schafer and Schroer 1999). These Arps function in the regulation of cell motility and cytoskeletal organization, or in chromatin remodeling, rather than as the main component of the cytoskeleton.

The unicellular alga Chlamydomonas reinhardtii has two actin genes, one of which codes for a conventional actin (amino acid sequence identity with mammalian actin: ~90%) (Sugase et al. 1996), and the other a highly divergent actin called the novel actin-like protein (NAP) (amino acid identity with mammalian actin: 64%) (Kato-Minoura et al. 1998; Lee et al. 1997). Although actins with sequence homology as low as 60% have been found in protozoa, C. reinhardtii is unique in that it has both conventional and unconventional actins. The expression and properties of NAP have been studied in the C. reinhardtii mutant ida5 that lacks the conventional actin. This mutant was first isolated as a mutant lacking several species of flagellar inner arm dyneins and later identified as a null mutant of a conventional actin gene (Kato et al. 1993; Kato-Minoura et al. 1997). Apparently ida5 lacks those dyneins because actin is an essential subunit of inner-arm dyneins (Piperno and Luck 1979). Interestingly, the loss of actin in ida5 is compensated by enhanced expression of NAP, which is only negligibly expressed in stationary wild type cells (Kato-Minoura et al. 1998). Thus, the expression of actin and NAP should be regulated in an alternative manner.

Functional difference between NAP and conventional actin has not been fully understood. The fact that ida5 proliferates as rapidly as wild type indicates that NAP can substitute for actin in fundamental cellular functions. However, ida5 was found to be deficient in the assembly of some inner arm dyneins, as well as in the production of the fertilization tubule, a structure containing a core of F-actin bundles. Thus NAP is a partial substitute for conventional actin. Puzzling, however, is why the NAP gene is present at all in the wild-type cell in which it is barely expressed. NAP may be present because it plays an essential function in a certain phase in the wild-type cell cycle. The recent finding that a significant amount of NAP is expressed in wild-type cells after deflagellation suggests involvement of NAP in the flageilar assembly process in wild-type cells (Hirono et al. 2003). However, the elucidation of the possible specific roles played by NAP during flagellation must await further studies using specific mutants that lack NAP.

If NAP plays an essential role of its own in wild-type C. reinhardtii, then a similar unconventional actin may well be present in other species to play a similar role. We thus asked in this study whether the presence of conventional and unconventional actins in a single organism is a feature shared by other volvocalean algae. Our Southern blot analyses suggested that NAP homologues are in fact present in several Volvocales. From two species (Volvox carteri and Chlamydomonas moewusii) we cloned NAP-related genes and determined their sequences. A phylogenetic tree constructed using those and other actin genes showed that the NAP-related sequences in these three species form a distinct group derived from conventional actins.

Materials and Methods

Strains and Cell Culture

Strains used in this study are listed in Table 1. Three Chlamydomonas strains were grown in Tris–acetic acid–phosphate (TAP) medium (German and Levine 1965). Eudorina elegans was maintained in CA medium (Erata 2000). Volvox steinii and Gonium pectorale were maintained in VTAC medium (Erata 2000). Volvox carteri was cultured in Standard-Volvox-medium (SVM) under 16 h light/8 h dark conditions (Kirk and Kirk 1983).

Table 1 Volvocale strains used in this study
Full size table

Southern Hybridization

Genomic DNA was prepared as described in Weeks et al. (1986). Briefly, cell pellets were suspended in TEN (10 mM Tris–HCl (pH 8.0), 10 mM EDTA, 150 mM NaCl) and extracted with SDS-extraction buffer (SDS-EB, 2% SDS, 400 mM NaCl, 40 mM EDTA, 100 mM Tris–HCl (pH 8.0)). For multicellular algae that produce extracellular polysacchalides, SDS-CTAB extraction buffer (SDS-CTAB-EB, 2% SDS, 1.4 M NaCl, 20 mM EDTA, 100 mM Tris–HCl (pH 8.0), 2% CTAB, and 1% polyvinyl pyrrolidone) (Miller et al. 1993) was used instead of SDS-EB. The SDS extracts were further extracted with phenol–chloroform and then fractionated by ultracentrifugation through a cushion of CsCl. Blotting and hybridization were performed using the digoxigenin (DIG) system (Roche Diagnostics, Mannheim, Germany) as described by Kato-Minoura et al. (1997). Actin (Sugase et al. 1996) and NAP cDNA clones (Lee et al. 1997) were used as probes. To avoid cross-reaction between actin- and NAP-encoding genes, hybridization and washing were performed at a higher than standard temperature (67°C).

Genomic Library Screening

A genomic library of V. carteri, constructed in Dr. David Kirk’s laboratory, was kindly supplied by Dr. I. Nishii (Washington University). A C. moewusii genomic library was produced using a λFIX II vector kit (Stratagene, La Jolla, CA). Libraries were screened by hybridization with the protein-encoding region of C. reinhardtii NAP or the whole cDNA of actin. For probing and detection, the DIG system (Roche Diagnostics) was used. Positive phage DNA was isolated and digested with restriction enzymes. Positive fragments were then subcloned into pBluescript (Stratagene) and sequenced using an ABI PRISM 310 genetic analyzer (Applied Biosystems, Foster City, CA) at the National Institute for Basic Biology (NIBB) Center for Analytical Instruments (Okazaki, Japan).

Phylogenetic Analysis

Sequences of actin and actin-related proteins in the following organisms were retrieved from GenBank/EMBL/DDBJ databases: Acanthamoeba castelanii (V0002), Achlya bisexualis (X59936), Arabidopsis thaliana (U39480), Chlamydomonas reinhardtii actin (D50839), Chlamydomonas reinhardtii NAP (U68060), Chondrus crispus (U03676), Cryptomonas ovata (AF284836), Coleochaete scutata (AF061019), Cyanidioschyzon merolae (D32140), Cyanophora paradoxa (U90325), Drosophila melanogaster ARP53D (X78487), Emiliania huxleyi (S64188), Euglena gracilis (AF057161), Fucus distichus (U11697), Giardia lamblia (L29032), Homo sapiens (M20543), Mesostigma viride (AF061020), Nannochloris bacillaris (AB013098), Physarum polycephalum (X07792), Saccharomyces cerevisiae actin (L00026), Saccharomyces cerevisiae Arp1 (NC_001140, SGD:S0001171), Scherffelia dubia (AF061018), Solanum tuberosum (S20098), Toxoplasma gondii (U10429), and Volvox carteri (M33963). The deduced amino acid sequences were aligned using CLUSTAL W, version 1.8 (Thompson et al. 1994), made available at the GenomeNet service by Kyoto University Bioinformatics Center (http://clustalw.genome.ad.jp/ ). For the following analyses 243 amino acid sites without indels were used. For bootstrap analysis, 10,000 datasets were created using the SEQBOOT program (PHYLIP 3.57c [Felsenstein 1997]). For each dataset, a tree was inferred by the neighbor-joining (NJ) method, and the maximum-likelihood (ML) distance was calculated using the Jones–Taylor–Thornton (JTT) model (Jones et al. 1992) and the ProtML program (MOLPHY [Adachi and Hasegawa 1996]). A consensus tree was then constructed using the CONCENSE program of PHYLIP 3.57c. The synonymous/nonsynonymous substitution rates of actin and NAP were calculated by the method of Ina (1995), after the deduced cDNA sequences from the three species of Volvocales were aligned using CLUSTAL W.

Results

NAP-Related Sequences Are Present in Several Volvocalean Species

As stated in the Introduction, Chlamydomonas reinhardtii is unique among known lower eukaryotes in that it has genes encoding conventional and unconventional actins. Is the presence of two diverse kinds of actin in the same organism a feature shared by other green algae? We therefore examined several species of Volvocales for the presence of genes that hybridize with the coding region of the Chlamydomonas reinhardtii NAP (Cr NAP) gene in Southern blotting analysis. Chlamydomonas moewusii, Gonium pectorale, and Volvox carteri exhibited discrete bands that hybridized with the NAP probe, as well as bands that hybridized with the probe of Chlamydomonas reinhardtii conventional actin (Cr actin) (Fig. 1), The NAP-positive and actin-positive bands differed in size. These results suggest that those Volvocales have genes of unconventional actin as well as conventional actin, as in C. reinhardtii.

Figure 1
figure 1

C. moewusii, G. pectorale, and V. carteri each have an NAP-related sequence in addition to a conventional actin gene. Southern blot analyses of Volvocales actin and NAP related sequences. Genomic DNA was digested with either PstI (Ps) or PvuII (Pv), loaded on 1% agarose–TAE gels, and probed with coding regions of NAP or CrA cDNA.

Full size image

Sequence of the NAP-Positive Genes in C. moewusii and V. carteri

Screening of genomic libraries of C. moewusii and V. carteri yielded clones that hybridized with the Cr NAP probe (Figs. 2 and 3). The sequence from V. carteri is very similar to Cr NAP (80.7% amino acid identity). BLAST search indicated that no other actin sequences in the database share greater than 70% amino acid identity with this protein. Thus, we concluded that it is the V. carteri NAP (Vc NAP). Since V. carteri has an actin gene whose sequence is very similar to that of the C. reinhardtii conventional actin gene (Cresnar et al. 1990; Sugase et al. 1996), we can now consider that V. carteni is similar to C. reinhardtii in having both conventional and unconventional actin genes. In contrast, the sequence of the C. moewusii genomic clone hybridizing with the NAP probe was more similar to Cr actin (amino acid sequence identity: 72%) or to various mammalian actins (about 70%) than to Cr NAP (67%). However, its low similarity to conventional actin clearly indicates that this C. moewusii protein should be regarded as an unconventional actin. To see whether this unconventional actin is an extra actin present in addition to a conventional actin, like the NAP in C. reinhardtii, we screened the C. moewusii genomic library with the Cr actin probe. The screen yielded a single gene, whose partial sequence showed a strong homology (95.3% in amino acid identity) to Cr actin gene (see Figs. 2 and 3). Hence, C. moewusii must have at least one conventional actin gene and a single divergent actin gene that hybridizes with the Cr NAP probe. The latter divergent actin gene of C. moewusii will be referred to as Cm NAP.

Figure 2
figure 2

The nucleotide sequences of cPLA-N (A, T) and cPLA-N(O) (O) and their amino acid sequences: 1′ for PLA-N (A, T) and 3′ for PLA-N(O) (O). A, Amami-Oshima; T, Tokunoshima; O, Okinawa. An asterisk indicates the position of nonsynonymous nucleotide substitution between cPLA-N and cPLA-N(O).

Full size image
Figure 3
figure 3

The aligned amino acid sequences of T. flavoviridis (Tf) and T. gramineus (Tg) venom PLA2’s and other Crotalinae venom neurotoxic PLA2’s. A, Amami-Oshima: T, Tokunoshima; O, Okinawa. Tg PLA2’s are PLA-I (Oda et al. 1991), PLA-II and PLA-III (Fukagawa et al. 1993), and PLA-V (Nakai et al. 1995). Crotalinae neurotoxic PLA2’s are trimucrotoxin from T. mucrosquamatus (Tm), agkistrotoxin from Agkistrodon halys pallas (Ag), and crotoxin B from Crotalus durissus terrificus (Cd) (Bouchier et al. 1991). Bovine pancreatic (Bp) [Asp49]PLA2 (Fleer et al. 1978) is included for alignment.

Full size image

Despite the rather low similarity of Cm NAP to Cr NAP or Vc NAP, several amino acid residues in these proteins are commonly substituted from the conventional actin sequence causing charge alterations (Table 2). Interestingly, those substitutions have been rarely observed with other actins. On the other hand, the 36 amino acid residues that have been shown to be invariant in all known actins (Sheterline et al. 1996) were found to be preserved in all of the three NAP sequences.

Table 2 Charge-altering substitutions between NAP and conventional actin sequences commonly observed in the three species of volvocalean algae
Full size table

Intron Positions

Similarity between actin genes in different species can be assessed by the position of introns. The total intron number was as many as seven to nine in all of the six sequences examined in this study. The presence of many introns has also been observed in other genes of Volvocales. The conventional actin genes in the three algae are strongly conserved in intron positioning. Five of the six introns in the partial sequence of Cm actin occur at the same positions as in the introns in the Cr actin gene, while only the one between exon 8 and exon 9 occurs at different positions (310-1 and 340-2). In the three NAP sequences, C. reinhardtii and V. carteri exhibit strong similarity; six of the seven introns occur at the same positions, while only the last ones occur at different positions (290-1 and 314-2). However, Cm NAP totally differs in intron positions from the NAPs of the other two species. One of its introns (123-3) occurs at the same position as the one in the conventional actin gene occurring between exon 4 and exon 5. Thus, in terms of intron positioning, the Cm NAP is more similar to conventional actin than to NAP in the other two species. As described below, these features are correlated with the phylogenetic relationship of these algae.

Phylogenetic Analyses

A phylogenetic tree was constructed by ML analysis from the six actin/NAP sequences examined in this study, together with other actin sequences (Fig. 4). Using 243 amino acid sites in 28 sequences, with the S. cerevisiae Arp1 gene as an outgroup, calculation was performed as described in Materials and Methods. All of the conventional actins in the three volvocalean species were found to position on appropriate branches in the algal lineage. The sequence of the C. moewusii NAP that hybridized with the Cr NAP probe was indeed found to form a distinct branch with Cr and Vc NAPs with high bootstrap values. The phylogenic relationship among these three species inferred from the NAP sequence is consistent with the results obtained by analysis of other genes (Hepperle et al. 1998; Larson et al. 1992). In all three of these volvocalean species, the synonymous substitution rate (d S) was about the same (~0.5) for actin and NAP, whereas the nonsynonymous substitution rate (d N) in NAP was ~6 times higher than that in actin (Table 3).

Table 3 Rates of synonymous and nonsynonymous substitutions (d S and d N)
Full size table
Figure 4
figure 4

Phylogenetic tree based on partial amino acid sequences of actin and NAP. S. cerevisiae actin-related protein 1 (Arp1) is used as an outgroup. Bootstrap probabilities are shownon branches where available.

Full size image

Discussion

In this study we have shown that several species of volvocalean algae have genomic sequences that hybridize with the coding sequence of NAP, an unconventional actin of C. reinhardtii (Fig. 1). In addition to these species, Chlamydomonas augustae, Eudorina elegans, and Volvox steinii also reacted with the NAP probe under conditions wherein actin and NAP probes do not cross-react in hybridization using the C. reinhardtii genome (data not shown). Thus, it is likely that NAP-related sequences are present in these species also. Analyses of the NAP genes in C. reinhardtii, V. carteri, and C. moewusii (Figs. 2 and 3) revealed that the three NAP sequences form a distinct group, which we call the NAP family (Fig. 4). In addition, we have shown that C. moewusii also has a gene encoding a conventional actin. Together with data from previous studies (Cresnar et al. 1990; Sugase et al. 1996), the present study thus suggests that the simultaneous presence of both conventional and unconventional actins in a single organism is a feature common to various species of Volvocales. There is a possibility that some of the genes analyzed here are pseudogenes, since we did not confirm the existence of their mRNA. However, the fact that no stop codon insertion was detected within the sequenced 73% (Cm actin) and 91 % (Cm NAP and Vc NAP) of the putative coding sequences suggests that they are functional genes. Considering that Cr NAP is expressed only in actin-null mutants (Kato-Minoura et al. 1997) or in cells engaged in flagellation (Hirono et al. 2003), we expect that Cm NAP and Vc NAP may also be expressed in only small amounts under normal conditions.

In the actin superfamily, all the members of the conventional actin subfamily have similar numbers and sequences of amino acid residues; therefore, they are presumed to share characteristic functional properties of actin, such as the ability to polymerize into thin filaments or to interact with myosin ATPases. Members of the families Arp1, Arp2, and Arp3, in contrast, significantly differ from conventional actins in structure (Schafer and Schroer 1999); in particular, they have several inserted sequences at different positions, making their molecular weights 10–20% higher than that of conventional actin. Consistent with this structural dissimilarity, the functions hitherto identified with those Arps are totally different from those of conventional actin. Namely, Arp1 functions in a protein complex, called the dynactin complex, that functions in combination with cytoplasmic dynein and microtubules, while Arp2/3 complex acts as a nucleus of actin polymerization and regulates actin filament dynamics in various kinds of cells. NAP has the same number of amino acid residues as conventional actins and maintains the same amino acid residues that have been shown invariant in all known actins (Sheterline et al. 1996); thus, NAP appears to belong to the actin subfamily. However, the three NAP sequences show novel kinds of amino acid substitutions from the conventional actin (Table 2). These structural features explain the observed functional properties of NAP as a partial substitute for conventional actin. Whether NAP has a specific function is an important issue that remains to be studied.

The origin of this highly divergent actin is a puzzle. In cryptophyte algae, the presence of two highly divergent actins was discovered and interpreted as originating from a symbiont of a red alga (Stibitz et al. 2000). Since the evolution of Volvocales does not seem to have involved eukaryotic symbiosis (Cavalier-Smith 1992), we must think of other mechanisms. One possibility is that Volvocales acquired the NAP gene from another organism by some horizontal gene transfer process. In this regard it is interesting that the NAP family formed a clade with Giardia lamblia actin, although this result could only be an artifact, caused by long branch attraction. Another possibility is that two actin genes originally existed in primeval eukaryotes but that one of them was eliminated except in Volvocales. Two actins might be retained in Volvocales because they have distinct functions in the cell. The fact that the nonsynonymous substitution rate (d N) of NAP was much higher than that of actin suggests a much faster evolutionary rate of change for NAP. In other words, functional constraints on NAP should have been much smaller than those on actin. Despite such an apparently loose functional requirement, NAP probably has its own function beneficial for the survival of Volvocales, since the synonymous substitution rate (d S) is higher than d N (Table 3). Again, the specific function of NAP remains an important question to explore in future studies.

Our present study thus poses a question regarding the origin of the NAP family: Whether members of this subfamily are present also in other orders or classes awaits further studies.