Plant Vegetative and Animal Cytoplasmic Actins ShareFunctional Competence for Spatial Developmentwith Protists W OA

Muthugapatti K. Kandasamy, Elizabeth C. McKinney, Eileen Roy, and Richard B. Meagher1

Department of Genetics, Davison Life Sciences Complex, University of Georgia, Athens, Georgia 30602

Actin is an essential multifunctional protein encoded by two distinct ancient classes of genes in animals (cytoplasmic andmuscle) and plants (vegetative and reproductive). The prevailing view is that each class of actin variants is functionallydistinct. However, we propose that the vegetative plant and cytoplasmic animal variants have conserved functionalcompetence for spatial development inherited from an ancestral protist actin sequence. To test this idea, we ectopicallyexpressed animal and protist actins in Arabidopsis thaliana double vegetative actin mutants that are dramatically altered incell and organ morphologies. We found that expression of cytoplasmic actins from humans and even a highly divergentinvertebrate Ciona intestinalis qualitatively and quantitatively suppressed the root cell polarity and organ defects of act8 act7mutants and moderately suppressed the root-hairless phenotype of act2 act8 mutants. By contrast, human muscle actinswere unable to support prominently any aspect of plant development. Furthermore, actins from three protists representingChoanozoa, Archamoeba, and green algae efficiently suppressed all the phenotypes of both the plant mutants. Remarkably,these data imply that actin’s competence to carry out a complex suite of processes essential for multicellular developmentwas already fully developed in single-celled protists and evolved nonprogressively from protists to plants and animals.

INTRODUCTION

Actin is a highly conserved ubiquitous protein that is essentialfor several important cellular processes, including endocytosis,cell motility, cell division and cytokinesis, vesicle and organellemovement, cell signaling, and the establishment and mainte-nance of cell junctions, cell shape, and cell polarity (Staiger andBlanchoin, 2006; Pollard and Cooper, 2009). In animals andplants, actin is often encoded as multiple variants that are allvital for normal cell growth and organ development. Modernestimations of the tree of life built from thousands of sequencecomparisons indicate that the plant and animal kingdoms shareda common protist ancestor more than 1.5 billion years ago (Figure1A) and the deuterostome animals and vascular plants evolvedindependently ;800 and 600 million years ago, respectively(DeVries et al., 2006; Keeling, 2007; Zimmer et al., 2007). Duringthe evolution from protist to higher eukaryotes, the sequencesof actin variants have been relatively conserved, yet plants andanimals evolved distinct classes of actin variants, perhaps tosupport their independent multicellular ontogenies.

In both animals and plants, most actin genes are essential forspecific aspects of development as evident from genetic studiesof various mutants. For example, deficiencies in individual actingenes in mice or Arabidopsis thaliana result in embryo lethality,

low viability, and/or severe developmental defects within dif-ferent organs and cell types (Asmussen et al., 1998; Gillilandet al., 1998; Shawlot et al., 1998; Crawford et al., 2002; Liuet al., 2003; Sonnemann et al., 2006; Kandasamy et al., 2009;Meagher et al., 2011). In Arabidopsis, deficiency in one of thethree constitutively expressed actins, ACT7, severely impairsroot organ development and results in minor shoot defects,whereas lack of the other two constitutive actins, ACT2 andACT8, affects root hair cell tip growth (Kandasamy et al., 2009).Missense mutations in human actins are also associated witha variety of gene-specific developmental pathologies, includingskeletal and smooth muscle myopathies (Sparrow et al., 2003;Donnell et al., 2008), midline malformations (Procaccio et al.,2006), neutrophil dysfunction (Nunoi et al., 1999), and deafness(Zhu et al., 2003; Bryan et al., 2006).Based on their sequence phylogeny and expression pattern,

the multiple actin variants encoded in plants and animals areeach grouped into distinct ancient classes. Accordingly, thecytoplasmic and muscle actin variants in humans or mice formtwo discrete classes that are 6 to 7% divergent from each otherin their amino acid sequences (Figure 1B; see SupplementalTable 1 online). These two classes of animal actins arose earlyin the deuterostome animal lineage and are readily identifiedas distinct protein classes in the sea squirt, Ciona intestinalis,which shared a common ancestor with humans ;700 millionyears ago. Similarly, the vegetative and reproductive classes ofactin variants in vascular plants, such as maize (Zea mays) orArabidopsis, arose early in the vascular plant lineage, ;450million years ago, and are 7 to 8% divergent from each other inprotein sequence and 10 to 14% divergent from the deutero-stome animal sequences (Figure 1B; see Supplemental Table 1online). The protein sequences of these four separate classes of

1 Address correspondence to meagher@uga.edu.The author responsible for distribution of materials integral to the findingspresented in this article in accordance with the policy described in theInstructions for Authors (www.plantcell.org) is: Richard B. Meagher(meagher@uga.edu).WOnline version contains Web-only data.OAOpen Access articles can be viewed online without a subscription.www.plantcell.org/cgi/doi/10.1105/tpc.111.095281

The Plant Cell, Vol. 24: 2041–2057, May 2012, www.plantcell.org ã 2012 American Society of Plant Biologists. All rights reserved.

actin variants are distinct from each other within their phyla andhighly conserved within their lineages. Because the commonancestral actin sequence shared by plants and animals must befrom a very ancient single-celled protist (Figure 1A), a long-heldview is that the different conserved classes of plant and animalactin variants each have evolved distinct functions that are es-sential for cell, tissue, and organ development (Ponte et al., 1983;Hightower and Meagher, 1986; Meagher, 1995).

This view that these four classes of actins are all functionallydistinct is further supported by the failure of divergent classes ofplant and animal actin variants to efficiently suppress actin nullmutations in different actins within or across species (Karlssonet al., 1991; Kumar et al., 1997; Fyrberg et al., 1998; Brault et al.,1999; Schildmeyer et al., 2000; Kandasamy et al., 2007). Forexample, in Drosophila melanogaster, a null mutation in the 88Fgene encoding an adult flight muscle actin produces a flightlessphenotype, which can be rescued by a wild-type copy of 88F orthe other flight muscle actin 79B, but not by the fly larval muscleor cytoplasmic actins or by human cytoplasmic b-actin Hs-ACTB (Fyrberg et al., 1998; Brault et al., 1999). In mice, anacardiac-actin knockout results in lethality during either embryonicor prenatal development, with profound disorganization ofcardiac myofibrils (Kumar et al., 1997). Overexpression of gsmooth-actin only partially rescues the acardiac-actin knockout, and in

those few mice that survive to adulthood, their hearts are weakand enlarged, demonstrating that acardiac-actin and gsmooth-actinvariants make distinct contributions to muscle cell function.Budding yeast actin null mutations are only partially suppressedby chicken b-actin, which supports yeast cell viability, but causeslow growth, aberrant morphology, and temperature-sensitivelethality (Karlsson et al., 1991). By contrast, we found that neithervegetative nor reproductive actins from Arabidopsis could sup-press the inviability phenotype caused by the yeast actin nullmutation (Kandasamy et al., 2007). In addition, the ectopic ex-pression of Arabidopsis reproductive actin, At-ACT1, but notthe overexpression of a vegetative actin, At-ACT2, in Arabidopsisvegetative tissues causes extremely aberrant development oforgans and cell types with altered organization of F-actin struc-tures (Kandasamy et al., 2002). These aberrant phenotypes areentirely suppressed by the overexpression of reproductive, butnot vegetative, actin binding proteins (Kandasamy et al., 2007).Thus, it is reasonable to propose that the loss of appropriateinteractions with actin binding proteins may be at the heart ofthe functional differences among actin variants. With thispreponderance of evidence from genetic suppression studies,the simplest deduction is that the ancient classes of plant andanimal actin protein variants are functionally distinct.However, none of these genetic suppression studies ex-

amined the possible conservation of developmental functionsamong animal cytoplasmic and plant vegetative classes of actinsand their ancestral protist actins. Based on experimental dataand comparative genomics, the animal cytoplasmic and plantvegetative actins direct numerous related cellular functions thatare essential for the spatial development of cells, tissues, andorgans. These functions include the cytoplasmic activities ofmaintaining cell polarity, movement of all organelles in thecytoplasm, such as trafficking of vesicles for exo- and endo-cytosis and the positioning of membrane receptors, and thenuclear activities of moving various RNA polymerases, traf-ficking nuclear structures and macromolecules, and remodel-ing chromatin. Taken together, all such activities may compriseactin’s ability to form dynamic three-dimensional cellularstructures and maintain them over various time periods to in-fluence development, which we define here as actin’s com-petence for spatial development. Comparative genomicsdata and the morphology of extant protists (e.g., Choanozoa,Archamoeba, and green algae; Figure 1A) suggest that an-cestral protists closely related to animals and/or plants alsocontain similar sophisticated actin systems capable of sup-porting many of these cytoplasmic and nuclear processes(Anderson et al., 2005; Derelle et al., 2006; Merchant et al.,2007; King et al., 2008; Fritz-Laylin et al., 2010). Conse-quently, we considered an alternative hypothesis that cyto-plasmic animal and vegetative plant actins inherited theircompetence for spatial development from a common ancestralprotist actin. We provide strong initial support for this hy-pothesis by demonstrating that a subset of divergent actinsfrom animals and protists (Figure 1B; see Supplemental DataSet 1 online), which have not shared common ancestry withplant actins for more than a billion years, efficiently suppresseda diverse suite of cell and organ developmental defects ofArabidopsis mutants deficient in vegetative actins, while more

Figure 1. Phylogenetic Relationships among Various Eukaryotes andActins.

(A) An organismal tree of life comparing the four eukaryotic kingdoms:Fungi, animals, plants, and protists (Hedges et al., 2004; Keeling, 2007).The approximate divergence times (billions of years ago [BYA]) froma common ancestry are indicated in a scale. The points of origin ofdeuterostome animals and vascular plants are marked with arrows.(B) An actin tree comparing the Arabidopsis plant, human, and C. in-testinalis animal and protist protein sequences that were examined ordiscussed in detail in this study. The scale bar represents percent di-vergence in amino acid sequences.

2042 The Plant Cell

specialized muscle actins failed to support markedly any plantdevelopmental functions.

RESULTS

Plant Vegetative Actin Double Mutants Exhibit DistinctDevelopmental Defects

Arabidopsis encodes eight actin variants, three vegetative andfive reproductive. Among the three vegetative actins, ACT7 isexpressed strongly in young seedlings and developing organsand regulated by plant hormones, whereas ACT2 is constitu-tively abundant in adult plants and mature organs (Meagheret al., 1999; Kandasamy et al., 2001, 2009; Gilliland et al., 2003).These two variants are more prevalent in all plant organs thanACT8, which is the most weakly expressed actin in both youngand mature organs. In shoots, 10 to 15% of the total actin isACT8 compared with 45 to 50% ACT2 and 40 to 45% ACT7(Meagher et al., 1999; Kandasamy et al., 2009). These variantsexhibit functional specificity such that the lack of ACT7 mainlyaffects cell polarity (Gilliland et al., 2003) and organ growth asshown for roots in Figures 2A to 2C, while the lack of ACT2(Gilliland et al., 2002) or ACT8 (Figure 2A) affects only root hairgrowth. Due to partial functional redundancy, the double ho-mozygous null mutants often exhibit stronger and more distinctdevelopmental phenotypes than the single mutants. For exam-ple, double mutants lacking in ACT8 and ACT7 (act8-2 act7-4;Kandasamy et al., 2009) are highly dwarfed with strong root andshoot growth phenotypes (Figure 2), but they are able to surviveand set seeds. Double mutants lacking in ACT2 and ACT8(act2-1 act8-2) develop into relatively normal plants, but withbald roots having no root hairs (Kandasamy et al., 2009). Thetrichoblast cells of act2-1 act8-2 roots show no tip growth fromthe root hair bulges (i.e., no cell elongation). On the other hand,double mutants null for ACT2 and ACT7 (e.g., act2-1 act7-4) areextremely dwarfed and nearly infertile and are difficult to ma-nipulate in the laboratory (Gilliland et al., 2002; Kandasamy et al.,2009). Therefore, the act8-2 act7-4 and act2-1 act8-2 doublemutants are ideal for genetic suppression studies to examinethe specific roles of diverse actins on shoot and root organdevelopment and root hair cell tip growth, respectively. Themorphological and cellular phenotypes of act2-1 act7-4 andact2-1 act8-2 double mutant plants were fully characterizedpreviously (Kandasamy et al., 2009); hence, only the defects ofthe act8-2 act7-4 plants that are used throughout this article arepresented in Figures 2A to 2O and described here in detail.

The act8-2 act7-4 double mutant seedlings had dwarf rootsthat were <25% the length of the wild type (Figures 2A to 2C).The double mutant root phenotype was significantly more se-vere than in any of the single mutants such as the ACT7 mutant(act7-4) that had roots 35 to 40% the length of wild type or theACT8 mutant (act8-2) that had root length similar to wild type.The root growth defects in the double mutants became morepronounced as the plants grew, severely affecting the wholeroot architecture (Figure 2C). The act8-2 act7-4 roots had anelongation zone that was almost three times smaller (the regionfrom the root tip to the first root hair) than the wild type and had

a rough surface with irregularly bulged epidermal cells com-pared with the smooth surface of wild-type root apices (Figures2D and 2E). Approximately 8% of act8-2 act7-4 seeds containedmultiple, often fused, embryos or aberrant embryos that upongermination produced seedlings with three or four cotyledonsinstead of the usual two, a phenotype seldom observed in thewild type (Figure 2F). Because the act8-2 act7-4 plants haddefects in the growth and elongation of primary roots, we wereinterested in determining if these plants were also deficient inadventitious root initiation and growth. Therefore, we inducedroots from leaves by incubating them with the phytohormoneauxin. Following auxin treatment, the act8-2 act7-4 mutantleaves produced a few root primordia, but they were unable togrow into long adventitious roots as in the wild type (Figures 2Iand 2J). These results demonstrate that the act8-2 act7-4 plantswere defective in the growth of both primary and adventitiousroots.To understand the possible mechanism behind the retarded

root growth of act8-2 act7-4 plants, we examined the cellularorganization of root tips. The act8-2 act7-4 mutant root tips hadirregular masses of cells as a result of oblique cell wall formationduring cell division with an apparent disruption in cell polarity(Figure 2H), as shown previously for the act7-4 single mutant(Gilliland et al., 2003). In the wild-type root, on the other hand,the elongation zone contained regular files of cells as a result oftransversely aligned planes of cell division (Figure 2G). In addi-tion to the obvious root development phenotypes, the act8-2act7-4 plants revealed dwarfed shoot (Figure 2K) and in-florescence (Figures 2L and 2M) phenotypes with distinctlysmaller rosette leaves (40% narrower than the wild type; Figure2K) and flowers (30% smaller than the wild type; Figures 2N and2O).

Human Cytoplasmic b-Actin ACTB, but Not a-SkeletalMuscle Actin ACTA1, Restores All the DevelopmentalDefects of act8-2 act7-4 Plants

To determine whether deuterostome actin variants could sub-stitute for plant vegetative actin functions, we expressed humancytoplasmic b-actin ACTB and skeletal muscle a1-actin ACTA1in the act8-2 act7-4 mutant background under control of theconstitutively active At-ACT2 regulatory sequences (A2p) (seeMethods). We screened more than 15 independent trans-formants each expressing the A2p:HsACTB and A2p:HsACTA1constructs for the suppression of mutant phenotypes. For thatscreening, seven distinct developmental parameters were ex-amined: root length of seedlings, root architecture of youngplants, cellular organization at the root apex, adventitious rootdevelopment from leaf explants, organization of embryos in de-veloping seeds, plant/leaf morphology, and inflorescence/flowerarchitecture (Figures 3A to 3O, 4A, 4B, 5A, and 5B).The human cytoplasmic b-actin variant ACTB fully rescued

the root morphology defects of 3-d-old act8-2 act7-4 seedlings(Figure 3A); hence, the transformed seedlings had root lengthssimilar to the wild type (Figure 3B). The mature roots of 12- to14-d-old act8-2 act7-4/A2p:HsACTB plants had normal rootarchitecture with the length of the primary roots and the numberof lateral roots equivalent to the wild type (Figures 3C and 4A).

Ancient Functional Competence of Actin 2043

Figure 2. Phenotypic Characterization of the act8-2 act7-4 Double Mutant.

(A) to (C) Comparison of root growth of the act8-2 (8-2) and act7-4 (7-4) single mutants and the act8-2 act7-4 (8/7) double mutant and the wild type(WT).(A) Three-day-old seedlings. Note the shorter root hairs in the act8-2 seedling than the wild type.(B) Root length of 3-d-old seedlings. The error bars represent SD.(C) Eight-day-old seedlings.(D) and (E) Root tips. Ez, elongation zone.(F) Aberrant mutant seedlings with three cotyledons (middle) or fused embryos with multiple radicals (right). Wild-type seedling with two cotyledons isshown on the left.(G) and (H) DAPI-stained root tips showing loss of cell polarity in the 8/7 double mutant relative to the wild type.(I) and (J) Adventitious root growth from leaves.(K) Twenty-six-day-old plants.(L) and (M) Inflorescences.(N) and (O) Flowers.Bars = 50 mm (D), (E), (G), and (H), 2 mm in (I) and (J), and 1 mm in (L) to (O).

2044 The Plant Cell

Figure 3. Rescue of act8-2 act7-4 Double Mutant Phenotypes by Human Cytoplasmic ACTB and Skeletal Muscle ACTA1 Actin Variants.

Hence, lateral root primordia were initiated at the appropriateintervals. The root tips of the suppressed plants had cellularorganization and cell polarity comparable to the wild type withcells in regular rows (Figures 2G and 3D). Human cytoplasmicb-actin also restored normal morphology and size to shoots

(Figure 3F), leaves (Figures 5A and 5B), inflorescences (Figures3H to 3J), and flowers (Figures 3L to 3N). Moreover, we foundthe application of auxin induced adventitious roots from leavesof b-actin suppression lines with an efficiency similar to that ofthe wild type (Figures 3G and 4B). Almost 100% of transgenic

Figure 3. (continued).

(A)Morphology of 3-d-old seedlings comparing the wild type (WT), act8-2 act7-4 (8/7) mutant, and the mutant expressing human ACTB (HsB) or ACTA1(HsA1).(B) Root lengths of 3-d-old seedlings. Two independent transformed lines are shown for each construct.(C) Root architecture of 14-d-old plants.(D) and (E) DAPI DNA-stained root tips showing cellular organization (cf. to Figures 2G and 2H).(F) Twenty-five-day-old plants.(G) Adventitious root growth from leaves.(H) to (K) Inflorescences.(L) to (O) Flowers.Bars = 50 mm in (D) and (E) and 1 mm in (H) to (O).

Figure 4. Suppression of Root Architecture and Adventitious Root Growth Phenotypes by Diverse Actins.

(A) Root length (white, left bar in each pair) versus number of lateral roots (black, right bar in each pair) produced from 14-d-old seedlings of wild-type(WT), act8-2 act7-4 double mutant (8/7), and various transgenic lines expressing diverse actins.(B) Adventitious root length (white) versus number of roots (black) produced from the leaf explants after 4 d of auxin treatment followed by 6 d of growthon MS medium. The data represent average values from three biological replicates with each experiment containing 10 samples for each line shown.The error bars represent SD. HsB, Hs-ACTB; HsG1, Hs-ACTG1; HsA1, Hs-ACTA1; HsA2, Hs-ACTA2 (human); CiB, Ci-ACTB1 (C. intestinalis); Mb, Mb-ACT (M. brevicollis); Ac, Ac-ACT (A. castellanii ); Cs, Cs-ACT (C. scutata).

2046 The Plant Cell

seeds also contained normal embryos with the usual two coty-ledons and one radical. We obtained similar results for all 15 in-dependent transgenic Hs-ACTB–expressing plant lines examined.On the other hand, expression of human a-skeletal muscle actintransgene A2p:HsACTA1 in act8-2 act7-4 plants failed to rescuequantitatively any of the seven developmental phenotypes weexamined, and the transformed plants resembled untransformedmutant plants in nearly all aspects (Figures 3 and 4). None of the15 independent transgenic Hs-ACT1 suppression lines we gen-erated showed restoration of any normal phenotypes, except forthe shoots, especially the size of leaves, which appeared some-what improved in most lines compared with the double mutant(Figures 3F, 5A, and 5B). Mutant suppression data are summa-rized in Table 1.

Immunoblot analysis with the monoclonal antibody MAbGEa,which detects both human cytoplasmic Hs-ACTB andmuscle Hs-ACTA1 actins (Figure 6A), revealed that all A2p:HsACTB and A2p:HsACTA1 transformed mutant plants examined expressed highlevels of human cytoplasmic b-actin or skeletal muscle a1-actin,respectively. The levels of cytoplasmic b-actin or muscle a1-actinin the transgenic plants ranged from being equivalent to wild-typeplant actin levels to up to threefold above wild-type levels. Actinexpression for two such lines for each construct is shown inFigure 6B. The two A2p:HsACTB lines shown, HsB-1 and HsB-2,expressed almost 1.5 times as much human b-actin comparedwith wild-type total plant actin levels, and both showed almost full

suppression of all phenotypes (Figure 3B). The A2p:HsACTA1–expressing lines HsA1-2 and HsA1-4 produced levels of a1-actinsimilar to the amount of b-actin produced by the A2p:HsACTBlines, but they were still unable to rescue the mutant phenotype(Figure 3B). We found that plants showing high levels of ex-pression of either human actin variant, approximately threefoldabove wild type levels, had slightly diminished shoot architec-ture relative to the wild type, but had normal root growth, similarto the results we observed with strong Arabidopsis ACT2 over-expression (Kandasamy et al., 2002). Because the antibody usedin these assays, MAbGEa (Kandasamy et al., 1999), reacts almostequally with the human cytoplasmic b-actin ACTB and skeletalmuscle a1-actin ACTA1 (Figure 6A), the failure of phenotypicsuppression by Hs-ACTA1 and successful rescue by Hs-ACTBare not due to differences in steady state protein levels.

Human Cytoplasmic ACTG1, but Not Aortic Muscle ACTA2,Suppress the Developmental Defects of Plant VegetativeActin Double Mutants

Our above studies demonstrate that the cytoplasmic Hs-ACTBis very similar to the plant vegetative actins in supporting variousdevelopmental functions. To generalize the functional equiva-lence of the animal cytoplasmic actins and the lack of equiva-lence for muscle actins with the plant vegetative actins, weexamined the suppression of the act8-2 act7-4 plant mutant

Figure 5. Suppression of the Leaf Morphology Phenotype by Diverse Actins.

(A) The largest rosette leaf from 25-d-old plants. WT, the wild type.(B) Leaf length (white) versus leaf width (black) of wild-type, act8-2 act7-4 double mutant (8/7), and various transgenic lines expressing diverse actins.Hs, human; Ci, C. intestinalis; Mb,M. brevicollis; Ac, A. castellanii; and Cs, C. scutata protist actins. The data correspond to average values from the twolargest leaf samples from 10 plants (24 d) for each line shown, and the error bars represent SD.

Ancient Functional Competence of Actin 2047

phenotypes by the other human cytoplasmic actin, ACTG1,and the most divergent muscle actin, ACTA2 (for aortic mus-cle a2-actin). Hs-ACTG1 is 1.1% divergent from Hs-ACTB,whereas Hs-ACTA2 is 2.2% divergent from Hs-ACTA1 and5.6% divergent from Hs-ACTG1 (Figure 1B; see SupplementalTable 1 online). Expression of Hs-ACTG1 fully restored rootgrowth to act8-2 act7-4 seedlings, as the young and maturetransgenic plants showed no significant difference in root lengthand architecture from the wild type (Figure 4A, Table 1; seeSupplemental Figure 1A online). In addition, Hs-ACTG1 almostfully suppressed all the other prominent developmental defectsin act8-2 act7-4 plants, such as adventitious root development,inflorescence/flower architecture, plant (leaf) morphology, androot cell organization and polarity (Figures 4B, 5A, and 5B, Table1). Thus, Hs-ACTG1 is as efficient as Hs-ACTB in supporting var-ious aspects of Arabidopsis plant development.

On the other hand, aortic muscle actin Hs-ACTA2, like skeletalmuscle Hs-ACTA1, cannot efficiently suppress any vegetativeplant actin deficiencies, with the exception of mild suppres-sion of the leaf size phenotype (Figures 4 and 5, Table 1; seeSupplemental Figure 1A online). Immunoblot analysis revealedthat all 15 transgenic A2p:HsACTA2 plants isolated expressedabundant levels of aortic muscle actin, but none showed sig-nificant suppression of mutant phenotypes. MAbGEa, whichreacts equally with recombinant Hs-ACTG1 (HsG1, Figure 6A)and Hs-ACTA2 (HsA2, Figure 6A) proteins on immunoblots,showed a slightly weaker reactivity with Hs-ACTA2 than Hs-ACTG1 expressed in plants (Figure 6C). Therefore, we tested theplant expression levels of Hs-ACTA2 and Hs-ACTG1 using anantihuman muscle actin antibody. This MAbHsACT2 antibody,which reacts strongly with recombinant muscle actins (HsA1and HsA2; Figure 6A) and weakly with cytoplasmic actins (HsBand HsG1; Figure 6A), revealed that the levels of Hs-ACTA2were similar to Hs-ACTA1 muscle actins expressed in trans-genic plants (Figure 6D). Thus, even with sufficient levels ofexpression, Hs-ACTA2 is unable to suppress nearly every one ofthe plant actin mutant developmental phenotypes.

Human Cytoplasmic Actins, but Not Muscle Actins, RestoreRoot Hair Development in the act2-1 act8-2 Mutant

In Arabidopsis, ACT2 and ACT8 are essential for normal elon-gation of root hairs (Kandasamy et al., 2009), which are one ofthe fastest growing cell types in plants. We wanted to test ifhuman cytoplasmic b-actin and skeletal muscle a1-actin couldsubstitute for these two vegetative actins and rescue the polartip growth and bald root hair phenotypes of act2-1 act8-2double mutant plants. The Hs-ACTB and Hs-ACTA1 variantseach partially suppressed the root hair cell tip growth defect(Figures 7A to 7C). However, suppression by Hs-ACTB wasrelatively efficient, with most plant lines having 50 to 60% ofwild-type root hair length, while At-ACTA1–suppressed plantsnever produced root hairs more than 15% of normal length(Figure 7C, Table 1). Immunoblot analysis showed that bothhuman actin variants were equally well expressed in the act2-1act8-2 mutant background (Figure 7D). The two lines shown foreach construct (A2p:HsACTB and A2p:HsACTA1) contained 2to 3 times more total actin than the wild type. The act2-1 act8-2mutant plants have almost the same level of total plant actin asthe wild type, because of the induction of endogenous ACT7expression in this actin mutant background (Kandasamy et al.,2009). The degree to which human cytoplasmic b-actin rescuedroot hair growth is similar to the results we obtained in sup-pression studies using overexpression of At-ACT7, which re-stores root hair length in this double mutant to no more than65% of wild-type levels (Kandasamy et al., 2009). Although theplants expressing b-actin revealed significantly improved roothair growth, high levels of b-actin (e.g., HsB-1) retarded shootdevelopment, as the plants had smaller rosette leaves than thewild type (Figure 7E).Moreover, we tested the effect of expressing Hs-ACTG1 and

Hs-ACTA2 on root hair cell elongation in act2-1 act8-2 plants.Similar to Hs-ACTB, Hs-ACTG1 also moderately supported roothair cell elongation (Figure 8; see Supplemental Figure 1B on-line). The transformed Hs-ACTG1–expressing plants had root

Table 1. Summary of Suppression of the Arabidopsis Vegetative Actin Double Mutant Phenotypes by Diverse Actins

Actin Transgenein Double Mutant

Root HairLength (µm)a

Root Length(mm)b,c Cell Polarityb

EmbryoDevelopmentb

Root Architecture(mm)b,d

InflorescenceArchitectureb

Adventitious RootDevelopmentb,e Plant Morphologyb,f

Hs-ACTA1 50 6 8 1.2 6 0.05 Aberrant Aberrant Dwarf (14 6 1/8 6 1) Aberrant Poor (9 6 1/9 6 2) SemidwarfHs-ACTA2 40 6 5 1.2 6 0.04 Aberrant Aberrant Dwarf (14 6 2/8 6 1) Aberrant Poor (8 6 2/9 6 1) SemidwarfHs-ACTB 283 6 23 4.4 6 0.1 Normal Normal Normal (69 6 2/23 6 2) Normal Robust (33 6 5/13 6 4) NormalHs-ACTG1 238 6 21 4.3 6 0.1 Normal Normal Normal (68 6 1/24 6 2) Normal Robust (33 6 6/12 6 2) NormalCi-ACTB1 125 6 4 4.1 6 0.05 Normal Normal Normal (69 6 3/25 6 2) Normal Robust (35 6 7/13 6 2) NormalMb-ACT 572 6 6 4.4 6 0.2 Normal Normal Normal (72 6 2/24 6 2) Normal Robust (32 6 3/12 6 2) NormalAc-ACT 478 6 30 4.7 6 0.2 Normal Normal Normal (70 6 2/25 6 1) Normal Robust (27 6 3/13 6 2) NormalCs-ACT 523 6 26 4.8 6 0.1 Normal Normal Normal (71 6 2/26 6 1) Normal Robust (33 6 7/17 6 3) NormalWild type 575 6 29 4.4 6 0.1 Normal Normal Normal (68 6 4/24 6 1) Normal Robust (36 6 8/14 6 1) Normal2/8 0 4.4 6 0.2 Normal Normal Normal (69 6 3/26 6 4) Normal Robust (35 6 5/16 6 4) Normal8/7 297 6 61 1.1 6 0.1 Aberrant Aberrant Dwarf (12 6 2/8 6 2) Aberrant Poor (6 6 1/9 6 2) Dwarf

aact2-1 8-2 double mutant background.bact8-2 7-4 double mutant background.cThree-day-old seedling.dFourteen-day-old plant root length versus number of lateral roots.eLength of adventitious roots produced from leaf explants versus number of roots produced.fSee Figure 5 for quantification of leaf length versus leaf width. At least 10 independent transformed lines for each transgene revealed this phenotype.

2048 The Plant Cell

hairs that were 40% the length of the wild type, which is ;10%less suppression than by Hs-ACTB. On the other hand, Hs-ACTA2–induced root hair cell growth to <10% of the wild type(Figure 8, Table 1).

Ancestral Ciona b-Actin Also Supports Plant Actin Functions

Arabidopsis and humans have completely independent and di-vergent ontogenies separated by more than a billion years (Figure1A); hence, it is extremely unlikely that Arabidopsis vegetativeand human cytoplasmic actin variants converged recently oncommon structures and appropriate protein interactions withthe myriad of actin binding proteins that are essential for normalactin cytoskeletal functions. Yet, both these classes of actins func-tioned similarly in supporting various aspects of plant development.Therefore, we suggest instead that during their evolution froma common ancestral sequence, the animal cytoplasmic andplant vegetative actins preserved ancient functions essential forspatial development. To support this idea, we selected a cyto-plasmic-like actin (Ci-ACTB1) from an ancestral deuterostome,C. intestinalis, and tested the suppression of act8-2 act7-4 andact2-1 act8-2 mutant plant phenotypes. The C. intestinalis ACTB1actin is a clear sequence homolog of human cytoplasmic actinsand is 3 to 4% divergent from Hs-ACTB and Hs-ACTG1 (Figure1B; see Supplemental Table 1 online). C. intestinalis has not shareda common ancestor with humans for 700 million years (DeVrieset al., 2006); therefore, the C. intestinalis actin has had ample timeto diverge in function and sequence from human actins.Expression of C. intestinalis cytoplasmic-like actin ACTB1 effi-

ciently suppressed the developmental defects of the act8-2 act7-4mutant. For example, the root length and architecture of Ci-ACTB1–expressing seedlings was indistinguishable from the wildtype or from human ACTB-suppressed plants (Figure 4A; seeSupplemental Figure 1A online). Ci-ACTB1 actin protein was alsoable to fully rescue the adventitious root development (Figure4B) and leaf morphology (Figure 5). Moreover, the embryo de-velopment, inflorescence and flower morphology, and the rootapical organization and cell polarity phenotypes of act8-2 act7-4were also restored to the wild type (Table 1). Immunoblot analysiswith MAbGEa, which reacted strongly with recombinant Ci-ACTB1actin, revealed that Ci-ACTB1 expression levels in all transgenicplant lines were similar to or even higher than total actin levels inwild-type plants (Figures 9A and 9B). However, Ci-ACTB1 ex-pression in act2-1 act8-2 seedlings produced mostly short roothairs that were just 20% the length of the wild type (Figure 8; seeSupplemental Figure 1B online). Thus, the cytoplasmic-like actinCi-ACTB1 fully supports plant growth and development, but onlyfunction partially in restoring root hair cell elongation.

Protist Actins Ancestral to Plants and Animals SupportNormal Root Hair Tip Growth and Plant Development

To test our hypothesis that the competence for spatial de-velopment exhibited by cytoplasmic animal and vegetative plantactins is inherited from a common ancestral protist actin se-quence, we examined the ability of protist actins to suppressplant actin deficiency phenotypes. We tested three protist actinsequences that are ancestral to plants and animals: Monosiga

Figure 6. Immunoblot Analysis of Human Cytoplasmic (Hs-ACTB andHs-ACTG1) and Muscle Actin (Hs-ACTA1 and Hs-ACTA2) Expression inTransgenic Plants.

(A) Reactivity of antiactin antibodies MAbGEa (top panel) and MAbH-sACTA2 (middle panel) with human (HsB and HsA1, native; HsG1 andHsA2, recombinant) and plant (At-ACT2, recombinant) actins. The num-bers on the Coomassie blue–stained gel shown in the bottom panel in-dicate the amount of actin protein loaded.(B) Expression of Hs-ACTB (cytoplasmic b-actin) and Hs-ACTA1 (musclea1-actin) in act8 act7 (8/7) A2p:Hs-ACTB and A2p:Hs-ACTA1 plants,respectively, versus total plant actin in 8/7 plants. WT, the wild type.(C) Expression of Hs-ACTG1 and Hs-ACTA2 in act8 act7 plants. The toppanels in (B) and (C) were probed with MAbGEa. The bottom panelsshow images of Coomassie blue–stained duplicate gels. Protein ex-pression for two independent transformed lines is shown for eachtransgene.(D) Protein blot probed with anti-Hs-ACTA2 antibody MAbHsACTA2.Note this antibody reacts strongly with both human muscle actins andpoorly with native plant actins or human cytoplasmic actins expressed inplants. All lanes were loaded with 25 µg of total plant protein.

Ancient Functional Competence of Actin 2049

brevicollis, Acanthamoeba castellanii, and Coleochaete scutata.M. brevicollis is a single-celled protist belonging to Choanozoa,a phylum that is particularly close to the animal kingdom (Kinget al., 2008) (Figure 1A). The MbACT actin protein sequence is2 to 3% divergent from deuterostome cytoplasmic actins, 7 to8% divergent from muscle actins, and 10 to 13% divergent fromplant actins (Figure 1B; see Supplemental Table 1 online). A.castellanii is a single-celled protist that belongs to Archamoeba,a phylum basal to both the fungal and animal kingdoms butmore distant from plants. Ac-ACT1 actin is 4 to 14% divergentfrom the other actins being examined and is also more closelyrelated to cytoplasmic actins than plant actins. C. scutata isa multicellular green algal protist belonging to the Charophytes,the phylum believed to contain the closest living protist relativesto vascular plants but quite distant from animals and fungi.

However, the Cs-ACT actin sequence is only slightly closer onthe average to plant actin than to animal actin sequences (Figure1B; see Supplemental Table 1 online).Transgenes expressing the three protist actins A2p:MbACT,

A2p:AcACT1, and A2p:CsACT were each effective at sup-pressing all the developmental defects of the act8-2 act7-4mutant. For example, seedlings and adult plants containingthese transgenes and strongly expressing their encoded pro-teins (Figure 9C) had normal roots that were of indistinguishablelength and architecture from those of wild-type plants (Figures4A and 10A; see Supplemental Figure 1C online). Other de-velopmental defects of the mutant that were also completelyrestored to the wild-type phenotype include poor adventitiousroot development (Figures 4B and 10B), aberrant root apical cellorganization and polarity (Figure 10C), dwarf leaf (Figures 5 and

Figure 7. Rescue of the Root Hair Growth Phenotype of the act2-1 act8-2 Double Mutant by Human ACTB and ACTA1 Actins.

(A) Thirty-six-hour-old seedlings. WT, the wild type.(B) Three-day-old seedlings.(C) Root hair length of 3-d-old seedlings. Two independent transformed lines are shown for each construct in (A) and (C). Twenty-five maximally grownroot hairs from at least 10 seedlings of each line were used for the measurement. 2/8, act2-1 act8-2; HsB, Hs-ACTB; HsA1, Hs-ACT1. Error barsrepresent SD.(D) Immunoblots showing the expression of Hs-ACTB and Hs-ACT1 in two transgenic lines each.(E) Morphology of 26-d-old act2-1 act8-2 seedlings expressing different levels of Hs-ACTB (see [D]).

2050 The Plant Cell

10D) and inflorescence/flower morphology (Figure 10E), andaberrant development of embryos (Table 1). Similarly, expres-sion of Mb-ACT, Ac-ACT1, and Cs-ACT actins efficiently sup-pressed the root hair cell elongation and tip growth defects ofthe act2-1 act8-2 mutant, with root hairs being 99, 83, and 90%of wild-type length, respectively (Figure 8; see SupplementalFigure 1D online). Surprisingly, the restoration of root hair elon-gation by all three protist actins significantly exceeded that bythe deuterstome cytoplasmic actins.

DISCUSSION

Mounting evidence from previous studies supported the ideathat the divergent actin protein variants encoded in multicellulareukaryotes evolved distinct roles essential to cell and organis-mal development. Contrary to this view, we have shown herethat several deuterostome cytoplasmic actin variants and protistactins are functionally indistinguishable from plant vegetativeactins in their ability to suppress multiple developmental defectsin Arabidopsis mutants lacking vegetative actins. Earlier geneticsuppression studies generally resulted in lethality or weak sup-pression, perhaps because they relied upon the rescue of de-fects in more specialized actins, such as mouse and Drosophilamuscle actins or the relatively divergent budding yeast actin. Bycontrast, this study examined combinations of animal cyto-plasmic actin, plant vegetative actin, and protist actin se-quences that are similarly involved in the control of spatial celldevelopment. We can account for the successful suppressionresults observed in this study only if the suppressing actins areessentially indistinguishable in their cellular functions from thevegetative plant actins. Our data lend support to a new hy-pothesis that during their independent evolution, cytoplasmic

deuterostome actin variants and vegetative plant actin variantsinherited their functional competence for spatial developmentfrom an ancestral protist actin.Although Arabidopsis encodes eight actin variants, only the

three vegetative actins are predominantly expressed throughoutthe major sporophytic phase of plant development. We dem-onstrated earlier by ectopic expression and suppression studiesthat these vegetative actins are functionally distinct from thereproductive class of actins expressed mainly in pollen andovules (Kandasamy et al., 2002, 2007). Moreover, genetic sup-pression studies suggest that At-ACT2 and At-ACT8 can sub-stitute for each other and for At-ACT7 function, but At-ACT7 canonly partially support At-ACT2 and At-ACT8 functions in theregulation of root hair tip growth, thus providing evidence forboth functional redundancy and divergence, even among theplant vegetative actin variants (Kandasamy et al., 2009). Here,we have shown that the three plant vegetative actins are func-tionally equivalent to at least three animal cytoplasmic actinsand three protist actins, as these nonplant actins are able to fullyrescue all the organ development phenotypes of plant actindouble mutants. However, unlike the protist actins that supportvery efficiently the highly specialized tip growth of root hairs, thehuman and C. intestinalis cytoplasmic actins are only able topartially rescue root hair growth. Thus, the cytoplasmic actinsbehaved very similarly to vegetative At-ACT7 in supporting onlymoderate growth of root hairs, although all of them completelyrescued plant development. Perhaps At-ACT7 and the animalcytoplasmic actins have partially lost their ability to supportrapid cell tip growth and elongation of specialized cells like roothairs, while the protist actins have preserved this property.It is surprising that Hs-ACTB and Hs-ACTG1, which differ

in four N-terminal residues and show dramatic ion-dependent

Figure 8. Suppression of the Root Hair Cell Growth Phenotype by Diverse Actins.

Root hair length of 3-d-old seedlings of wild-type (WT), act2-1 act8-2 double mutant (2/8), and various transgenic lines expressing diverse actins. Thedata represent average values from three biological replicates, with each experiment containing 25 samples for each line shown. The error bars indicateSD. HsB, Hs-ACTB; HsG1, Hs-ACTG1; HsA1, Hs-ACTA1; HsA2, Hs-ACTA2 (human); CiB, Ci-ACTB1 (C. intestinalis); Mb, Mb-ACT (M. brevicollis);Ac, Ac-ACT (A. castellanii ); Cs, Cs-ACT (C. scutata).

Ancient Functional Competence of Actin 2051

(e.g., Ca2+ and Mg2+) polymerization differences (Bergeron et al.,2010) and apparently serve different functions in humans, areboth capable of suppressing all the plant developmental phe-notypes. The only distinction between these two human cyto-plasmic actins is minor differences in the suppression of roothair cell growth, which is known to be Ca2+ dependent (Wymeret al., 1997; Fan et al., 2011). Hs-ACTB induces root hair growth10 to 20% better than Hs-ACTG1, consistent with HsACTBbeing more Ca2+ responsive than Hs-ACTG1. This differentialsuppression of root hair growth among these two cytoplasmicactins suggests that unlike most of the plant organs or tissues,these specialized root cells are able to recognize the functionaldifferences in these two human sequences. This difference alsomay be due to actin binding protein interaction with the N ter-minus or allosteric effects of the N terminus on the rest of themolecule.

The functional homology observed among animal cytoplas-mic, plant vegetative, and extant protist actins, as revealed byour positive developmental suppression data, fits well with whathas been described as nonprogressive sequence evolution,where ancient protein structures and functions are conservedbut the protein sequences are not (Castrodeza, 1979). As plantvegetative actins evolved from common ancestral sequences,they do not appear to have evolved many new functions that arenot present in some protist actins or deuterostome cytoplasmic

actins. Several observations support this model. First, phylo-genetic clustering of the actin sequences examined demon-strated that the animal cytoplasmic actins and two of the protistactin sequences (i.e., Mb-ACT and Ac-ACT1), which suppressedloss of plant actin function, are quite distinct in their sequencesfrom the plant actins (Figure 1B; see Supplemental Figure 2online). Second, in both animal and yeast actins, a functional

Figure 9. Immunoblot Analysis of C. intestinalis and Protist Actin Ex-pression in act8-2 act7-4 Plants.

(A) Reactivity of MAbGEa with C. intestinalis cytoplasmic-like actinACTB1 (CiB) and diverse protist actins. Ac, A. castellanii; Mb, M. brevi-collis; Cs, C. scutata. About 1 mg of crude recombinant actin was loadedper lane.(B) Expression of Ci-ACTB1 in act8-2 act7-4 plants. Three independentlines are shown. WT, the wild type.(C) Expression of protist actin proteins in act8-2 act7-4 plants. Two in-dependent lines are shown for each protist actin examined.

Figure 10. Rescue of act8-2 act7-4 Double Mutant Phenotypes byProtist Actins.

(A) Root architecture of 14-d-old seedlings. WT, the wild type; 8/7, act8-2act7-4; Ac, A. castellanii; Mb, M. brevicollis; Cs, C. scutata.(B) Adventitious roots developed from leaves.(C) DAPI-stained root tips showing cellular organization.(D) Twenty-six-day-old plants.(E) Flowers from adult plants.Bars = 50 mm in (C) and 1 mm in (E).

2052 The Plant Cell

role has been shown for the N-terminal region containing severalacidic amino acids (Lasa et al., 1997; Hinz et al., 2002; Wonget al., 2002; Lu et al., 2005). Therefore, we were surprised toobserve that the animal cytoplasmic actins (i.e., Hs-ACTG1, Hs-ACTB, and Ci-ACTB1) and the protist actins (i.e., Ac-ACT1 andMb-ACT), which are two amino acids shorter than all plant actinsat the N terminus (see Supplemental Figure 2 online), efficientlysubstituted for plant actins. By contrast, the muscle actins, whichare the same length in the N-terminal acidic region as plant actins,did not support plant actin function. In this case, actin function,but not the length at the N terminus, was conserved among actinsequences that rescued mutant phenotypes. Third, changes inprotein sequence under a nonprogressive evolutionary modelmay be viewed as intragenic suppression, where a change inone residue with a deleterious effect is compensated by selec-tion of a second change in sequence that restores the originalfunction. This process of selecting for intragenic suppressormutations to maintain original structure and function has beenproposed to explain paired mutations in HIV-1 protease (Pareraet al., 2009), cytochrome c oxidase subunit I (Acín-Pérez et al.,2003), b-subunit of RNA polymerase (Malik et al., 1999), andaminoacyl-tRNA synthetase (Ador et al., 2004) and for thecomplementary bases in the double-stranded stems of struc-tural RNAs (e.g., tRNA, rRNA, small nuclear RNA, etc.). A com-parison of the sequences and three-dimensional structures (seeSupplemental Figures 2 and 3 online) of At-ACT2 and Hs-ACTBsuggests that their evolution from a common ancestral actinoccurred by the codivergence of sets of adjacent amino acids.For example, on one face of actin’s structure depicted inSupplemental Figure 3 online, human b-actin ACTB differs fromArabidopsis ACT2 in five groups of clustered residue changes:Ile-75/Thr-77, Ser-365/Ser-368, Phe-223/Leu-211, Ala-231/Ser-232/Ala-228/Asn-252, and Ser-271/Cys-272/Phe-279. Suchclustered changes in sequence account for the majority of theiramino acid differences and may represent similar cases of in-tragenic suppression to maintain wild-type activity. Humanskeletal muscle actin ACTA1, which did not suppress plant actinfunctions, is similarly divergent in many pairs or small sets ofadjacent residues, perhaps adjusting for distinct new musclefunctions, while maintaining other actin functions, such as nu-cleotide binding and exchange. Previous studies examining theability of Drosophila cytoplasmic actin 5C to suppress the lossof flight muscle actin 88F indicated that the combined activitiesof a number of sequence changes comprised the functionaldifferences of these two variants (Fyrberg et al., 1998). Finally, inspite of the sequence divergence of the cytoplasmic animal andprotist actins from the vegetative plant actins, they must haveconserved appropriate strength and specificity to their bindingof many classes of actin binding proteins. However, the plantactin binding proteins are all quite divergent from those in otherkingdoms. For example, plant profilins and actin depolymerizingfactors (cofilins) are only 15 and 31% identical in protein se-quence to their closest human homologs. Yet the efficient func-tion of animal cytoplasmic actins in plants demonstrates that theymust interact appropriately with plant profilins and actin depoly-merizing factors.

Although the muscle actins differed by 10 to 14% (37 to 52amino acid residues) from the plant sequences and the animal

cytoplasmic and protist actins differed by only 7 to 13% (Table1), the suppressing sequences are not phylogenetically muchcloser to the plant sequences (Figure 1B). However, among thesuppressing actins, we found a few conserved amino acid res-idues that distinguished them from the nonsuppressing actins(see Supplemental Figure 2 online). These residues could beessential to actin’s conservation of competence for spatial de-velopment. Based on the residue numbering for At-ACT2, Met-18,Val-78, Thr-164, Leu-178, Thr-203, and Val-288 were perfectlyconserved among the nonplant actins that suppressed the act8-2act7-4 developmental phenotypes or partially suppressed roothair elongation in act2-1 act8-2, while these residues had di-verged in the muscle actins. One of these residues, Val-78, is anIle (Ile-78) in the deuterostome muscle actins we examined andin the protistome Drosophila flight muscle actin 88F. Conversionof Ile-78 in flight muscle actin to Val-78 (i.e., the cytoplasmicanimal actin residue) resulted in a flightless phenotype showingthe importance of this residue to muscle actin function (Fyrberget al., 1998). However, the role of Val-78 in supporting the de-velopmental activities of nonmuscle actins is not known. We willtest a number of mutant actins in the plant system to determinethe importance of these and other residues for cell growth andorgan developmental functions.Two human muscle actins failed to support root organ de-

velopment and growth in the act8-2 act7-4 mutant, althoughthere was a minor improvement in shoot morphology. Addi-tionally, the muscle actins poorly supported root hair cell tipgrowth in the act2-1 act8-2 mutant. Thus, there can be littledoubt that muscle actins in the deuterostome lineage lost theircompetence for spatial development. While we can only spec-ulate as to the reasons for this lack of conserved function, a fewaspects of their divergence seem particularly worthy of mention.The deuterostome muscle actins are highly specialized in theircellular functions (Khaitlina, 2001) and more divergent in se-quence than cytoplasmic actin, as revealed by the longer branchlengths for muscle actins in some phylogenetic analyses of actinsequences (Figure 1B). Cytoplasmic actins are expressed in allcells, including muscle cells, while the muscle actins are all tis-sue specific. Knockout mice lacking cytoplasmic g-actin ACTG1initially develop what appear to be normal skeletal muscles butacquire a progressive myopathy with the subsequent degen-eration of skeletal muscle fibers (Sonnemann et al., 2006). Thus,while many cells and tissues develop normally using only cy-toplasmic actins, muscle tissues require both cytoplasmic andmuscle actin variants, suggesting muscle actins may no longerbe competent in fulfilling all cellular and developmental func-tions. Moreover, the folding of newly synthesized actin intoa functional monomer is a chaperone-mediated process. Theinability of muscle actins to suppress plant actin function couldbe due to the divergence of plant chaperone machinery andinappropriate folding into active molecules. For example, a yeastin vitro actin folding system will assemble functional humancytoplasmic ACTB but could not fold functional muscle-specificACTA1 (Altschuler et al., 2009). Changing Hs-ACTA1 Asn-299 tothe Val residue found in yeast ACT1 allowed normal folding ofHs-ACTA1 in this system. Asn-299 is a deuterostome muscle-specific actin residue, while cytoplasmic animal and two protistactins we examined have Thr, and plants and the green alga

Ancient Functional Competence of Actin 2053

protist have an Ile in this position. Although the muscle actinswere well expressed in plants and equally soluble with the cy-toplasmic actins in the low salt extraction buffers used for im-munoblot analysis in this study, it will be important to confirmthe normal folding of muscle actins expressed in plant cells.Moreover, we are in the process of examining the organizationof the actin cytoskeleton in the leaf and root tissues of muscleactin expressing plants that are transformed with the fluorescentfimbrin reporter ABD2-GFP to see whether and how these actinsare integrated into the actin filaments in these mildly suppressedand nonsuppressed cell types.

Human cytoplasmic actins appear to function normally in theArabidopsis model system. It would be interesting to seewhether any of the vegetative plant actins or the ancestral protistactins can function in animals and suppress the embryo lethalphenotype of mice lacking ACTB (Shawlot et al., 1998; Bunnellet al., 2011) or the delayed embryonic development and poorsurvival of mice lacking ACTG1 (Bunnell and Ervasti, 2010).Moreover, because the human cytoplasmic actins are functionalin Arabidopsis, missense mutations in these human actins that areassociated with a large number of diseases, including blindness,deafness, myopathies, immune deficiency, and aberrant kidneyfunction, could be easily tested in this plant model. For example,the influence of dominant missense mutations in b-actin on thestructure of lamellapodia in lymphoblasts (Procaccio et al., 2006)and in g-actin on the maintenance of stereocilia in the ear (Morínet al., 2009) might be studied in Arabidopsis root hairs. By con-trast, examining actin functions using transgenic mouse modelswould be orders of magnitude more expensive and time con-suming. Our positive suppression data indicate that the humancytoplasmic actins and some protist actins have the necessaryinteractions with a variety of plant actin binding proteins to con-duct normal spatial cellular development. However, a significantunresolved issue raised by our results is how the diverse actinfunctions necessary to support multicellular development couldhave evolved in and are still maintained in single-celled protists.

METHODS

Plant Material and Generation of Transgenic Plants

Arabidopsis thalianawild-type (Wassilewskija or Columbia), act2-1 act8-2and act8-2 act7-4 double null mutants (Kandasamy et al., 2009), andvarious suppression lines were grown at 22°C with 16 h of light and 8 hdarkness. Root morphologies and root hair cell growth were examined onsterile plants grown vertically on Murashige and Skoog (MS) medium(Murashige and Skoog, 1962) solidified with 1% agar, and abovegroundplant morphologies were observed on soil-grown plants. Fourteen-day-old plants were used for examination of lateral root formation and rootarchitecture, and 24-d-old plants were used for quantitative analysis ofleaf size. Roots and root hairs from 25 and 10 T2 generation seedlings (3 dold) of each plant line were examined to quantify root length and root hairlength, respectively.

To induce adventitious roots, leaves from sterile seedlings were cul-tured on solid MS medium supplemented with 0.5 mg/L naphthaleneacetic acid and 0.33mg/L indole-3-acetic acid for 4 d. Leaf explants werethen transferred to MS plates without hormones and incubated verticallyfor 6 d for adventitious root growth measurements. For suppressionstudies, the double mutant plants were transformed with various con-structs by the Agrobacterium tumefaciens–mediated floral dip method

(Clough, 2005). Fifteen to 20 independent transformed lines in eachmutant genetic background were isolated for every transgene by platingsterilized seeds on medium supplemented with suitable antibiotics.

Comparison of Actin Variant Sequences

The actin protein sequences examined include At-ACT2, At-ACT8, At-ACT1, and At-ACT7 (Arabidopsis); Hs-ACTB, Hs-ACTG1, Hs-ACTA1, andHs-ACTA2 (Homo sapiens); Ci-ACTB1 (Ciona intestinalis); Mb-ACT (Mon-osiga brevicollis); Ac-ACT1 (Acanthamoeba castellanii); and Cs-ACT (Co-leochaete scutata). The previously uncharacterized C. intestinalis sequenceCi-ACTB1 was selected from the C. intestinalis family of actin sequencesbased on its similarity to human ACTB. The unrooted neighbor-joining treepresented in Figure 1Bwas constructed usingMEGA4 (Tamura et al., 2007).The alignment of the various actin protein sequences was performed usingPileUp in GCG version 11.1 (Accelrys), and the multiple-alignment outputwas made using BOXSHADE version 3.31 C.

Plasmid Construction

The native cDNAs encoding the actin proteins described above (Hs-ACTB,Hs-ACTG1, Hs-ACTA1, Hs-ACTA2, Ci-ACTB1, Mb-ACT, Ac-ACT1, andCs-ACT) were synthesized and cloned into pUC57 by GenScript. Each ofthese constructs was then moved into the NcoI/BamHI sites of the Arab-idopsis ACTIN2 vector (ACT2pt, A2p) for ubiquitous expression in mutantplants to study suppression. This fragment was also moved into pET15bfrom EMD4Biosciences for actin protein expression in bacteria to examinethe reactivity of the antiactin monoclonal antibodies with different actinvariants. The various steps involved in cloning the constructs into ACT2ptwere identical to those described previously for the A2P:A2 construct(Kandasamy et al., 2002). The expression plasmids were mobilized intothe Agrobacterium strain C58C1 before plant transformation. The actinclones in pET15b were transformed into the expression host BL21 Star(DE3) from Invitrogen for recombinant actin protein expression as describedearlier (Kandasamy et al., 1999).

Immunoblot Analysis

To detect actin on immunoblots, total protein from 50 mg of frozen leafsamples was ground frozen in liquid nitrogen and extracted in 250 mLof extraction buffer containing 25 mM Tris-HCl, pH 7.5, 10 mM NaCl,10 mM MgCl2, 5 mM EDTA, and protease inhibitor cocktail (RocheDiagnostics; one tablet/10 mL). After centrifugation, the supernatantwasmixed 1:1 with 23 sample buffer (O’Farrell, 1975) and boiled (5 min),and ;20 mL was loaded per well (i.e., ;25 mg protein) on a 10% SDS-polyacrylamide gel. The human, C. intestinalis, and protist actins on im-munoblots were detected with a mouse monoclonal antibody MAbGEa(Thermo Scientific; MA1-744), which recognized all the eukaryotic ac-tins equally (Kandasamy et al., 1999). Moreover, an anti-HsACT2 an-tibody (MAbHsACTA2) produced in mouse (Sigma-Aldrich; 4A8-2H3)was used to confirm the expression levels of human muscle actins.Both antibodies were used at a dilution of 1:500 in the blocking solutioncontaining TBST (10 mM Tris-HCl, pH 7.5, 150 mM NaCl, and 0.05%Tween 20), 10% goat serum, and 5% dry milk. Purified rabbit skeletalmuscle actin (Hs-ACTA1) and human cytoplasmic b-actin (Hs-ACTB)that were used for control experiments to show equal reactivity ofMAbGEa with both these cytoplasmic, and muscle actins were obtainedfrom Cytoskeleton, Inc. The crude bacterial extracts containing re-combinant human (Hs-ACTG1 and Hs-ACTA2), C. intestinalis (Ci-ACTB1), and protist (Ac-ACT1, Mb-ACT, and Ce-ACT) actins made inEscherichia coli were assayed on immunoblots to confirm the reactivityof the antiactin antibodies. Coomassie Brilliant Blue staining of dupli-cate gels was used to monitor equal loading of proteins and adjustloading if necessary.

2054 The Plant Cell

Microscopy

Roots used for the examination of the cellular organization and cell lin-eages at the apex were fixed with 4% paraformaldehyde in 50 mM PIPESbuffer, pH 7.0, containing 5 mM EDTA, 1 mM MgCl2, 0.5% casein, and0.05%Triton X-100 for 30min to 1 h. The root segments were stained with49,6-diamidino-2-phenylindole (DAPI; 0.1 mg/mL) before observation witha Leica fluorescence microscope (Kandasamy et al., 2009). For root androot hair length measurements, 3-d-old seedlings were photographedusing a Leica stereomicroscope fitted with a color digital camera.

Accession Numbers

Sequence data from this article can be found in the Arabidopsis GenomeInitiative or GenBank/EMBL databases under the following accessionnumbers: NP_188508.1, NP_175350, NP_001031504.1, NP_196543.1(Arabidopsis), P60709.10, CAA27723.1, NP_001091.1, NP_001604.1 (H.sapiens), NP_001027674.1 (C. intestinalis), XP_001747496.1 (M. brevi-collis), P02578.1 (A. castellanii ), and 065315.1 (C. scutata).

Supplemental Data

The following materials are available in the online version of this article.

Supplemental Figure 1. Suppression of Root and Root Hair GrowthPhenotypes of Arabidopsis Double Mutants by Human and C.intestinalis Animal and Protist Actins.

Supplemental Figure 2. Comparison of the Sequence of DiverseActin Variants.

Supplemental Figure 3. Comparison of the Structure of Arabidopsisand Human Actin Variants.

Supplemental Table 1. Fraction of Amino Acid Differences among theActins Examined in This Study.

Supplemental Data Set 1. Text File of the Actin Sequences Used forConstruction of the Phylogenetic Tree Shown in Figure 1B.

ACKNOWLEDGMENTS

We thank Mark Farmer, Jessica Kissinger, and Jenna Oberstaller for theirassistance in the consideration of protists and Nancy Manley for herhelpful comments concerning evolution and development throughoutthis project. We acknowledge the University of Georgia’s AdvancedComputing Resource Center for their facilities and technical expertise.This study was supported by a grant from the National Institutes ofHealth (GM36397-25) to R.B.M. and by National Institutes of HealthGenetics Training Grant GM 07103-35 to E.R.

AUTHOR CONTRIBUTIONS

M.K.K. and R.B.M. designed the experiments and wrote the article. M.K.K.and E.C.M. performed the laboratory research and analyzed the data.E.R. contributed to the construction of actin phylogenetic trees.

Received December 31, 2011; revised April 2, 2012; accepted April 26,2012; published May 15, 2012.

REFERENCES

Acín-Pérez, R., Bayona-Bafaluy, M.P., Bueno, M., Machicado, C.,Fernández-Silva, P., Pérez-Martos, A., Montoya, J., López-

Pérez, M.J., Sancho, J., and Enríquez, J.A. (2003). An intragenicsuppressor in the cytochrome c oxidase I gene of mouse mito-chondrial DNA. Hum. Mol. Genet. 12: 329–339.

Ador, L., Jaeger, S., Geslain, R., Martin, F., Cavarelli, J., and Eriani,G. (2004). Mutation and evolution of the magnesium-binding site ofa class II aminoacyl-tRNA synthetase. Biochemistry 43: 7028–7037.

Altschuler, G.M., Dekker, C., McCormack, E.A., Morris, E.P., Klug,D.R., and Willison, K.R. (2009). A single amino acid residue is re-sponsible for species-specific incompatibility between CCT andalpha-actin. FEBS Lett. 583: 782–786.

Anderson, I.J., Watkins, R.F., Samuelson, J., Spencer, D.F., Majoros,W.H., Gray, M.W., and Loftus, B.J. (2005). Gene discovery in theAcanthamoeba castellanii genome. Protist 156: 203–214.

Asmussen, M.A., Gilliland, L.U., and Meagher, R.B. (1998). De-tection of deleterious genotypes in multigenerational studies. II.Theoretical and experimental dynamics with selfing and selection.Genetics 149: 727–737.

Bergeron, S.E., Zhu, M., Thiem, S.M., Friderici, K.H., andRubenstein, P.A. (2010). Ion-dependent polymerization differencesbetween mammalian beta- and gamma-nonmuscle actin isoforms.J. Biol. Chem. 285: 16087–16095.

Brault, V., Reedy, M.C., Sauder, U., Kammerer, R.A., Aebi, U., andSchoenenberger, C. (1999). Substitution of flight muscle-specificactin by human (beta)-cytoplasmic actin in the indirect flight muscleof Drosophila. J. Cell Sci. 112: 3627–3639.

Bryan, K.E., Wen, K.K., Zhu, M., Rendtorff, N.D., Feldkamp, M.,Tranebjaerg, L., Friderici, K.H., and Rubenstein, P.A. (2006). Ef-fects of human deafness gamma-actin mutations (DFNA20/26) onactin function. J. Biol. Chem. 281: 20129–20139.

Bunnell, T.M., Burbach, B.J., Shimizu, Y., and Ervasti, J.M. (2011).b-Actin specifically controls cell growth, migration, and the G-actinpool. Mol. Biol. Cell 22: 4047–4058.

Bunnell, T.M., and Ervasti, J.M. (2010). Delayed embryonic de-velopment and impaired cell growth and survival in Actg1 null mice.Cytoskeleton (Hoboken) 67: 564–572.

Castrodeza, C. (1979). Non-progressive evolution, the Red Queenhypothesis, and the balance of nature. Acta Biotheor. 28: 11–18.

Clough, S.J. (2005). Floral dip: Agrobacterium-mediated germ linetransformation. Methods Mol. Biol. 286: 91–102.

Crawford, K., Flick, R., Close, L., Shelly, D., Paul, R., Bove, K.,Kumar, A., and Lessard, J. (2002). Mice lacking skeletal muscleactin show reduced muscle strength and growth deficits and dieduring the neonatal period. Mol. Cell. Biol. 22: 5887–5896.

Derelle, E. et al. (2006). Genome analysis of the smallest free-livingeukaryote Ostreococcus tauri unveils many unique features. Proc.Natl. Acad. Sci. USA 103: 11647–11652.

DeVries, M.E., Kelvin, A.A., Xu, L., Ran, L., Robinson, J., andKelvin, D.J. (2006). Defining the origins and evolution of the che-mokine/chemokine receptor system. J. Immunol. 176: 401–415.

Donnell, A.M., Doi, T., Hollwarth, M., Kalicinski, P., Czauderna, P.,and Puri, P. (2008). Deficient alpha-smooth muscle actin as a causeof functional intestinal obstruction in childhood. Pediatr. Surg. Int.24: 1191–1195.

Fan, J.L., Wei, X.Z., Wan, L.C., Zhang, L.Y., Zhao, X.Q., Liu, W.Z.,Hao, H.Q., and Zhang, H.Y. (2011). Disarrangement of actin fila-ments and Ca2+ gradient by CdCl2 alters cell wall construction inArabidopsis thaliana root hairs by inhibiting vesicular trafficking. J.Plant Physiol. 168: 1157–1167.

Fritz-Laylin, L.K. et al. (2010). The genome of Naegleria gruberi illu-minates early eukaryotic versatility. Cell 140: 631–642.

Fyrberg, E.A., Fyrberg, C.C., Biggs, J.R., Saville, D., Beall, C.J., andKetchum, A. (1998). Functional nonequivalence of Drosophila actinisoforms. Biochem. Genet. 36: 271–287.

Ancient Functional Competence of Actin 2055

Gilliland, L.U., Kandasamy, M.K., Pawloski, L.C., and Meagher,R.B. (2002). Both vegetative and reproductive actin isovariantscomplement the stunted root hair phenotype of the Arabidopsisact2-1 mutation. Plant Physiol. 130: 2199–2209.

Gilliland, L.U., McKinney, E.C., Asmussen, M.A., and Meagher,R.B. (1998). Detection of deleterious genotypes in multigenerationalstudies. I. Disruptions in individual Arabidopsis actin genes. Ge-netics 149: 717–725.

Gilliland, L.U., Pawloski, L.C., Kandasamy, M.K., and Meagher,R.B. (2003). Arabidopsis actin gene ACT7 plays an essential role ingermination and root growth. Plant J. 33: 319–328.

Hedges, S.B., Blair, J.E., Venturi, M.L., and Shoe, J.L. (2004). Amolecular timescale of eukaryote evolution and the rise of complexmulticellular life. BMC Evol. Biol. 4: 1–9.

Hightower, R.C., and Meagher, R.B. (1986). The molecular evolutionof actin. Genetics 114: 315–332.

Hinz, B., Gabbiani, G., and Chaponnier, C. (2002). The NH2-terminalpeptide of alpha-smooth muscle actin inhibits force generation bythe myofibroblast in vitro and in vivo. J. Cell Biol. 157: 657–663.

Kandasamy, M.K., Burgos-Rivera, B., McKinney, E.C., Ruzicka,D.R., and Meagher, R.B. (2007). Class-specific interaction of pro-filin and ADF isovariants with actin in the regulation of plant de-velopment. Plant Cell 19: 3111–3126.

Kandasamy, M.K., Gilliland, L.U., McKinney, E.C., and Meagher,R.B. (2001). One plant actin isovariant, ACT7, is induced by auxinand required for normal callus formation. Plant Cell 13: 1541–1554.

Kandasamy, M.K., McKinney, E.C., and Meagher, R.B. (1999). Thelate pollen-specific actins in angiosperms. Plant J. 18: 681–691.

Kandasamy, M.K., McKinney, E.C., and Meagher, R.B. (2002).Functional nonequivalency of actin isovariants in Arabidopsis. Mol.Biol. Cell 13: 251–261.

Kandasamy, M.K., McKinney, E.C., and Meagher, R.B. (2009). Asingle vegetative actin isovariant overexpressed under the controlof multiple regulatory sequences is sufficient for normal Arabidopsisdevelopment. Plant Cell 21: 701–718.

Karlsson, R., Aspenström, P., and Byström, A.S. (1991). A chickenbeta-actin gene can complement a disruption of the Saccharomy-ces cerevisiae ACT1 gene. Mol. Cell. Biol. 11: 213–217.

Keeling, P.J. (2007). Genomics. Deep questions in the tree of life.Science 317: 1875–1876.

Khaitlina, S.Y. (2001). Functional specificity of actin isoforms. Int.Rev. Cytol. 202: 35–98.

King, N. et al. (2008). The genome of the choanoflagellate Monosigabrevicollis and the origin of metazoans. Nature 451: 783–788.

Kumar, A. et al. (1997). Rescue of cardiac alpha-actin-deficient miceby enteric smooth muscle gamma-actin. Proc. Natl. Acad. Sci. USA94: 4406–4411.

Lasa, I., Gouin, E., Goethals, M., Vancompernolle, K., David, V.,Vandekerckhove, J., and Cossart, P. (1997). Identification oftwo regions in the N-terminal domain of ActA involved in the actincomet tail formation by Listeria monocytogenes. EMBO J. 16:1531–1540.

Liu, R.Y., Ferrenberg, A.M., Gilliland, L.U., Meagher, R.B., andAsmussen, M.A. (2003). Detection of deleterious genotypes inmultigenerational studies. III. Estimation of selection components inhighly selfing populations. Genet. Res. 82: 41–53.

Lu, X., Bryant, M.K., Bryan, K.E., Rubenstein, P.A., and Kawai, M.(2005). Role of the N-terminal negative charges of actin in forcegeneration and cross-bridge kinetics in reconstituted bovine car-diac muscle fibres. J. Physiol. 564: 65–82.

Malik, T., Ahmad, K., Buyukuslu, N., Cromie, K., and Glass, R.E. (1999).Intragenic suppression of trans-dominant lethal substitutions in theconserved GEME motif of the beta subunit of RNA polymerase:

Evidence for functional cooperativity within the C-terminus. Genes Cells4: 501–515.

Meagher, R.B. (1995). The impact of historical contingency on genephylogeny: Plant actin diversity. In Evolutionary Biology, M. Hecht,R. MacIntyre, and M. Clegg, eds (New York: Plenum Press), pp.195–215.

Meagher, R.B., Kandasamy, M.K., and King, L. (2011). Actin func-tions in the cytoplasmic and nuclear compartments. In Plant Cy-toskeleton. Advances in Plant Biology, Vol. 2, B. Liu, ed (Dordrecht,The Netherlands: Springer Science/Business Media), pp. 3–32.

Meagher, R.B., McKinney, E.C., and Vitale, A.V. (1999). The evolu-tion of new structures: Clues from plant cytoskeletal genes. TrendsGenet. 15: 278–284.

Merchant, S.S. et al. (2007). The Chlamydomonas genome revealsthe evolution of key animal and plant functions. Science 318:245–250.

Morín, M., Bryan, K.E., Mayo-Merino, F., Goodyear, R., Mencía, A.,Modamio-Høybjør, S., del Castillo, I., Cabalka, J.M., Richardson,G., Moreno, F., Rubenstein, P.A., and Moreno-Pelayo, M.A.(2009). In vivo and in vitro effects of two novel gamma-actin (ACTG1)mutations that cause DFNA20/26 hearing impairment. Hum. Mol.Genet. 18: 3075–3089.

Murashige, T., and Skoog, F. (1962). A revised medium for rapidgrowth and bioassays with tobacco tissue culture. Plant Physiol.15: 473–497.

Nunoi, H., Yamazaki, T., Tsuchiya, H., Kato, S., Malech, H.L.,Matsuda, I., and Kanegasaki, S. (1999). A heterozygous mutationof beta-actin associated with neutrophil dysfunction and recurrentinfection. Proc. Natl. Acad. Sci. USA 96: 8693–8698.

O’Farrell, P.H. (1975). High resolution two-dimensional electropho-resis of proteins. J. Biol. Chem. 250: 4007–4021.

Parera, M., Perez-Alvarez, N., Clotet, B., and Martínez, M.A. (2009).Epistasis among deleterious mutations in the HIV-1 protease. J.Mol. Biol. 392: 243–250.

Pollard, T.D., and Cooper, J.A. (2009). Actin, a central player in cellshape and movement. Science 326: 1208–1212.

Ponte, P., Gunning, P., Blau, H., and Kedes, L. (1983). Human actingenes are single copy for alpha-skeletal and alpha-cardiac actin butmulticopy for beta- and gamma-cytoskeletal genes: 39 untranslatedregions are isotype specific but are conserved in evolution. Mol.Cell. Biol. 3: 1783–1791.

Procaccio, V. et al. (2006). A mutation of beta -actin that alters de-polymerization dynamics is associated with autosomal dominantdevelopmental malformations, deafness, and dystonia. Am. J. Hum.Genet. 78: 947–960.

Schildmeyer, L.A., Braun, R., Taffet, G., Debiasi, M., Burns, A.E.,Bradley, A., and Schwartz, R.J. (2000). Impaired vascular con-tractility and blood pressure homeostasis in the smooth musclealpha-actin null mouse. FASEB J. 14: 2213–2220.

Shawlot, W., Deng, J.M., Fohn, L.E., and Behringer, R.R. (1998).Restricted beta-galactosidase expression of a hygromycin-lacZgene targeted to the beta-actin locus and embryonic lethality ofbeta-actin mutant mice. Transgenic Res. 7: 95–103.

Sonnemann, K.J., Fitzsimons, D.P., Patel, J.R., Liu, Y., Schneider,M.F., Moss, R.L., and Ervasti, J.M. (2006). Cytoplasmic gamma-actin is not required for skeletal muscle development but its ab-sence leads to a progressive myopathy. Dev. Cell 11: 387–397.

Sparrow, J.C., Nowak, K.J., Durling, H.J., Beggs, A.H., Wallgren-Pettersson, C., Romero, N., Nonaka, I., and Laing, N.G. (2003).Muscle disease caused by mutations in the skeletal muscle alpha-actin gene (ACTA1). Neuromuscul. Disord. 13: 519–531.

Staiger, C.J., and Blanchoin, L. (2006). Actin dynamics: Old friendswith new stories. Curr. Opin. Plant Biol. 9: 554–562.

2056 The Plant Cell

Tamura, K., Dudley, J., Nei, M., and Kumar, S. (2007). MEGA4:Molecular Evolutionary Genetics Analysis (MEGA) software version4.0. Mol. Biol. Evol. 24: 1596–1599.

Wong, W.W., Gerson, J.H., Rubenstein, P.A., and Reisler, E. (2002).Thin filament regulation and ionic interactions between the N-terminalregion in actin and troponin. Biophys. J. 83: 2726–2732.

Wymer, C.L., Bibikova, T.N., and Gilroy, S. (1997). Cytoplasmic freecalcium distributions during the development of root hairs of Ara-bidopsis thaliana. Plant J. 12: 427–439.

Zhu, M., Yang, T., Wei, S., DeWan, A.T., Morell, R.J., Elfenbein,J.L., Fisher, R.A., Leal, S.M., Smith, R.J., and Friderici, K.H.(2003). Mutations in the gamma-actin gene (ACTG1) are associatedwith dominant progressive deafness (DFNA20/26). Am. J. Hum.Genet. 73: 1082–1091.

Zimmer, A., Lang, D., Richardt, S., Frank, W., Reski, R., andRensing, S.A. (2007). Dating the early evolution of plants: Detectionand molecular clock analyses of orthologs. Mol. Genet. Genomics278: 393–402.

Ancient Functional Competence of Actin 2057

DOI 10.1105/tpc.111.095281; originally published online May 15, 2012; 2012;24;2041-2057Plant Cell

Muthugapatti K. Kandasamy, Elizabeth C. McKinney, Eileen Roy and Richard B. MeagherDevelopment with Protists

Plant Vegetative and Animal Cytoplasmic Actins Share Functional Competence for Spatial

 This information is current as of September 29, 2020

 

Supplemental Data /content/suppl/2012/05/15/tpc.111.095281.DC2.html /content/suppl/2012/05/15/tpc.111.095281.DC1.html

References /content/24/5/2041.full.html#ref-list-1

This article cites 59 articles, 22 of which can be accessed free at:

Permissions https://www.copyright.com/ccc/openurl.do?sid=pd_hw1532298X&issn=1532298X&WT.mc_id=pd_hw1532298X

eTOCs http://www.plantcell.org/cgi/alerts/ctmain

Sign up for eTOCs at:

CiteTrack Alerts http://www.plantcell.org/cgi/alerts/ctmain

Sign up for CiteTrack Alerts at:

Subscription Information http://www.aspb.org/publications/subscriptions.cfm

is available at:Plant Physiology and The Plant CellSubscription Information for

ADVANCING THE SCIENCE OF PLANT BIOLOGY © American Society of Plant Biologists