REVIEW

GENETIC TOOLBOX

Tetrahymena as a Unicellular Model Eukaryote: Genetic and Genomic Tools

Marisa D. Ruehle,*,1 Eduardo Orias,† and Chad G. Pearson* *Department of Cell and Developmental Biology, University of Colorado Anschutz Medical Campus, Aurora, Colorado, 80045, and †Department of Molecular, Cellular, and Developmental Biology, University of California, Santa Barbara, California 93106

ABSTRACT Tetrahymena thermophila is a ciliate model organism whose study has led to important discoveries and insights into both conserved and divergent biological processes. In this review, we describe the tools for the use of Tetrahymena as a model eukaryote, including an overview of its life cycle, orientation to its evolutionary roots, and methodological approaches to forward and reverse genetics. Recent genomic tools have expanded Tetrahymena’s utility as a genetic model system. With the unique advantages that Tetrahymena provide, we argue that it will continue to be a model organism of choice.

KEYWORDS Tetrahymena thermophila; ciliates; model organism; genetics; amitosis; somatic polyploidy

ENETIC model systems have a long-standing history as cost-effective laboratory handling, and its accessibility to both Gimportant tools to discover novel genes and processes in forward and reverse genetics. Despite its affectionate reference cell and developmental biology. The ciliate Tetrahymena ther- as “pond scum” (Blackburn 2010), the beauty of Tetrahymena mophila is a model system that combines the power of for- as a genetic model organism is displayed in many lights. ward and reverse genetics with a suite of useful biochemical Tetrahymena has a long and distinguished history in the and cell biological attributes. Moreover, Tetrahymena are discovery of broad biological paradigms (Figure 2), begin- evolutionarily divergent from the commonly studied organ- ning with the discovery of the first microtubule motor, dynein isms in the opisthokont lineage, permitting examination of (Gibbons and Rowe 1965). Others include the Nobel Prize both unique and universally conserved biological processes. winning discoveries of catalytic RNA (Kruger et al. 1982) and Here we highlight the advantages of Tetrahymena as a model telomere structure and telomerase (Greider and Blackburn system, such as its unique and easily manipulated life cycle, 1985). The first histone-modifying enzyme (histone acetyl that have contributed to important discoveries. The growing transferase) and its role as a transcription factor were discov- suite of molecular-genetic and genomic tools described here ered in Tetrahymena (Brownell et al. 1996), which gave birth provides a system to couple gene discovery to mechanistic to the “histone code” and the field of epigenetic control of dissections of gene function in the cell. gene expression by chromatin modification. The role of small T. thermophila (Figure 1) is a ciliate model organism interfering RNAs in heterochromatin formation and the mas- whose study has led to fundamental biological insights cov- sive, programmed excision of transposon-related DNA from ering the central dogma and beyond. Indeed, this single- the somatic genome (Mochizuki et al. 2002; Taverna et al. celled, motile eukaryote provides the tools and techniques 2002) is another major Tetrahymena contribution. These fun- not only for novel gene discovery, but also for unveiling the damental discoveries in Tetrahymena have helped usher in ’ important molecular mechanisms behind those genes func- the modern era of molecular and cellular biology. Many of the ’ tions. As such, Tetrahymena s utility as a genetic model reasons why Tetrahymena was an advantageous system for organism is revealed by its short life cycle, easy and these groundbreaking discoveries are the same that make Tetrahymena useful today and in the future, augmented by the ever-expanding toolkit available to Tetrahymena Copyright © 2016 by the Genetics Society of America doi: 10.1534/genetics.114.169748 researchers. Manuscript received March 1, 2016; accepted for publication April 8, 2016. In this review, we discuss major biological questions to 1Corresponding author: Department of Cell and Developmental Biology, University of Colorado Denver School of Medicine, 12801 E. 17th Ave., Mail Stop 8108, Aurora, which T. thermophila is amenable and the genetic tools avail- CO 80045. E-mail: [email protected] able to answer them. It begins with the general biology of

Genetics, Vol. 203, 649–665 June 2016 649 genome consists of 200 different chromosomes, each main- tained at a ploidy of 45 during the G1 phase of the cell cycle. The MIC is transcriptionally silent, and all gene expression is driven from the MAC DNA. Despite the fact that Tetrahymena are unicellular organisms, this germline/soma separation is reminiscent of metazoans where distinct germ cells and so- matic cells are maintained. Germline/soma separation is ex- tremely rare among unicellular eukaryotes, though it is also found in certain Foraminifera (in the Rhizarian evolutionary group). To replicate these distinct genomes and to maintain ge- netic diversity, the T. thermophila life cycle employs both asex- ual, vegetative cell division and sexual reorganization through a process called conjugation. During vegetative division, the MAC divides amitotically (random chromosome segregation), the MIC divides mitotically, and binary cell fission produces two daughter cells. However, under conditions of starvation, pairs of sexually mature cells with different mating types un- dergo conjugation. T. thermophila cells can have one of seven different mating types, and each mating type can conjugate with another mating type, but not with itself (Elliott and Gruchy 1952; Nanney and Caughey 1953). Conjugation in- volves highly conserved eukaryotic processes: meiosis, game- Figure 1 Tetrahymena cell image, illustrating its crystal-like organization togenesis, and gamete nucleus fusion to form the fertilization of ciliary units. A single Tetrahymena cell labeled for basal bodies (a-centrin, (or “zygote”) nuclei of progeny cells (Figure 3). The fertiliza- et al. fi red) (Stemm-Wolf 2005), kinetodesmal bers (5D8, green) (Jerka- tion nucleus divides mitotically and then differentiates into a Dziadosz et al. 1995), and DNA (Hoerscht-33342, blue). Note that basal bodies are packed together in the four ciliary “membranelles” of the oral new MIC as well as a new MAC, and the parental MAC is apparatus, for which the genus is named. Bar, 10 mm. eliminated. Conjugation has important consequences: Mende- lian genetic inheritance and increasing genotypic diversity.The genome’s maintenance and expression depend on widely con- Tetrahymena and its unique advantages as an experimental served mechanisms and pathways across all lineages of life, model system. We then describe both forward and reverse making Tetrahymena a useful organism in which to study these genetic strategies in Tetrahymena to facilitate gene discovery universal processes. and to interrogate the mechanistic underpinnings of those Evolutionary context genes. Ultimately, we seek to engage future researchers by describing the wealth of experimental advantages (both his- As ciliates, Tetrahymena are a part of the Alveolate group of torical and modern) that Tetrahymena can provide and pre- the Stramenopile–Alveolate–Rhizarian (SAR) lineage (Fig- view its promising future. ure 4). Dinoflagellates and apicomplexans are also a part of the Alveolate group, but are more closely related to each other than to the ciliates. Thus, there is a large evolutionary Tetrahymena Biology distance from ciliates to their sister Alveolate groups, and an Tetrahymena are unicellular, ciliated eukaryotes that live in even greater distance to other commonly used eukaryotic model fresh water over a wide range of conditions. In the wild, organisms which, like humans, are a part of the opisthokont Tetrahymena feed on bacteria, but laboratory strains typically group (yeast, worms, flies, mice, and many others). There is a live as axenic cultures in nutrient-rich media (derived from noticeable dearth of model organisms outside of the opisthokont tissue extracts) or chemically defined media. Populations group, making research of nonopisthokonts such as Tetrahy- grow quickly, with cells dividing every 2–3 hr under optimal mena an attractive means to explore the remarkable diversity conditions. Each cell is large, 30–50 mminlength,mak- of eukaryotic cell biology (Lynch et al. 2014). Nevertheless, ing them ideal for light and electron microscopy-based Tetrahymena utilize many universally conserved eukaryotic investigations. processes, making them also useful for illuminating these Tetrahymena, like all ciliates, are nuclear dimorphic, conserved features (Briguglio and Turkewitz 2014; Lynch which means that the germline genome and the somatic ge- et al. 2014). nome exist within two separate nuclei in one cell: the micro- Tetrahymena’s closest model organism relative (though nucleus (MIC) and the macronucleus (MAC), respectively not its closest relative outright) is Paramecium, which, along (Table 1, Table 2, and Figure 1). The diploid MIC genome with Tetrahymena, is a part of the Oligohymenophorea class consists of five pairs of chromosomes, whereas the MAC (Figure 4, bottom). However, based on deviations in small

650 M. D. Ruehle, E. Orias, and C. G. Pearson and later T. pyriformis syngen 1, before receiving its current name (Nanney and McCoy 1976). Isogenic, inbred lines of T. thermophila have been derived (Allen 1967). By conven- tion, most genetic analysis and genome sequencing has used T. thermophila inbred strain B cells, while most genetic map- ping has utilized DNA polymorphisms between inbred strains B and C3. The sibling Tetrahymena species found in the original T. pyriformis group provide an invaluable resource for com- parative genomic analyses, as evolutionary DNA sequence conservation is a useful predictor of biological function under natural selection (Romero and Blackburn 1991; Coyne and Yao 1996). It is particularly useful for the detection of lineage- specific genes of unknown function and to identify conserved DNA and protein sequences within genes. Lineage-specific gene family expansions are interesting properties of any sequenced genome, as they can provide clues to the importance of particular cellular processes. In T. thermophila, gene family expansion has occurred mainly by localized gene duplication (Eisen et al. 2006). This is in contrast to Paramecium tetraurelia where two recent whole- genome duplications (WGDs) occurred after the divergence from the Tetrahymenine ciliates contributing to its gene fam- ily expansion (Aury et al. 2006). It has been suggested that an older WGD, clearly detectable in the Paramecium genome occurred prior to the divergence of Paramecium and Tetrahy- mena (Aury et al. 2006). However, no remaining evidence of a WGD in the Tetrahymena genome has yet been found. Ma- jor gene family expansions in the Tetrahymena genome in- volve genes whose products function in “sensing and responding to environmental cues (e.g. protein kinases, membrane transporters), mobilizing resources from the en- vironment (e.g. membrane transporters and traffic compo- nents, proteases), and maintaining complex cell structure and movement (e.g. microtubule components, motors, and regulators, membrane trafficcomponents)” (Eisen et al. 2006). The function of closely related gene family members Figure 2 Tetrahymena areas of research. Fields of study that Tetrahy- can be difficult to dissect experimentally because of redun- mena has impacted. dancy or significant functional similarity. Ciliates have evolved lineage-specific changes in the nu- subunit ribosomal RNA (rRNA) sequences between Parame- clear genetic code. Tetrahymena uses only a single codon cium and Tetrahymena, the two likely shared a common (UGA) to terminate MAC-encoded protein synthesis. The ancestor several hundred million years ago and their evolu- other stop codons in the universal genetic code, UAA and tionary distance exceeds that between rat and brine shrimp UAG, encode glutamine amino acids, along with the canon- (Greenwood et al. 1991; Frankel 2000). Thus, while much of ical glutamine codons, CAA and CAG. While the variant ge- the biology of Tetrahymena has been illuminated by studies netic code poses no special problem when expressing in Paramecium and vice versa, differences between them are heterologous genes in Tetrahymena, the expression of Tetra- expected and often observed (Frankel 2000). hymena protein coding genes in heterologous systems re- T. thermophila belongs to a group of Tetrahymena species quires first mutating UAA and UAG codons to either CAA or that are morphologically indistinguishable from one another. CAG. This hurdle, as well as that of optimizing codon usage These species were originally lumped together as a single (described later), is generally addressed by gene synthesis. species called T. pyriformis. The discovery of Tetrahymena Two nuclei in one cell mating types allowed dissection of the group based on sexual isolation and led to the recognition of T. thermophila and Tetrahymena cells have a MIC that houses the germline ge- other sibling species as distinct Tetrahymena species. Along nome, and a MAC that contains the somatic genome. Al- the way, T. thermophila was first called T. pyriformis variety 1 though the separation of germline and soma sounds

Genetics Toolbox Review 651 Table 1 Tetrahymena thermophila glossary Term Definition

Chromosomal breakage site (CBS) MIC chromosome location occupied by the highly conserved 15-bp chromosome breakage sequence, which is the necessary and sufficient DNA element for programmed chromosome fragmentation during MAC development. Cytogamy A mating strategy in which gamete pronucleus exchange between conjugants is blocked; the two sister gamete pronuclei in each conjugant instead fuse to one another (double self-fertilization), generating whole-genome homozygous fertilization nuclei. The postzygotic events are normal. Fertilization nucleus The nucleus that forms as a result of gamete pronuclear fusion during conjugation. In Figure 3F the fertilization nuclei are depicted as half white, half black small circles in each conjugant cell. Also referred to as the zygote nucleus. Genomic exclusion A self-fertilization mating strategy to generate a whole-genome homozygote that uses a cross to a strain with a defective MIC (star strain). Heterokaryon A cell in which the MIC and MAC genotypes are different. Homokaryon A cell in which the MIC and MAC genotypes are the same. Karyonides The four cells formed by the first fission of the two exconjugants of a pair, represented in Figure 3J. Each karyonide acquires an independently differentiated new MAC. MAC Macronucleus. MAC anlagen The newly developing MACs mitotically derived from zygote nucleus during the postzygotic steps of conjugation. MAC KO Gene knockout in the macronucleus. MAC maturation or development The process, during conjugation, whereby an initially diploid MAC anlage completes its structural and functional differentiation from the MIC. Meiotic segregant A sexual progeny cell whose MIC and MAC are derived from meiotic products of a double heterozygote during conjugation; its genotype is classified as parental or recombinant. MIC Micronucleus. MIC KO Gene knockout in the micronucleus. Nullisomic A cell whose MIC lacks (is null for) both copies of an entire chromosome or chromosome arm. Such strains show no phenotype when they are “covered” by a diploid MAC. Phenotypic assortment The phenomenon in which a MAC initially heterozygous for an allelic variant becomes homozygous through successive rounds of amitosis during vegetative multiplication. Postzygotic mitoses The two divisions of the fertilization nucleus that generate the two MIC and MAC anlagen during conjugation. Prezygotic mitosis Division of the surviving haploid meiotic product that generates the migratory and stationary gamete pronuclei during conjugation. Segmental deletion Loss of a large MIC chromosome segment. Deletion homozygotes show no phenotype when they are covered by a diploid MAC. Star strain A cell having a severely defective MIC, which fails to generate any meiotic products during conjugation. Uniparental cytogamy (UPC) A mating strategy to generate a whole-genome homozygote by inducing self-fertilization in a cross between a normal and a star strain. The exconjugant derived from the star strain dies.

familiar to us metazoans, the presence of two different nu- MAC genome (Mochizuki et al. 2002; Duharcourt et al. clear genomes in the same cell is foreign. How does the MAC 2009; Fass et al. 2011; Schoeberl et al. 2012). Most of the develop from the MIC, how are they similar or different, and eliminated sequences are repetitive elements likely derived what experimental advantages does nuclear dimorphism from transposons (genomic parasites that randomly insert in confer? the genome, causing deleterious genomic instability if not The diploid MIC genome consists of five pairs of chromo- silenced). Thus, eliminating these sequences from the MAC somes and its size is estimated to total 220 Mb of DNA protects the expressed genome of the organism. The newly (Table 2) (Yao and Gorovsky 1974). When the newly forming generated MAC chromosomes are then endoreplicated (rep- MAC (MAC anlagen) develops during conjugation (Figure 3, licated multiple times without nuclear division), leading to H and I) these 10 chromosomes are fragmented at specific the production of 45 copies of each chromosome per MAC. sites called chromosomal breakage sequences (CBSs) and A unique and important exception to this copy number is the they lose some internal sequences (Yao et al. 1987), collec- chromosome that encodes the 35S rRNA precursor gene tively referred to as internal eliminated sequences (IESs). (Tetrahymena homolog of the 45S rRNA precursor gene in Fragmentation and DNA elimination generates many new, other eukaryotes). This is the smallest MAC chromosome smaller chromosomes. The CBS is a conserved 15-bp motif, (20 kb) and is present at 9000 copies per MAC. which is necessary and sufficient for DNA breakage (Yao et al. Mating type determination in the sexual progeny occurs 1987). Telomeric repeat addition occurs at the ends of the during MAC development (Figure 3, H and I). It occurs in new chromosomes. The locations of IESs are marked through each of the four genetically identical developing MACs of an elegant small RNA-dependent pathway, which ensures a conjugating pair independently of one another or of that these segments of DNA are not incorporated into the the parental mating types. This is because mating type

652 M. D. Ruehle, E. Orias, and C. G. Pearson Table 2 Genome statistics amitosis as a result of unequal chromosome segregation during No. of chromosomes Ploidy Size (Mb) MAC division (Figure 5B). Daughter cells stochastically receive a greater or lesser number of copies of an allele, and MACs MIC 5 2 220a MAC 200 45 105 ultimately become pure for either allele during subsequent cell divisions (Orias and Flacks 1975; Nanney and Preparata 1979; a A more accurate estimate will become available upon the pending publication of the MIC genome sequencing project. MerriamandBruns1988).Thesteady-staterateofassort- ment is mathematically represented as 1 / (2n 2 1) per fission (Schensted 1958), where n is the number of chromo- determination is based on stochastically selected, alternative some copies just after MAC division (45 in this case). In the DNA rearrangements at the mating type locus of the devel- wild, this phenomenon permits rapid somatic adaptation to oping MAC (Cervantes et al. 2013). This process is unrelated selective pressures, and in the lab, it is useful for isolating to chromosome fragmentation or IES excision. A cell’smat- mutations or phenotypes of interest. Despite the random as- ing type is faithfully inherited during vegetative division. Prog- sortment of chromosomes in the MAC, a mechanism to control eny cells can mate only after a period of sexual immaturity chromosome copy number has been inferred (Doerder and that lasts 40–70 cell divisions after conjugation. The molec- DeBault 1978), although its molecular basis remains unknown. ular basis of this maturation is unknown. The mating type locus is already present in multiple copies at the time of mating type determination in the developing Two genetic systems in one cell MAC. Therefore, more than one mating type can be expressed The organization and behavior of the germline vs. the somatic in a progeny cell line immediately after conjugation. This nuclei define two types of genetic systems at work in the life of condition is typically resolved by the time the mixed mating every cell. The diploid MIC lineage undergoes meiosis and fer- type progeny cells reach sexual maturity, when most cells tilization to bestow Mendelian genetic inheritance across sexual have become fixed for a single mating type by phenotypic generations. On the other hand, during vegetative cell division, assortment. Sexually mature cells whose MAC remain mixed theMAC,withfragmentedchromosomes that undergo amitotic are called “selfers,” because they can pair with other mem- distribution, generates a distinct genetic phenomenon called bers of the same clone when starved. If undetected, selfers phenotypic assortment. This is akin to the genetics of high copy can present a challenge to downstream experiments that re- number bacterial plasmids. The interplay of these two genetic quire mating. If it is absolutely necessary to work with a selfer systems generates natural versatility that can be exploited to line, additional vegetative cell divisions permit the comple- great experimental advantage in the laboratory. Because tion of phenotypic assortment and allow mating type to be- the less familiar genetic phenomena associated with amitotic come fixed in a subclonal population of these cells. division are critical to mutational and other genetic approaches, The remarkable genetic features made possible by germline/ they are examined in more detail in the next section. soma differentiation in Tetrahymena provide a variety of ex- perimental advantages. First, recessive mutations within a het- Amitosis and phenotypic assortment erogeneous MAC can ultimately come to complete expression During vegetative S phase, MAC DNA is replicated once because phenotypic assortment will enhance the number of (Andersen and Zeuthen 1971) and then is distributed to daugh- recessive mutant alleles and reduce the number of dominant ter MACs by amitosis. In this process, the MAC chromosomes wild-type alleles in a subset of the progeny. Second, whole- are not equally segregated to daughter cells; rather chromo- genome homozygotes can be generated quickly and easily by some segregation is stochastic. Familiar features of mitosis in self-fertilization methods. As a consequence, recessive muta- animals and plants, such as bipolar spindle formation, extreme tions present in the MIC can be isolated as efficiently as if the chromosome condensation, and microtubule attachment to germline were haploid and are expressed in the MAC without centromeres, are all missing during MAC division; thus the assortment. Third, production of heterokaryon cells (cells kinetochore-based, equal chromatid separation in normal mito- whose MIC and MAC genotypes are different; Figure 6) allow sis does not play a role in MAC division (Flickinger 1965; lethal mutations to be maintained in the MIC during vegetative Cervantes et al. 2006). Microtubules are nonetheless required growth until induction of mating. Similarly, nullisomics and during amitosis to elongate the MAC before it divides (Williams segmental deletion homozygotes can be maintained as hetero- and Williams 1976; Kushida et al. 2011). The molecular mech- karyons, which allow for the physical mapping of mutations, anisms that govern amitosis remain to be elucidated. DNA polymorphisms, or MIC-limited DNA elements, such as Becauseamitosisdoesnotinvolvethehigh-fidelity, equal CBSs. These are advantages that Tetrahymena researchers segregation of sister chromatids into daughter cells as in mito- commonly exploit to identify and study genes of interest. sis, a phenomenon called phenotypic assortment is observed dur- ing Tetrahymena vegetative growth (Figure 5A) (Allen and Why Use Tetrahymena as a Genetic Model Nanney 1958; Doerder et al. 1975; Orias and Flacks 1975; Organism? Nanney and Preparata 1979). Phenotypic assortment is the abil- ity of a heterozygous MAC to come to homozygosity or lead to Model systems are typically chosen based on their utility in fixation of either allele. This occurs over multiple rounds of identifying new avenues of research or in contributing to

Genetics Toolbox Review 653 Figure 3 Tetrahymena conjugation. The life cycle of Tetrahymena includes a sexual phase called conjugation. Cells are represented as ovals with a micronucleus (small circle) and a macronucleus (large circle) whose DNA content (ploidy) is indicated. Images of cells at corresponding steps are shown on the periphery of the illustration and are lettered accordingly. To begin the sexual phase of the life cycle, two cells of different mating types (A), homozygous for black and white alleles, respectively, undergo pairing (B). Completion of two rounds of meiosis (C) leads to the production of four haploid products (half circles, indicated as 1n). One of these meiotic products is positioned at the anterior cytoplasm of each conjugant, while the remainingthree are targeted for elimination (red outline) at the conjugant’s posterior end. Subsequently, mitosis of the surviving meiotic product generates two gamete pronuclei (D). Each migratory pronucleus is transferred to the opposite conjugant in a process called pronuclear exchange (E). The incoming migratory pronuclei fuse with the stationary pronuclei (pronuclear fusion), restoring the diploid character of the MIC and thus generating the fertilization (or zygote) nucleus (F). Thus, each fertilization nucleus gets one haploid genome from each parent (black and white semicircles). The fertilization nucleus undergoes two postzygotic mitoses (G), leading to the production of four genetically identical diploid nuclei in each conjugant. (H) The two anterior nuclei (MAC anlagen) develop into new MACs, while the two posterior nuclei become the new MICs. New MAC maturation involves ploidy increase and programmed DNA rearrangement (see text). The parental MACs are eliminated apoptotically (red outline) and contribute no DNA to the progeny. (I) The exconjugant cells separate and one of the two MICs is eliminated. When the exconjugants divide, one of the two fully developed MACs and a mitotic copy of the surviving MIC in each exconjugant are segregated to their daughter cells, called karyonides because each one gets an independently developed MAC (J). Having completed the conjugation process, these cells must sexually mature through vegetative cell division before they can conjugate. the mechanistic understanding of important biological genomes sequenced. The germline MIC exhibits Mendelian problems. Tetrahymena have the advantages of a “do-it- genetics, and drug resistance markers exist for positive all” model system; this is especially true as a genetic model selection. This suite of genetic advantages yields an effec- system. Tetrahymena grow fast (2- to 3-hr doubling tive model system to complement its powerful biochemical time),canundergolargescaleandsynchronousmatings, and cytological attributes. Indeed, both forward and re- genetic markers have been mapped, and its two nuclear verse genetic tools are well-developed in Tetrahymena.

654 M. D. Ruehle, E. Orias, and C. G. Pearson Figure 4 Evolutionary relationship of Tetrahy- mena to other eukaryotes. (Top) Eukaryotic su- pergroups and their subgroups. Adapted from Lynch et al. (2014). (Bottom) Expanded tree of the Alveolates showing the lineage leading to Tetrahymena (red lines) (Cavalier-Smith 1993). Line lengths do not represent evolutionary distances.

Furthermore, the National Tetrahymena Stock Center, lo- genes without associated phenotypes. To then express the cated at Cornell University, maintains and readily makes mutant phenotype, genetically identical cells are mated to available many cell lines as well as plasmids, protocols, and one another and all progeny will express the mutation of other resources to facilitate these studies. interest. These strategies allow for sophisticated reverse ge- netic studies. Forward and reverse genetics Advantages to repetitive elements Forward genetic strategies have been useful for generating Tetrahymena mutants. Mutagenesis screens with accom- An important advantage to Tetrahymena is their large num- panying phenotypic screens have identified defects in bers of repeating structural elements, such as the repeating biological processes including cortical patterning, metab- cortical architecture composed of polarized individual cili- olism, motility, cell cycle progression, RNA positioning, ary units that are essential for cellular motility (Figure 1). drug resistance, phagocytosis, and exocytosis (Bruns and This is a defining feature of ciliate cellular organization. Sanford 1978; Ahmed et al. 1998; Bowman et al. 2005). Motility is a simple readout for experiments relating to cor- Such mutations can then be mapped to the genome. It was tical architecture, requiring minimal technological tools. previously difficult to identify the mutations associated with Additionally, clathrin-dependent endocytosis is organized a phenotype due to difficulties in producing genome- in an array at the cell cortex and is readily studied in wide libraries to rescue mutant phenotypes. However, Tetrahymena (Elde et al. 2005). The relatively large next generation sequencing (NGS) is now an effective Tetrahymena cells (20 3 50 mm) with repeating elements tool to rapidly identify mutations in Tetrahymena and enables rapid and high-quality cytological work using both in other genomes (Galati et al. 2014; Schneeberger light and electron microscopy and provides copious material 2014). The development of NGS to identify mutations cre- for biochemical and structural studies of these repeating ated in Tetrahymena forward genetic screens greatly in- elements. creases the power of this system for new mechanistic The Tetrahymena community has developed molecular research. tools and microscopy techniques to visualize Tetrahymena’s To study the functions of known genes, Tetrahymena is also diverse cellular processes. Protein tags allow for biochemical amenable to reverse genetic strategies by the introduction of analyses and live and fixed cell visualization of protein local- precisely targeted and heritable gene knockouts (KOs) and ization and cellular structure. The regular, crystal-like repe- disruptions. Moreover, potentially deleterious genomic tition of structural elements enables signal amplification knockouts and gene mutations can be introduced into and in these localization studies, resulting in rapid and high- propagated in the germline while maintaining a normal MAC resolution detection of cortical architecture using light and genome. Thus, cells may propagate mutations in essential electron microscopy.

Genetics Toolbox Review 655 Figure 5 Amitotic macronuclear division and phenotypic assortment. (A) Phenotypic assortment: During vegetative growth, the MIC undergoes mitotic division, whereas the MAC divides amitotically, meaning that chromosome copies are randomly distributed to the daughter macro- nuclei. When the MAC is heterozygous, this leads to unequal segregation of allelic genetic information, a phenomenon called phenotypic assortment, illustrated by the all-white or all-black MACs, the two alternative endpoints of the assortment process. The mitotically dividing MIC remains heterozygous. (B) Amitosis leads to phenotypic assortment: A vegetatively growing cell whose MAC is represented in (a) one of the chromosome copies has a mutant allele, represented by a star. (For simplicity of illustration, only one MAC chromosome is shown, and the G1 ploidy is 4 instead of 45 copies.) One possible sequence of events is illustrated. (b) The number of copies has doubled during S phase. The two mutant copies can be distributed in one of two equally probable ways: one mutant copy goes to each daughter MAC (no change, not shown), or both mutant copies go to the same daughter (c), so that the other daughter becomes fixed for the wild-type allele. Additional cell cycles (d and e) generate MACs fixed for the mutant allele (f). Once a cell with a pure MAC is generated, every MAC in the resulting subclone remains fixed. Thus, the fraction of cells with pure MACs continuously increases in the clone. At steady state, the probability that a cell with a mixed MAC will generate a daughter with a pure MAC is 1 / (2n 2 1), where n is the ploidy at G1 stage, or 0.011/fission for n = 45. The initial allele ratio in the progenitor cell’s MAC (a) is called the input ratio. The output ratio is the ultimate ratio of cells whose MACs have become pure for either allele after many fissions. In the absence of selection, the output ratio should be identical to the input ratio. Heterozygous MACs produced by crossing, as in Figure 3, will generally have an 1:1 input ratio. In contrast, when MAC transformation is carried out, the initial transformant is likely to have an input ratio highly biased against the transforming allele, which is compensated for by introducing selection for the transformed allele.

In silico toolbox compiles these data into a user-friendly, searchable format. Researchers new to the field will appreciate the community- Informatics tools exist to stimulate the study of the Tetrahy- annotated genome page where genes are assigned with Gene mena genome and its expression (Table 3). These include the MIC and MAC genome sequences from the Broad Institute Ontology (GO) terms, relevant literature references, and links “ and The Institute for Genome Research (TIGR: now the to GenBank entries. The site also maintains a Textpresso ” J. Craig Venter Institute), respectively, a list of annotated for Tetrahymena function for full text searches. The combi- genes, transcriptome-wide RNA sequencing (RNA-seq) data nation of genomic resources with the genetic power of the at different life cycle stages (http://tfgd.ihb.ac.cn/), compar- organism and its accompanying molecular and biochemical ative genomes of organisms closely related to T. thermophila, tools, primes Tetrahymena for elucidating additional funda- and a community database (TGD wiki; ciliate.org) that mental biological processes.

656 M. D. Ruehle, E. Orias, and C. G. Pearson Figure 6 Useful heterokaryons. (A) Drug resistance heterokaryon: often used to eliminate unmated parental cells (drug sensitive) and recover only true progeny (drug resistant) in a mass cross. (B) Heterokaryon for a lethal mutation: used to propagate a homozygous lethal mutation in the MIC, while “covered” by a wild-type MAC. The phenotype is expressed when two such cell lines are crossed to one another. (C) Nullisomic heterokaryons: the lethal MIC genotype is the absence of both copies of one MIC chromosome or chromosome arm (arrow); such heterokaryons are useful for genetically mapping mutations and DNA polymorphisms to chromosome locations. (D) Gene knockout: the lethal mutation is a deleterious or lethal gene KO, in which the coding sequence of the KO’s gene has been replaced by a cassette expressing drug resistance (to paromomycin, related to neomycin). The KO comes to expression in the progeny when such heterokaryons are crossed while the neo cassette allows elimination of wild-type parental and untransformed progeny cells.

Genetic Studies in Tetrahymena Mutagenesis: Random mutations are induced by a brief exposure to a chemical mutagen (methylmethane sulfonate Forward genetic studies identify both important biological or nitrosoguanidine) during vegetative growth (Pennock processes and the genes associated with them (Forsburg 2000). Random mutagenesis leads to dominant and recessive 2001; Patton and Zon 2001; Page and Grossniklaus 2002; mutations in the somatic and/or germline genomes. Only Casselton and Zolan 2002; Jorgensen and Mango 2002; St mutations present in the MAC lead to immediately observ- Johnston 2002; Shuman and Silhavy 2003; Kile and Hilton able phenotypes because the MIC is transcriptionally inac- 2005; Candela and Hake 2008). These approaches have been tive. Germline mutations only come to expression after the combined with reverse genetics methods to investigate de- mutagenized cells are crossed, and the progeny MACs carry tails of how the discovered genes function. In Tetrahymena, the mutation. Using a mating strategy that generates homo- this powerful approach has illuminated the biology of cilia, zygous progeny (such as uniparental cytogamy; Figure 7B) fi vesicular traf cking and exocytosis, telomeres, RNA enzymes, even recessive mutations can be phenotypically observed af- and chromatin, to name a few examples (Figure 2). Many ter a single cross (Karrer 2000). Tetrahymena genes are conserved with humans; indeed hu- mans share more orthologs with Tetrahymena than with Mutant isolation: New mutations in the MAC are immedi- other unicellular eukaryotes with a closer phylogenetic re- ately expressed and their phenotypes become observable lationship (Eisen et al. 2006), making these methods a pow- when thenumber of chromosomal copies carrying themutation erful way to study ciliate and human biology alike. increasessufficientlyby phenotypic assortment.However,these mutations are lost upon mating when the old MAC is destroyed, Forward genetics: unbiased gene discovery which makes MAC mutations less accessible to traditional ge- Forward genetic screens employ random mutagenesis of the netic strategies. MIC mutations, on the other hand, can be genome, followed by screening for a mutant phenotype of propagated and used for traditional Mendelian genetic exper- interest and then identifying the mutated gene(s). The iments. However, expression of the new mutation occurs only method makes no preconceptions about what genes can be after it is transmitted to a new MAC following mating. identified. Below we describe the general strategies and work- The isolation of recessive MIC mutations creates a special flow for forward genetics approaches in Tetrahymena. problem because a simple cross of the mutagenized cell to

Genetics Toolbox Review 657 Table 3 In silico toolbox and other resources Resource Source Location Reference

Genomics resources Genome browser TGD: Tetrahymena genome http://www.Ciliate.org Stover et al. 2006 database NCBI genome http://www.ncbi.nlm.nih.gov/genome/222 Eisen et al. 2006 Functional genomics TetraFGD: Tetrahymena Functional http://tfgd.ihb.ac.cn/ Xiong et al. 2013 Genomics Database

Experimental resources Strains Tetrahymena stock center https://tetrahymena.vet.cornell.edu/ Cassidy-Hanley 2012

Methods resources Asai D. J. and J. D. Forney (eds.) 1999 Tetrahymena thermophila. Methods in Cell Biology, vol. 62

Collins K. (ed.) 2012 Tetrahymena thermophila, 1st edition. Methods in Cell Biology, vol. 109

Teaching resources ASSET teaching resource ASSET: Advancing secondary https://tetrahymenaasset.vet.cornell.edu/ Smith et al. 2012 science education thru Tetrahymena Tetrahymena SUPRDB SUPRDB: Student/Unpublished http://suprdb.org/ Wiley and Stover 2014 Results Ciliate genomics Consortium Community/Wiley http://faculty.jsd.claremont.edu/ewiley/

Biotechnology resources Cilian http://www.cilian.de/ (website in German) Tetragenetics http://www.tetragenetics.com/ Tetratox http://www.vet.utk.edu/TETRATOX/index.php Schultz 1997

Additional resources Tetrahymena history http://www.life.illinois.edu/nanney/index.html Community listserv https://listserv.uga.edu/cgi-bin/wa?A0=ciliatemolbio-l allow expression of the mutation in progeny MACs will result density gradients for feeding defects have been successfully in heterozygous progeny showing the wild-type pheno- employed (Nilsson and van Deurs 1983; Hünseler et al. type. This problem is elegantly resolved by inducing self- 1987; Pennock et al. 1988; Tiedtke and Rasmussen 1988). fertilization, where mutations are made homozygous for the mutant allele in both the MIC and MAC genomes allowing Sorting the mutant collection: Identical mutant phenotypes for expression of the mutant phenotype. Two approaches are can obviously result from mutations in more than one gene. To commonly used to obtain Tetrahymena whole-genome homo- tease apart biological pathways following a screen, Tetrahy- zygotes: genomic exclusion and uniparental cytogamy (UPC) mena is amenable to complementation tests to identify (Figure 7) (Allen 1967; Cole and Bruns 1992). UPC is the whether two independent mutations are in the same or differ- most efficient way to accomplish self-fertilization for mutant ent genes (Frankel et al. 1976; Gutiérrez and Orias 1992). In isolation purposes. Recessive mutations are thus isolated as the simplest scenario, homozygous cell lines of two mutants efficiently as in other microbes with a haploid genome. The of independent origin are mated to one another and pheno- progeny are then screened (or in favorable cases selected) for types of the progeny are determined. Rescue of the wild-type the desired mutant phenotype. phenotype is expected if the mutations are in different genes. More complex genetic interactions like intragenic complemen- Screening mutants: Phenotypic screening of mutants en- tation and nonallelic, noncomplementation are also accessible tails the rapid assessment of cells for classes of phenotypes to these analyses. Complementation studies thus establish (Forsburg 2001; Patton and Zon 2001; Casselton and Zolan the minimum predicted number of genes involved in a path- 2002; Jorgensen and Mango 2002; Page and Grossniklaus way and potential genetic interactions. 2002; St Johnston 2002; Shuman and Silhavy 2003; Kile and Hilton 2005; Candela and Hake 2008). These simple Identifying mutations by NGS: In the era of high-throughput assays are designed to sort through hundreds to thousands NGS, comparative genomics is used to map mutations at of mutant cells. Creative strategies exist (or may need to be nucleotide-level resolution in a “mapping-by-sequencing” developed) to identify a mutant with a phenotype of inter- fashion (Schneeberger 2014). This has been successful in est. Notably, swimming assays for cilia defects, fluorescent various model organisms, including the ciliates Tetrahymena detection of secreted proteases for exocytosis defects, and and Paramecium (Sarin et al. 2008; Blumenstiel et al. 2009;

658 M. D. Ruehle, E. Orias, and C. G. Pearson Figure 7 Genetic approaches to generate whole-genome homozygotes from a heterozygote. Symbols are as in Fig- ure 3. Asterisk denotes a mutation or allele of interest, such as a drug resistance gene. (A) Genomic exclusion. A cell containing a heterozygous MIC (represented by black and white) for a mutation in the MIC (asterisk) is crossed to a cell line with a defective MIC—called a “star” strain—which is incapable of contributing germline DNA to sexual progeny (round I). At meiosis, the remnant of its MIC disintegrates without cytological trace (stage b). The MIC of the nonstar mate undergoes normal meiosis, ga- metogenic mitosis, and gamete pronucleus transfer (stages c and d). The gamete pronuclei diploidize and the exconjugants separate without undergoing postzygotic nuclear differentiation (stage e). The two exconjugants are heterokaryons: their MICs are both whole-genome homozygotes derived from the single, dip- loidized meiotic product, and their MACs are those of the two parental cells that remain unchanged. The two types of whole-genome homozygotes are obtained in 1:1 ratio (only the white, mutated is shown). If two exconjugants from the same round 1 pair are crossed to one another (round II; stage f), whole-genome homozygous homo- karyon progeny are obtained (stage g). (B) UPC. In this method, a cell heterozygous for a mutation is crossed to a star cell (stages a and b are identical to stages a and b in genomic exclusion), but gamete pronucleus transfer is blocked by a suitably timed hyperosmotic shock (between stages c and d). As a consequence, the two gamete pro- nuclei fuse to one another, generating a diploid, whole- genome homozygous zygote nucleus (stage d). Unlike round I of the genomic exclusion cross, postzygotic nu- clear differentiation and old MAC destruction occur nor- mally. The star mate dies for lack of a MIC and MAC (stage e). The resulting cell line has MICs and MACs that are whole-genome homozygous for the haploid genome of the surviving meiotic product at stage b (homokaryons). UPC is the method of choice to isolate mutants homozy- gous for recessive mutations after mutagenesis.

Birkeland et al. 2010; Cuperus et al. 2010; Galati et al. 2014; linked to the mutant phenotype in Tetrahymena or other or- Marker et al. 2014; A. Turkewitz and J. Gaertig, personal com- ganisms? Candidates are also tested by cloning and Sanger munication). To map by sequencing, the mutant is backcrossed sequencing. The final candidate is confirmed by expression to the wild-type parent and a panel of homozygous F2 meiotic of a wild-type copy of the candidate gene to rescue the mu- segregants is obtained by UPC. Alternatively, F1 progeny of tant phenotype. different mating types can be crossed to one another and F2’s displaying the mutant phenotype can be selected (Galati et al. Auxiliary tools: mapping mutations to chromosome segments: 2014; A. Turkewitz, personal communication). A pool of whole- Prior to the advent of NGS, the physical location of mutations cell DNA extracted from mutant F2’sandwhole-cellDNA within the genome was determined using deletion mapping. extracted from the wild-type parental line are sequenced. Se- This remains a useful complementary approach to narrow quencing reads are aligned to the MIC and MAC reference candidate mutations from mapping-by-sequencing experi- genomes, which are fully assembled and coaligned and variants ments. In deletion mapping, the physical location of mutations identified. Candidate mutations are those that are found in within the genome is determined by crossing a homozygous 100% of both MAC and MIC reads. mutant to a nullisomic cell line (Figure 6) (Gutiérrez and Orias The candidate mutations are prioritized, using one or more 1992). Nullisomic cell lines are heterokaryons with a func- functional criteria: Is the mutation within an annotated gene tional MAC, but lack both MIC copies of a chromosome or a sequence? Does the candidate gene have a function previously chromosome arm. Crossing a mutant to a series of nullisomic

Genetics Toolbox Review 659 lines that collectively span the entire germline genome allows protein encoded by the cDNA. The plasmids can then be re- one to rapidly identify the chromosome location of the muta- covered and sequenced to identify the genes that encode tion. Upon mating of a homozygous mutant with a nullisomic proteins that localize to regions of interest. Variations on this strain, the fertilization nuclei will be monosomic, i.e.,theywill experiment include using different promoters that have dif- have only one copy of the chromosome (arm) missing in the ferent cell cycle and developmental timing of expression to nullisomic. If the mutation is recessive and maps to that chro- analyze protein localization at specific stages of the cell cycle mosome (arm), then there will be no wild-type allele to rescue or during conjugation (Yao et al. 2007). the mutant phenotype and the progeny will display the mutant Tetrahymena cDNA libraries also have been utilized to pro- phenotype; otherwise the progeny phenotype will be wild duce “antisense ribosomes” (Sweeney et al. 1996). Antisense type. Nullisomic mapping eliminates the vast majority of false ribosomes are ribosomes that display the reverse comple- candidates in NGS experiments, since only variants in the MAC ment of a portion of a messenger RNA (mRNA) of interest contigs that map to the MIC chromosome arm determined to (or, in this case, an entire cDNA library) inserted at a harmless harbor the mutation need to be considered. Segmental dele- position in the large rRNA subunit. The target mRNA is trans- tion homozygotes allow even finer mapping by the same lationally repressed. This is an effective method to knock method. Dominant mutations can also be deletion mapped down a library of genes, which can be followed by phenotypic in a similar manner. However, because all progeny will display screening and identification of the genes by sequencing the the dominant mutant phenotype, cells must be assorted for recovered antisense ribosomal DNA (rDNA) vectors. This many generations. If the mutation lies within the chromosome method was used to identify genes involved in secretion region that is missing in the nullisomic parent, the location will (Chilcoat et al. 2001). This study highlights the first example be recognized because the progeny will not have a wild-type of a Tetrahymena gene identified and cloned solely based on allele and the wild-type phenotype will never be recovered its mutant phenotype. after assortment. Reverse genetics: analysis of identified genes If still necessary, a candidate mutation’s location can be mapped at higher resolution using classical genetic meiotic Once identified, gene functions can be further illuminated recombination linkage analysis, regardless of whether the using a suite of modern molecular tools in Tetrahymena. Here, mutation is dominant or recessive. DNA polymorphisms we describe a variety of methods for gene perturbation and called randomly amplified polymorphic DNAs (RAPDs) are study. commonly used for this analysis. Differences in RAPD loca- tions within inbred strains B and C3 have been mapped in the Gene perturbation: heritable transgenics: DNA can be sta- Tetrahymena MIC genome. Additionally, nearly every MAC bly targeted and integrated in the somatic or germline ge- chromosome has a mapped DNA polymorphism (Lynch nomes of Tetrahymena using homologous recombination. et al. 1995; Brickner et al. 1996; Wickert and Orias 2000). The DNA cassettes described below enable gene knockouts, After crossing the mutant (usually generated in inbred strain knockins, and mutations. B) with a wild-type cell (usually strain C3) the DNA of every Transformation of DNA into Tetrahymena is performed

F2 homozygous segregant (mutant and wild type) is tested by using microinjection, electroporation, or biolistics. Transfor- PCR amplification to identify whether the RAPDs are from mations require little DNA (microgram quantities) and posi- strain C3 or B. For efficiency, this testing is limited to the MAC tive transformants are acquired in ,1 week. Transformation chromosomes that were derived from the MIC chromosome of DNA into the MAC or the MIC depends on the method arm or smaller segment previously determined to carry the used; in microinjections, needles can specifically deliver mutation. The proximity of RAPD loci to the mutation of in- DNA to the MAC, whereas electroporation and biolistics rely terest is determined by linkage analysis, where the recombi- on the precise timing of the transformation during the cell nation frequency is proportional to the distance between the cycle or during conjugation. Recovery of MIC transformants two loci, allowing identification of a MAC chromosome seg- requires mating, while MAC transformants can be selected ment containing the mutation of interest. directly. Transformations using biolistics are simple and effi- cient, making it the method of choice. This is especially im- Molecular approaches to forward genetics: As an alterna- portant for low-efficiency transformations that target MIC tive to random mutagenesis in forward genetics experiments, DNA integration (Cassidy-Hanley et al. 1997; Hai and complementary DNA (cDNA) library screens have also led to Gorovsky 1997). the unbiased discovery of proteins involved in a variety of DNA is typically introduced into Tetrahymena in one of processes in Tetrahymena and other organisms. For example, two forms: as rDNA vectors that are autonomously replicat- GFP-fusion libraries have been constructed where cDNAs are ing, extrachromosomal MAC DNA, or by integration into and fused in frame with the coding sequence of GFP (Sawin and precise replacement of a targeted chromosomal segment by Nurse 1996; Fujii et al. 1999; Rolls et al. 1999; Ding et al. homologous recombination. The rDNA vectors are high copy 2000; Misawa et al. 2000; Escobar et al. 2003; Yao et al. vectors (9,000 copies) containing a drug resistance marker 2007). Upon transformation of these constructs, the localiza- and are used for exogenous protein overexpression. Gene tion of GFP within the cell identifies the localization of the perturbations, however, are typically introduced into the

660 M. D. Ruehle, E. Orias, and C. G. Pearson Tetrahymena genome by homologous recombination. In this complementary to both the IES sequences and to the target process, linear DNA sequences containing a drug resistance sequence they flank. These small RNAs can then simulta- marker flanked by homologous regions to a gene of interest neously target most of the copies of the endogenous MAC are used to specifically target and replace that gene (Yao and locus of the gene of interest and induce its elimination during Yao 1991; Hai et al. 2000). These knockout constructs confer MAC development. Two independent targets can be included resistance to cycloheximide, paromomycin, or blasticidine. in the same CoDel vector. This promising method allows for Upon transformation into Tetrahymena cells, the knockout fast, targeted gene disruption in Tetrahymena. construct replaces the endogenous gene by exact homolo- gous recombination (Yao and Yao 1991; Kahn et al. 1993). Gene perturbation: RNA-based knockdowns: Several RNA- For somatic MAC transformations, the DNA cassette in- based perturbation strategies are accessible in Tetrahymena. serts into only one or a few copies of the targeted locus while Antisense ribosomes were first reported as an effective tool the remaining copies possess a wild-type gene (Hai et al. for reducing the expression of genes (Sweeney et al. 1996), as 2000). Once transformed, phenotypic assortment is used to described above for cDNA libraries. Rather than using a li- generate MACs pure for somatic MAC knockouts or knock- brary of antisense cDNAs, an antisense DNA fragment to the downs containing a drug resistance gene (Figure 5B). Step- 59 untranslated region of a specific mRNA is inserted into the wise increases in drug concentration are used to select for rRNA gene. The antisense RNA is displayed on every ribo- cells that stochastically increase the gene copies containing some and inhibits translation of the gene. the knockout cassette during division while killing those that As discussed previously, Tetrahymena RNA interference do not. For nonessential genes, homozygous MAC genomic (RNAi) pathways control programmed DNA elimination from knockouts are obtained. Essential genes can also be selected the MAC genome during MAC development (Chalker et al. through phenotypic assortment and driven toward knock- 2013). Likewise, related RNAi machinery can be hijacked down, although incompletely because the loss of all copies throughout the cell cycle and development to induce gene of the wild-type allele is lethal. This produces a knockdown silencing. RNA hairpin constructs can be expressed by strong genotype and is a major advantage of Tetrahymena since this transcriptional promoters (Howard-Till and Yao 2006; Beales is not easily achieved in other organisms. When desired, ho- et al. 2007; Awan et al. 2009; Howard-Till et al. 2011, 2013). mozygous KOs for essential genes are generated by MIC The RNA hairpins then become processed into 23- and 24- transformation, as described below. nucleotide sequences to induce gene silencing by mRNA Germline MIC mutations are produced by transforming degradation. mating cells during meiosis, when recombination is highest in the MIC genome (Cassidy-Hanley et al. 1997). This intro- Exogenous gene expression: The introduction of DNA for duces at least one copy of the drug resistance gene into the controlled protein expression in Tetrahymena cells is utilized diploid MIC genome, producing a heterozygote that ex- to rescue mutant phenotypes, express mutant forms of a gene presses it in the newly developed MAC, and can be selected or exogenously tagged protein, and large-scale protein pro- for during subsequent vegetative growth. Confirmed MIC duction. Such DNA cassettes are introduced as either rDNA integrants are made homozygous heterokaryons using round integrating constructs for high copy number insertion and I of genomic exclusion (Figure 7A). The knockout is thus protein expression, or they can be integrated at specific ge- propagated in the MIC without a phenotype. To reveal the nomic loci by homologous recombination. These systems al- phenotype, two such homozygous heterokaryons of different low for fine-tuned exogenous expression of proteins, but as a mating types are mated to produce progeny that are homo- general consideration, all exogenous sequences including zygous MAC genomic knockouts (round II of genomic exclu- tags and fusions may need to be codon optimized for sion). An advantage to this strategy is that lethal mutations Tetrahymena. can be propagated in phenotypically wild-type cells and then Exogenous expression of Tetrahymena genes is controlled knockout conditions can be induced by mating in synchro- by either constitutive or inducible promoters. In Tetrahymena, nous and large populations of cells for phenotypic analyses. the major constitutive promoter utilized is the histone H4 pro- Recently,a rapid method for MAC gene knockouts has been moter (Kahn et al. 1993). Different promoters can be used for developed that does not depend on homologous recombina- selective cell cycle- and developmental-specificgeneexpres- tion (Hayashi and Mochizuki 2015; Noto et al. 2015). Codeletion sion. Additionally, strongly inducible promoters provide direct (CoDel) harnesses the endogenous mechanism for IES DNA control over the temporal expression and the relative protein lev- elimination during MAC development to remove 1kbofa els. This is typically controlled by using various metallothionein gene of interest. While this method leads to only a knock- (MTT) gene promoters, which can be induced by the addition down of essential genes, it is a highly effective knockout of cadmium, mercury, copper, or zinc. The MTT1 promoter, strategy for nonessential genes. The target sequence for induced by cadmium chloride (CdCl2) (Shang et al. 2002), is knockout is flanked by known IES sequences in a high copy the most commonly used; however, copper inducible versions number rDNA extrachromosomal vector that is introduced also exist (Boldrin et al. 2008). In addition, the Tetrahymena into conjugating cells undergoing MAC differentiation. The Hsp70-2 promoter is highly inducible by a short heat shock IES sequences trigger the production of small RNAs that are (Yu et al. 2012). As the repertoire of inducible promoters

Genetics Toolbox Review 661 expands (including tetracycline inducible systems), their util- processes and will build systems-level genetic networks. Ad- ity in exogenous gene expression will become more accessible. ditionally, Tetrahymena researchers will benefit from im- proved next generation sequencing methods and their Protein tagging: A diverse collection of reagents for protein accompanying computational tools. Because many of these tagging are available and freely shared by the Tetrahymena technologies have not necessarily been developed for Tetra- research community. These reagents facilitate the generation hymena, adapting them to the organism will be a major focus, of constructs for protein localization using light and electron and one that is quite promising and reachable. microscopy, biochemical purifications, and protein dynamics Although Tetrahymena was established as an advanta- studies. Many of these reagents and Tetrahymena strains are geous model organism in the early 1920s (Collins and available through the National Tetrahymena Stock Center. Gorovsky 2005), its utility to researchers has only increased Codon optimized EGFP and mCherry constructs are available over the last century. Tetrahymena continues to be a vehicle for live and fixed cell imaging of protein localization in for groundbreaking discoveries in structural, molecular, and Tetrahymena cells using both fluorescence localization and cellular biology because of its ability to be genetically manip- immuno-EM (IEM) to visualize the localization of proteins ulated, biochemically deconstructed, and visually inspected. (Kataoka et al. 2010; Stemm-Wolf et al. 2013). Several of Together, these unique advantages give us an unprecedented the localization tags are also amenable to biochemical puri- view into the inner workings of the ciliate’s complex life, and fications with localization and purification (LAP) tags that it, in turn, has taught us about the mechanisms that govern have been adapted from mammalian LAP tags and then our own. And, the end is not in sight; as more genetics and codon optimized for Tetrahymena (Rigaut et al. 1999; genomics tools emerge, we see Tetrahymena continuing to Cheeseman and Desai 2005; Pearson et al. 2009; Winey keep its place as one of the fastest, most accessible, and rel- et al. 2012). Additional purification strategies have been de- evant model organisms to address crucial biological ques- veloped apart from the LAP tag (Couvillion and Collins tions that must be answered at the lightning speed of 2012). These tags provide the ability to couple powerful science today. in vivo cell biology experiments with the detail gleaned from reductionist biochemical approaches. Acknowledgments The authors thank Domenico “Nick” Galati, Brian Bayless, Conclusions and Perspectives Aaron Turkewitz, and Alex Stemm-Wolf for critical reading Tetrahymena’s evolutionary divergence from the more com- and comments, and A. Turkewitz and J. Gaertig for commu- monly studied model organisms, while retaining most of the nicating unpublished results. C.G.P. is funded by the Na- cell biology inherited from the last eukaryotic common ances- tional Institutes of Health–National Institute of General tor, is a major advantage as a model system. Within the context Medical Sciences GM099820, Pew Biomedical Scholars Pro- of its unusual life cycle, genetic and molecular manipulations gram, and the Boettcher Foundation. E.O. is funded by a that are impossible in other organisms enable Tetrahymena to University of California Santa Barbara Academic Senate Re- reveal universally conserved cellular processes. Harnessing search Award. this unique power continues to underlie Tetrahymena’s utility to researchers working on diverse scientificproblems. What does the future hold for Tetrahymena research? Cer- Literature Cited tainly, advances will be built upon existing methods, and the Ahmed, S., H. Sheng, L. Niu, and E. Henderson, 1998 Tetrahymena quest for more information, higher resolution, and less am- mutants with short telomeres. Genetics 150: 643–650. biguous results is never ending. The Tetrahymena research Allen, S. L., 1967 Genomic exclusion: a rapid means for inducing community strives to improve molecular tools such as RNAi homozygous diploid lines in Tetrahymena pyriformis, syngen 1. knockdown strategies, to develop clustered regularly inter- Science 155: 575–577. spaced short palindromic repeat (CRISPR) technology ame- Allen, S. L., and D. L. Nanney, 1958 An analysis of nuclear differ- – nable to MIC genome editing, and to expand the repertoire of entiation in the selfers of Tetrahymena. Am. Nat. 92: 139 160. Andersen, H. A., and E. Zeuthen, 1971 DNA replication sequence inducible promoter systems. Additionally, optogenetic and in Tetrahymena is not repeated from generation to generation. dimerization targeting tools will be used to induce protein Exp. Cell Res. 68: 309–314. localization or gene expression upon light and chemical ex- Aury, J.-M., O. Jaillon, L. Duret, B. Noel, C. Jubin et al., posure. Along with the development of quantitative fixed and 2006 Global trends of whole-genome duplications revealed live cell imaging methods (a significant hurdle for a motile by the ciliate Paramecium tetraurelia. Nature 444: 171–178. organism), these efforts will expand the reverse genetics Awan, A., A. J. Bell, and P. Satir, 2009 Kin5 knockdown in Tetra- hymena thermophila using RNAi blocks cargo transport of Gef1. tools in the Tetrahymena toolbox. In forward genetics ap- PLoS ONE 4: e4873. proaches, initiating synthetic genetic array (SGA) analyses Beales, P. L., E. Bland, J. L. Tobin, C. Bacchelli, B. Tuysuz et al., similar to those that were successfully performed in other 2007 IFT80, which encodes a conserved intraflagellar trans- model organisms (for review see Dixon et al. 2009) will be port protein, is mutated in Jeune asphyxiating thoracic dystro- invaluable to identify genes that contribute to similar phy. Nat. Genet. 39: 727–729.

662 M. D. Ruehle, E. Orias, and C. G. Pearson Birkeland, S. R., N. Jin, A. C. Ozdemir, R. H. Lyons, L. S. Weisman Couvillion, M. T., and K. Collins, 2012 Biochemical approaches et al., 2010 Discovery of mutations in Saccharomyces cerevisiae including the design and use of strains expressing epitope- by pooled linkage analysis and whole-genome sequencing. Ge- tagged proteins. Methods Cell Biol. 109: 347–355. netics 186: 1127–1137. Coyne, R. S., and M. C. Yao, 1996 Evolutionary conservation of Blackburn, E. H., 2010 Telomeres and telomerase: the means to the sequences directing chromosome breakage and rDNA palindrome end (Nobel lecture). Angew Chem. Int. Ed. Engl. 49: 7405–7421. formation in tetrahymenine ciliates. Genetics 144: 1479–1487. Blumenstiel, J. P., A. C. Noll, J. A. Griffiths, A. G. Perera, K. N. Cuperus,J.T.,T.A.Montgomery,N.Fahlgren,R.T.Burke,T.Townsend Walton et al., 2009 Identification of EMS-induced mutations et al., 2010 Identification of MIR390a precursor processing- in Drosophila melanogaster by whole-genome sequencing. Ge- defective mutants in Arabidopsis by direct genome sequencing. netics 182: 25–32. Proc. Natl. Acad. Sci. USA 107: 466–471. Boldrin, F., G. Santovito, A. Formigari, Y. Bisharyan, D. Cassidy- Ding, D. Q., Y. Tomita, A. Yamamoto, Y. Chikashige, T. Haraguchi Hanley et al., 2008 MTT2, a copper-inducible metallothionein et al., 2000 Large-scale screening of intracellular protein local- gene from Tetrahymena thermophila. Comp. Biochem. Physiol. ization in living fission yeast cells by the use of a GFP-fusion C Toxicol. Pharmacol. 147: 232–240. genomic DNA library. Genes Cells 5: 169–190. Bowman,G.R.,N.C.Elde,G.Morgan,M.Winey,andA.P.Turkewitz, Dixon,S.J.,M.Costanzo,A.Baryshnikova, B. Andrews, and 2005 Core formation and the acquisition of fusion competence C. Boone, 2009 Systematic mapping of genetic interaction net- are linked during secretory granule maturation in Tetrahymena. works. Annu. Rev. Genet. 43: 601–625. Traffic 6: 303–323. Doerder, F. P., and L. E. DeBault, 1978 Life cycle variation and Brickner, J. H., T. J. Lynch, D. Zeilinger, and E. Orias, regulation of macronuclear DNA content in Tetrahymena ther- 1996 Identification, mapping and linkage analysis of randomly mophila. Chromosoma 69: 1–19. amplified DNA polymorphisms in Tetrahymena thermophila. Ge- Doerder, F. P., J. H. Lief, and L. E. Doerder, 1975 A corrected table netics 143: 811–821. for macronuclear assortment in Tetrahymena pyriformis, syngen Briguglio, J. S., and A. P. Turkewitz, 2014 Tetrahymena thermo- 1. Genetics 80: 263–265. phila: a divergent perspective on membrane traffic. J. Exp. Zool. Duharcourt, S., G. Lepère, and E. Meyer, 2009 Developmental ge- B Mol. Dev. Evol. 322: 500–516. nome rearrangements in ciliates: a natural genomic subtraction Brownell, J. E., J. Zhou, T. Ranalli, R. Kobayashi, D. G. Edmondson mediated by non-coding transcripts. Trends Genet. 25: 344–350. et al., 1996 Tetrahymena histone acetyltransferase A: a homo- Eisen, J. A., R. S. Coyne, M. Wu, D. Wu, M. Thiagarajan et al., log to yeast Gcn5p linking histone acetylation to gene activa- 2006 Macronuclear genome sequence of the ciliate Tetrahy- tion. Cell 84: 843–851. mena thermophila, a model eukaryote. PLoS Biol. 4: e286. Bruns, P. J., and Y. M. Sanford, 1978 Mass isolation and fertility Elde, N. C., G. Morgan, M. Winey, L. Sperling, and A. P. Turkewitz, testing of temperature-sensitive mutants in Tetrahymena. Proc. 2005 Elucidation of clathrin-mediated endocytosis in Tetrahy- Natl. Acad. Sci. USA 75: 3355–3358. mena reveals an evolutionarily convergent recruitment of Candela, H., and S. Hake, 2008 The art and design of genetic dynamin. PLoS Genet. 1: e52. screens: maize. Nat. Rev. Genet. 9: 192–203. Elliott, A. M., and D. F. Gruchy, 1952 The occurrance of mating Casselton, L., and M. Zolan, 2002 The art and design of genetic types in Tetrahymena. Biol. Bull. 103: 301. screens: filamentous fungi. Nat. Rev. Genet. 3: 683–697. Escobar, N. M., S. Haupt, G. Thow, P. Boevink, S. Chapman et al., Cassidy-Hanley, D., J. Bowen, J. H. Lee, E. Cole, L. A. VerPlank 2003 High-throughput viral expression of cDNA-green fluores- et al., 1997 Germline and somatic transformation of mating cent protein fusions reveals novel subcellular addresses and Tetrahymena thermophila by particle bombardment. Genetics identifies unique proteins that interact with plasmodesmata. 146: 135–147. Plant Cell 15: 1507–1523. Cassidy-Hanley, D. M., 2012 Tetrahymena in the laboratory: Fass, J. N., N. A. Joshi, M. T. Couvillion, J. Bowen, M. A. Gorovsky strain resources, methods for culture, maintenance, and storage. et al., 2011 Genome-scale analysis of programmed DNA elim- Methods Cell Biol. 109: 237–276. ination sites in Tetrahymena thermophila. G3 (Bethesda) 1: Cavalier-Smith, T., 1993 Kingdom protozoa and its 18 phyla. Mi- 515–522. crobiol. Rev. 57: 953–994. Flickinger, C. J., 1965 The fine structure of the nuclei of Tetrahy- Cervantes, M. D., R. S. Coyne, X. Xi, and M.-C. Yao, 2006 The mena pyriformis throughout the cell cycle. J. Cell Biol. 27: 519– condensin complex is essential for amitotic segregation of bulk 529. chromosomes, but not nucleoli, in the ciliate Tetrahymena ther- Forsburg, S. L., 2001 The art and design of genetic screens: yeast. mophila. Mol. Cell. Biol. 26: 4690–4700. Nat. Rev. Genet. 2: 659–668. Cervantes, M. D., E. P. Hamilton, J. Xiong, M. J. Lawson, D. Yuan Frankel, J., 2000 Cell biology of Tetrahymena thermophila. Meth- et al., 2013 Selecting one of several mating types through gene ods Cell Biol. 62: 27–125. segment joining and deletion in Tetrahymena thermophila. PLoS Frankel, J., L. M. Jenkins, F. P. Doerder, and E. M. Nelsen, Biol. 11: e1001518. 1976 Mutations affecting cell division in Tetrahymena pyriformis. Chalker, D. L., E. Meyer, and K. Mochizuki, 2013 Epigenetics of I. Selection and genetic analysis. Genetics 83: 489–506. ciliates. Cold Spring Harb. Perspect. Biol. 5: a017764. Fujii, G., R. Tsuchiya, E. Ezoe, and S. Hirohashi, 1999 Analysis of Cheeseman, I. M., and A. Desai, 2005 A combined approach for nuclear localization signals using a green fluorescent protein- the localization and tandem affinity purification of protein com- fusion protein library. Exp. Cell Res. 251: 299–306. plexes from metazoans. Sci. STKE 2005: pl1. Galati, D. F., S. Bonney, Z. Kronenberg, C. Clarissa, M. Yandell Chilcoat, N. D., N. C. Elde, and A. P. Turkewitz, 2001 An anti- et al., 2014 DisAp-dependent striated fiber elongation is re- sense approach to phenotype-based gene cloning in Tetrahy- quired to organize ciliary arrays. J. Cell Biol. 207: 705–715. mena. Proc. Natl. Acad. Sci. USA 98: 8709–8713. Gibbons, I. R., and A. J. Rowe, 1965 Dynein: a protein with aden- Cole, E. S., and P. J. Bruns, 1992 Uniparental cytogamy: a novel osine triphosphatase activity from cilia. Science 149: 424–426. method for bringing micronuclear mutations of Tetrahymena Greenwood, S. J., M. L. Sogin, and D. H. Lynn, 1991 Phylogenetic into homozygous macronuclear expression with precocious sex- relationships within the class Oligohymenophorea, phylum ual maturity. Genetics 132: 1017–1031. Ciliophora, inferred from the complete small subunit rRNA Collins, K., and M. A. Gorovsky, 2005 Tetrahymena thermophila. gene sequences of Colpidium campylum, Glaucoma chattoni, Curr. Biol. 15: R317–R318. and Opisthonecta henneguyi.J.Mol.Evol.33:163–174.

Genetics Toolbox Review 663 Greider, C. W., and E. H. Blackburn, 1985 Identification of a spe- Merriam, E. V., and P. J. Bruns, 1988 Phenotypic assortment in cific telomere terminal transferase activity in tetrahymena ex- Tetrahymena thermophila: assortment kinetics of antibiotic- tracts. Cell 43: 405–413. resistance markers, tsA, death, and the highly amplified rDNA Gutiérrez, J. C., and E. Orias, 1992 Genetic characterization of locus. Genetics 120: 389–395. Tetrahymena thermophila mutants unable to secrete capsules. Misawa, K., T. Nosaka, S. Morita, A. Kaneko, T. Nakahata et al., Dev. Genet. 13: 160–166. 2000 A method to identify cDNAs based on localization of Hai, B., and M. A. Gorovsky, 1997 Germ-line knockout hetero- green fluorescent protein fusion products. Proc. Natl. Acad. karyons of an essential alpha-tubulin gene enable high- Sci. USA 97: 3062–3066. frequency gene replacement and a test of gene transfer from Mochizuki, K., N. A. Fine, T. Fujisawa, and M. A. Gorovsky, somatic to germ-line nuclei in Tetrahymena thermophila. Proc. 2002 Analysis of a piwi-related gene implicates small RNAs Natl. Acad. Sci. USA 94: 1310–1315. in genome rearrangement in Tetrahymena. Cell 110: 689–699. Hai, B., J. Gaertig, and M. A. Gorovsky, 2000 Knockout hetero- Nanney, D. L., and J. W. McCoy, 1976 Characterization of the karyons enable facile mutagenic analysis of essential genes in species of the Tetrahymena pyriformis complex. Trans. Am. Mi- Tetrahymena. Methods Cell Biol. 62: 513–531. crosc. Soc. 95: 664–682. Hayashi, A., and K. Mochizuki, 2015 Targeted gene disruption by Nanney, D. L., and P. A. Caughey, 1953 Mating type determina- ectopic induction of DNA elimination in Tetrahymena. Genetics tion in Tetrahymena pyriformis. Proc. Natl. Acad. Sci. USA 39: 201: 55–64. 1057–1063. Howard-Till, R. A., and M.-C. Yao, 2006 Induction of gene silencing Nanney, D. L., and R. M. Preparata, 1979 Genetic evidence con- by hairpin RNA expression in Tetrahymena thermophila reveals a cerning the structure of the Tetrahymena thermophila macronu- second small RNA pathway. Mol. Cell. Biol. 26: 8731–8742. cleus. J. Protozool. 26: 2–9. Howard-Till, R. A., A. Lukaszewicz, and J. Loidl, 2011 The recom- Nilsson, J. R., and B. van Deurs, 1983 Coated pits with pinocytosis binases Rad51 and Dmc1 play distinct roles in DNA break repair in Tetrahymena. J. Cell Sci. 63: 209–222. and recombination partner choice in the meiosis of Tetrahy- Noto, T., K. Kataoka, J. H. Suhren, A. Hayashi, K. J. Woolcock et al., mena. PLoS Genet. 7: e1001359. 2015 Small-RNA-mediated genome-wide trans-recognition net- Howard-Till, R. A., A. Lukaszewicz, M. Novatchkova, and J. Loidl, work in Tetrahymena DNA elimination. Mol. Cell 59: 229–242. 2013 A single cohesin complex performs mitotic and meiotic Orias, E., and M. Flacks, 1975 Macronuclear genetics of Tetrahy- functions in the protist Tetrahymena. PLoS Genet. 9: e1003418. mena. I. Random distribution of macronuclear genecopies in Hünseler, P., G. Scheidgen-Kleyboldt, and A. Tiedtke, 1987 Isolation T. pyriformis, syngen 1. Genetics 79: 187–206. and characterization of a mutant of Tetrahymena thermophila Page, D. R., and U. Grossniklaus, 2002 The art and design of blocked in secretion of lysosomal enzymes. J. Cell Sci. 88(Pt 1): genetic screens: Arabidopsis thaliana. Nat. Rev. Genet. 3: 124– 47–55. 136. Jerka-Dziadosz, M., L. M. Jenkins, E. M. Nelsen, N. E. Williams, R. Patton, E. E., and L. I. Zon, 2001 The art and design of genetic Jaeckel-Williams et al., 1995 Cellular polarity in ciliates: per- screens: zebrafish. Nat. Rev. Genet. 2: 956–966. sistence of global polarity in a disorganized mutant of Tetrahy- Pearson, C. G., D. P. S. Osborn, T. H. Giddings, P. L. Beales, and M. mena thermophila that disrupts cytoskeletal organization. Dev. Winey, 2009 Basal body stability and ciliogenesis requires the Biol. 169: 644–661. conserved component Poc1. J. Cell Biol. 187: 905–920. Jorgensen, E. M., and S. E. Mango, 2002 The art and design of Pennock, D. G., 2000 Selection of motility mutants. Methods Cell genetic screens: Caenorhabditis elegans. Nat. Rev. Genet. 3: 356– Biol. 62: 281–290. 369. Pennock, D. G., T. Thatcher, J. Bowen, P. J. Bruns, and M. A. Kahn, R. W., B. H. Andersen, and C. F. Brunk, 1993 Transformation Gorovsky, 1988 A conditional mutant having paralyzed cilia of Tetrahymena thermophila by microinjection of a foreign gene. and a block in cytokinesis is rescued by cytoplasmic exchange Proc. Natl. Acad. Sci. USA 90: 9295–9299. in Tetrahymena thermophila. Genetics 120: 697–705. Karrer, K. M., 2000 Tetrahymena genetics: two nuclei are better Rigaut, G., A. Shevchenko, B. Rutz, M. Wilm, M. Mann et al., than one. Methods Cell Biol. 62: 127–186. 1999 A generic protein purification method for protein com- Kataoka, K., U. E. Schoeberl, and K. Mochizuki, 2010 Modules for plex characterization and proteome exploration. Nat. Biotech- C-terminal epitope tagging of Tetrahymena genes. J. Microbiol. nol. 17: 1030–1032. Methods 82: 342–346. Rolls, M. M., P. A. Stein, S. S. Taylor, E. Ha, F. McKeon et al., Kile, B. T., and D. J. Hilton, 2005 The art and design of genetic 1999 A visual screen of a GFP-fusion library identifies a new screens: mouse. Nat. Rev. Genet. 6: 557–567. type of nuclear envelope membrane protein. J. Cell Biol. 146: Kruger, K., P. J. Grabowski, A. J. Zaug, J. Sands, D. E. Gottschling 29–44. et al., 1982 Self-splicing RNA: autoexcision and autocycliza- Romero, D. P., and E. H. Blackburn, 1991 A conserved secondary tion of the ribosomal RNA intervening sequence of tetrahymena. structure for telomerase RNA. Cell 67: 343–353. Cell 31: 147–157. Sarin, S., S. Prabhu, M. M. O’Meara, I. Pe’er, and O. Hobert, Kushida, Y., K. Nakano, and O. Numata, 2011 Amitosis requires 2008 Caenorhabditis elegans mutant allele identification by g-tubulin-mediated microtubule assembly in Tetrahymena ther- whole-genome sequencing. Nat. Methods 5: 865–867. mophila. Cytoskeleton (Hoboken) 68: 89–96. Sawin, K. E., and P. Nurse, 1996 Identification of fission yeast Lynch, M., M. C. Field, H. V. Goodson, H. S. Malik, J. B. Pereira-Leal nuclear markers using random polypeptide fusions with green et al., 2014 Evolutionary cell biology: two origins, one objec- fluorescent protein. Proc. Natl. Acad. Sci. USA 93: 15146–15151. tive. Proc. Natl. Acad. Sci. USA 111: 16990–16994. Schensted, I. V., 1958 Appendix model of subnuclear segregation Lynch, T. J., J. Brickner, K. J. Nakano, and E. Orias, 1995 Genetic in the macronucleus of ciliates. Am. Nat. 92: 161–170. map of randomly amplified DNA polymorphisms closely linked Schneeberger, K., 2014 Using next-generation sequencing to iso- to the mating type locus of Tetrahymena thermophila. Genetics late mutant genes from forward genetic screens. Nat. Rev. 141: 1315–1325. Genet. 15: 662–676. Marker, S., Q. Carradec, V. Tanty, O. Arnaiz, and E. Meyer, 2014 A Schoeberl,U.E.,H.M.Kurth,T.Noto,andK.Mochizuki, forward genetic screen reveals essential and non-essential RNAi 2012 Biased transcription and selective degradation of small factors in Paramecium tetraurelia. Nucleic Acids Res. 42: 7268– RNAs shape the pattern of DNA elimination in Tetrahymena. 7280. Genes Dev. 26: 1729–1742.

664 M. D. Ruehle, E. Orias, and C. G. Pearson Schultz, T. W., 1997 Tetratox: Tetrahymena pyriformis popula- Wickert, S., and E. Orias, 2000 Tetrahymena micronuclear ge- tion growth impairment endpoint: a surrogate for fish lethality. nome mapping. a high-resolution meiotic map of chromosome Toxicol. Methods 7: 289–309. 1l. Genetics 154: 1141–1153. Shang, Y., X. Song, J. Bowen, R. Corstanje, Y. Gao et al., Wiley, E. A., and N. A. Stover, 2014 Immediate dissemination of 2002 A robust inducible-repressible promoter greatly fa- student discoveries to a model organism database enhances cilitates gene knockouts, conditional expression, and over- classroom-based research experiences. CBE Life Sci. Educ. 13: expression of homologous and heterologous genes in 131–138. Tetrahymena thermophila.Proc.Natl.Acad.Sci.USA99: Williams, N. E., and R. J. Williams, 1976 Macronuclear division 3734–3739. with and without microtubules in Tetrahymena. J. Cell Sci. 20: Shuman, H. A., and T. J. Silhavy, 2003 The art and design of 61–77. genetic screens: Escherichia coli. Nat. Rev. Genet. 4: 419–431. Winey, M., A. J. Stemm-Wolf, T. H. Giddings, and C. G. Pearson, Smith,J.J.,E.A.Wiley,andD.M.Cassidy-Hanley, 2012 Cytological analysis of Tetrahymena thermophila. Meth- 2012 Tetrahymena in the classroom. Methods Cell Biol. ods Cell Biol. 109: 357–378. 109: 411–430. Xiong, J., Y. Lu, J. Feng, D. Yuan, M. Tian et al., 2013 Tetrahymena Stemm-Wolf,A.J.,G.Morgan,T.H.Giddings,E.A.White,R.Marchione functional genomics database (TetraFGD): an integrated resource et al., 2005 Basal body duplication and maintenance require for Tetrahymena functional genomics. Database (Oxford) 2013: one member of the Tetrahymena thermophila centrin gene fam- bat008. ily. Mol. Biol. Cell 16: 3606–3619. Yao, M. C., and C. H. Yao, 1991 Transformation of Tetrahymena Stemm-Wolf, A. J., J. B. Meehl, and M. Winey, 2013 Sfr13, a to cycloheximide resistance with a ribosomal protein gene member of a large family of asymmetrically localized Sfi1-repeat through sequence replacement. Proc. Natl. Acad. Sci. USA 88: proteins, is important for basal body separation and stability in 9493–9497. Tetrahymena thermophila. J. Cell Sci. 126: 1659–1671. Yao, M. C., and M. A. Gorovsky, 1974 Comparison of the sequences St Johnston, D., 2002 The art and design of genetic screens: Dro- of macro- and micronuclear DNA of Tetrahymena pyriformis. sophila melanogaster. Nat. Rev. Genet. 3: 176–188. Chromosoma 48: 1–18. Stover, N. A., C. J. Krieger, G. Binkley, Q. Dong, D. G. Fisk et al., Yao, M. C., K. Zheng, and C. H. Yao, 1987 A conserved nucleotide 2006 Tetrahymena Genome Database (TGD): a new genomic sequence at the sites of developmentally regulated chromo- resource for Tetrahymena thermophila research. Nucleic Acids somal breakage in tetrahymena. Cell 48: 779–788. Res. 34: D500–D503. Yao, M.-C., C.-H. Yao, L. M. Halasz, P. Fuller, C. H. Rexer et al., Sweeney, R., Q. Fan, and M. C. Yao, 1996 Antisense ribosomes: 2007 Identification of novel chromatin-associated proteins in- rRNA as a vehicle for antisense RNAs. Proc. Natl. Acad. Sci. USA volved in programmed genome rearrangements in Tetrahymena. 93: 8518–8523. J. Cell Sci. 120: 1978–1989. Taverna, S. D., R. S. Coyne, and C. D. Allis, 2002 Methylation of Yu, T., S. Barchetta, S. Pucciarelli, A. La Terza, and C. Miceli, histone h3 at lysine 9 targets programmed DNA elimination in 2012 A novel robust heat-inducible promoter for heterologous Tetrahymena. Cell 110: 701–711. gene expression in Tetrahymena thermophila. Protist 163: 284– Tiedtke, A., and L. Rasmussen, 1988 Lysosomal enzymes in ex- 295. tracellular digestion in the unicellular eukaryote Tetrahymena. J. Cell. Physiol. 136: 554–556. Communicating editor: O. Hobert

Genetics Toolbox Review 665