Ciliate pheromones: primordial self/non-self recognition signals
Running title: Ciliate pheromones
 
Adriana Vallesi, Claudio Alimenti, Pierangelo Luporini
Laboratory of Eukaryotic Microbiology and Animal Biology, School of Biosciences and Veterinary Medicine, University of Camerino, 62032 Camerino (MC) Italy
E-mail:
Pierangelo Luporini: piero.luporini@unicam.it
Claudio Alimenti: claudio.alimenti@unicam.it
Adriana Vallesi: adriana.vallesi@unicam.it
 
Summary
As is common among multi-cellular life forms, also single-cell organisms use pheromones to communicate among members of the same species. In protist ciliates, pheromones have been identified in functional association with mating systems. Consistently with this association, they have for long time been regarded only as non-self signals committed to elicit a mating response of cells to which they bind in paracrine-like (heterologous) fashion. However, their spectrum of activity has revealed wider borders. It also includes a self activity which promotes the vegetative growth of the same cells from which pheromones are secreted, and to which they continuously bind in autocrine fashion. This double self and non-self activity is made possible by the pheromone ability to compete with one another in cell binding reactions. In Euplotes species, which synthesize pheromone families under the control of multiple series of alleles at a single locus, this ability is ensured by the relationships of the structural homology that link these molecules into species-specific globular, disulfide-rich protein families having a common three-helix fold.
 
Key Words
Diffusible signaling proteins; globular helical proteins; cell-cell communication; protein-protein interactions; ciliate mating and growth   
 Introduction
Independently of their prokaryotic or eukaryotic nature, cells are evolutionarily driven to socialize to acquire the temporary or stable multi-cellular organization necessary to increase their size and engage in functional specialization and partition of labor (1). This ancient impulse to socialize implicates a very early evolution of the cell’s ability to discriminate between self and non-self, which is a sine qua non condition for pursuing effective cell-cell communication, cooperation and structure-function integration. In its most basic aspects, this ability is primarily directed to the recognition of self signaling molecules, as exemplified by the quorum sensing phenomena in which bacteria perceive their environmental population densities and activities and, consistently, vary their behavior and metabolism (2, 3). It then undergoes vast functional diversification among the unicellular eukaryotes in which it governs impressive phenomena of cell aggregation based on kin discrimination in social amoebas (4, 5). Its maximal complexity is eventually shown by animals with the evolution of innate and adaptive immunity systems that are primarily directed to recognizing and processing non-self molecules for the defense of body integrity (6, 7).
This chapter focuses attention on the biology and structure of diffusible protein signals, nowadays known as pheromones and earlier as mating-type factors/substances or gamones, that control self/non-self recognition in protist ciliates. Together with Dinoflagellates and Apicomplexa, ciliates (Ciliophora) form the group of Alveolata which clusters with those of Stramenopiles (diatoms) and Rhizaria (foraminifers and radiolarians) into the so called SAR supergroup (8).  
 
Pheromone identification
Ciliate pheromones are the chemical markers of genetically distinct vegetative cell classes—only two like sexes in some species, or multiple with indefinite numbers in others (9, 10)—that have been described as mating types because their mixing determines a ciliate-specific mating phenomenon of conjugation. Ciliate conjugation is a sexual phenomenon as it involves a temporary union in mating pairs between cells which mutually exchange gamete-nuclei, undergo fertilization, and develop new micro- and macronuclei from the synkarya. However, it has nothing to do with general phenomena of fertilization between gametes of opposite sexes. One major reason for opposing this parallelism is that many species of ciliates may form mating pairs between genetically different cells (heterotypic pairs), as well as between genetically alike (clonal) cells (homotypic mating pairs). Although destined to generate self-fertilization, these homotypic pairs are fully fertile (the Blepharisma case excepted) just like the heterotypic ones.
It is precisely by virtue of this unique capability that ciliates have to form homotypic (intraclonal, or selfing) mating pairs that ciliate pheromones were identified more than half a century ago (11). By investigating mating interactions in Euplotes patella—the first Euplotes species used to study the Mendelian genetics of ciliate multiple mating systems controlled by series of alleles co-dominantly expressed at a single genetic locus (annotated as mat locus)--Kimball (11) observed that an experimental condition sufficient to induce the formation of homotypic mating pairs was the simple suspension of cell cultures with cell-free filtrates from other cultures of different mating types. From this observation it became evident that in E. patella, and ciliates in general, the mating-type factors (pheromones) can be freely released into the extracellular environment and that the presence of these molecules in solution can be promptly detected by assaying the mating induction activity of cell-free filtrates.
By applying this mating-induction assay to cell-free filtrates, pheromones have been identified in species of Blepharisma, Dileptus, Oxytricha and Tokophrya besides that in other Euplotes species (12, 13, as reviews). However, studies of structure-function characterization have been carried out to varying degrees of complexity only on pheromones isolated from B. japonicum, E. raikovi, E. octocarinatus, E. nobilii and E. crassus. These studies are reviewed in this chapter with particular attention to the cross-reactions that some Euplotes pheromones show with the signaling system components of more modern organisms.
 
Pheromone structures  
The first two pheromones to be purified and structurally characterized belong to the pink-colored freshwater ciliate, B. japonicum. Although they have been reported as being distinctive of two mating types (I and II) defined as “complementary”, of which nothing is known about their genetic determination, their molecular structures are chemically unrelated. The pheromone isolated from the type-I cells (originally designated as Gamone 1, or Blepharmone) is an unstable glycoprotein of 272 amino acids and six N-linked sugars (14). It is particularly rich in polar and aromatic amino acids (Asn 10.3%, Ser 10.3%, Tyr 8.1%, Trp 5.1%), synthesized in the form of a cytoplasmic precursor, and active in inducing mating between type II cells at picogram concentrations. In contrast, the second pheromone distinctive of type-II cells (also designated as Gamone 2, or Blepharismone) is a very stable tryptophan derivative, namely a calcium-3-(2’-formyl-amino-5’-hydroxybenzoil) lactate (15). This has also been obtained by chemical synthesis and is active at nano-molar concentrations in attracting type-I cells besides that for inducing their mate (16).
Quite different is the picture that emerges for the pheromone structures determined in four species of Euplotes, namely E. raikovi, E. nobilii, E. octocarinatus and E. crassus. In all these species, pheromones have been shown to form species-specific families of structurally homologous proteins. Among these pheromone families, a more in-depth structural knowledge has been obtained for E. raikovi and E. nobilii, which secrete comparatively larger amounts of pheromones. Up to 200 mg of pure protein/liter can be recovered from E. raikovi cell-free filtrates (17). This has first greatly facilitated chemical and genetic approaches to the determination of significant numbers of pheromone primary amino acid sequences. It subsequently opened the way to NMR and crystallographic analyses of the pheromone three-dimensional structures based on the use of proteins at natural isotope composition purified from cell-free filtrate preparations.
In E. raikovi, nine distinct amino acid pheromone sequences in their form of cytoplasmic precursors (pre-pro-pheromones) and six three-dimensional structures of the secreted molecules are known (18_24) (Fig. 1). In contrast to the tight structural conservation of the strongly hydrophobic pre segment (or signal peptide) of 19 amino acids and (to a slightly lesser extent) of the pro segment of 16_18 amino acids, the secreted pheromones of 37_40 amino acids (51 only in pheromone Er-23) show minimal sequence conservation. Only the positions of six cysteines are fully conserved and the sequence dissimilarity between any two pheromones may be as high as 74%. In any case, despite this wide variability in amino acid sequence all the pheromones (pheromone Er-23 in part excepted) mimic one another in their three-dimensional conformation. This is based on a fold of three nearly parallel, right-handed a helices (increased to five with the addition of two single-turn helices in Er-23), that remain tightly associated with each other by the close proximity and strategic disposition of their disulfide bonds. As is usually required for water-borne signaling molecules that must maintain long-term activity in the environment, the stability of the globular conformation of E. raikovi pheromones is such that even heating these proteins up to 95 °C does not cause unfolding and disruption of their secondary structures (25). However, in spite of showing closely comparable molecular architectures and equivalent stabilities, each pheromone has its own unmistakable cell-type structural earmarks that are likely deputed to confer specificity to its signaling activity. These earmarks are imposed by the evolution of variations that are most apparent for the shape, extension and spatial arrangement of the carboxy-terminal domain, as well as for the geometry of the inter-helix loops of the molecules.
The overall structural organization of E. nobilii pheromones strictly matches the E. raikovi pheromone organization, consistently with the close phylogenetic kinship that links these two species (26_27). Nevertheless, E. nobilii pheromones, for which seven amino acid sequences of the precursor forms (28) and four NMR structures have been determined (29_31) (Fig. 2), have longer sequences (from 52 to 63 amino acids), include eight instead of six cysteines and, more important, show structural specificities that appear closely correlated with the different ecology that distinguishes E. nobilii from E. raikovi. While E. raikovi thrives in temperate waters, E. nobilii has colonized the Antarctic and Arctic seas (32_33). Hence, E. nobilii pheromones are cold-adapted, psychrophilic proteins. With respect to E. raikovi pheromones, their overall contents of polar and hydrophobic amino acids is markedly different, most likely in functional correlation with their improved interactions with the solvent (34). They are much richer in Thr (11.7% vs 5.7%), Asn (7.7% vs 4.2%), and Ser (8.6% vs 5.7%) residues, and poorer in Leu (1.1% vs 7.3%), Pro (4.7% vs 8.9%) and Ile (1.6% vs 5.7%). Structurally more apparent are however the reduced extensions that E. nobilii pheromones show for the three helices of the molecular architecture with respect to E. raikovi pheromones. These reductions determine a spatial preponderance of regions devoid of a regular secondary organization, implying that the activity of E. nobilii pheromones in the thermodynamically adverse environment of the polar waters is facilitated by an improved flexibility of their molecular backbones. Indeed, E. nobilii pheromones have been measured to be much less thermo-stable than E. rakovi pheromones, unfolding upon heating to only 55_70 °C (25_35).
Much less information is available on the structures of E. octocarinatus and E. crassus pheromones for which only the amino acid sequences are known on the basis of their respective DNA coding sequences. The determination of three-dimensional conformations of these pheromones has been made impractical by the very low concentrations in which they are secreted; 0.5 mg of pure protein/liter of cell-free filtrate are usually obtained in E. ocrocarinatus and 15_20 mg in E. crassus (36_37).
Nine distinct pheromone amino acid sequences have been determined in E. octocarinatus (Fig. 3). They are twice as long as the E. raikovi and E. nobilii pheromone sequences, extending from 85 to 109 amino acids with ten cysteines (eight in the pheromone Phr4) (38_39). The positions of only four of the ten cysteines have counterparts in E. raikovi and E. nobilii pheromones, implying that the E. octocarinatus pheromones have only two disulfide bonds and few structural motifs in common with those of E. raikovi and E. nobilii. They also appear to be unique among Euplotes pheromones in functioning as chemo-attractants between cells of different mating types (40).
In E. crassus, pheromones have been structurally characterized only recently (37, 41). They were a long time thought to be represented by insoluble membrane-bound proteins, and as such difficult to extract and characterize from cell-membrane preparations. Evidence that E. crassus pheromones are, instead, constitutively secreted into the extracellular environment as in other Euplotes species was obtained from inter-specific mating induction assays (37), suggested by the notion that cross-mating reactions are relatively frequent between Euplotes species in correlation with their high-multiple (virtually open) mating systems (9, 42). The genetic control of these E. crassus pheromones appears not to conform with the general Euplotes pattern provided by multiple series of alleles at a single mat locus. Results from molecular approaches have been found to be more consistently explained by assuming that this control involves a phenomenon of mat gene duplication responsible for the production of two distinct E. crassus pheromone sub-families. One includes the cell type-specific pheromones, and the other includes a new pheromone species that appears to be synthesized in common among cells of compatible mating types (37, 41). The functions of this ‘shared’ pheromone, of which only a single 56-amino acid sequence with eight cysteines (designated Ec-a) has been determined, are still obscure. Preliminary evidence suggests that it might behave like an adaptor, or scaffold protein that selectively interacts with the cell type-specific pheromones to mediate and/or reinforce their receptor binding reactions (41). This possibility is supported by an unusually strong propensity of the Ec-a pheromone to oligomerize when it is purified by chromatographic separation, and is reinforced by the comparison of its sequence with the three 45-amino acid sequences with ten cysteines (designated Ec-1, Ec-2 and Ec-3) that have been determined for the sub-family of the cell type-specific pheromones (Fig. 4). This comparison makes it evident that a major distinctive structural trait of the Ec-a pheromone is the evolution of an unusually extended hydrophilic domain lying centrally in the molecule. This domain lacks Cys residues, abounds in Gly residues and likely takes a random-coil conformation which is presumptive for a specific capability of the Ec-a pheromone to interact with the other proteins.
 
Pheromone activity
As outlined above, it is essentially due to the application of mating induction assays that ciliate pheromones have been identified. This application has thus decisively contributed to support the generalization that ciliate pheromones function exclusively as “sexual factors” committed to stimulate cell mating by binding in paracrine (or heterologous) fashion only to cells other than those from which they are synthesized and secreted (12). However, observations on the physiology of pheromone secretion in E. raikovi turned out to counter this view, because cells were shown to secrete their pheromones throughout their life cycle independently of their incompetence to mate during the immaturity stage, to vary their rates of pheromone secretion in relation to the environmental concentrations of their pheromone, as well as to refrain from forming homotypic mating pairs in the presence of environmental excess of their pheromone (43). A more coherent view with these observations thus appeared to be one predicting that ciliate pheromones, in addition to acting as non-self mating signals, also (and most likely, primarily) act as self signals committed to promoting the vegetative proliferation (mitotic growth) of their same source cells to which they bind in autocrine (or autologous) fashion (43, 44).
A decisive support for an autocrine, cell growth-promoting activity of ciliate pheromones was provided by the finding that E. raikovi cells bind their pheromones to membrane-bound isoforms that arise from a splicing mechanism of intron-like sequences of the same genes which encode the pheromones in the transcriptionally active cell macronucleus (45, 46). These isoforms (functionally regarded as pheromone-receptors) incorporate the full sequence of the cytoplasmic pheromone precursor and, using the signal (pre) peptide (normally expected to be proteolytically removed before the pheromone secretion) as trans-membrane anchoring domain, take a spatial orientation typical of membrane proteins of type-II which have the C-terminal end directed to the cell outside and the N-terminal end directed to the cell inside. Given this structural equivalence between pheromones and extracellular binding domains of the twin receptor molecules, the way pheromones interact with their receptors on the cell surface has been thought to be fully mimicked by the protein-protein interactions that determine the molecular packing of the E. raikovi pheromone Er-1 into crystals (22). In the bi-dimensional plane of the crystal, the Er-1 molecules (that may be figured as pyramids with a triangular base) associate with one another by means of all three of their faces (ab and c, delimited by helices 1 and 2, 2 and 3, and 1 and 3, respectively), through the cooperative utilization of initially weak interactions that arise from the formation of two distinct types of dimer. One of these involves face a of each monomer, and is symmetrical with the two monomers related by a twofold axis; the other involves faces b and c, and is asymmetrical with the two monomers related by a twofold screw axis (Fig. 5). This association is such that half of the molecules direct their carboxy-terminus to one side of the plane, and may thus reflect the spatial orientation of the pheromone-receptors on the cell surface. The second half of the molecules, instead, direct their amino-terminus to the opposite side of the plane and may thus be regarded as free pheromone molecules that bind to their receptors. In line with a cooperative model of protein-protein interactions, the homotypic pheromone/pheromone-receptor complexes formed by cells which grow suspended with their own (self) pheromone have been observed to undergo clustering and internalization via endocytotic vesicles, whereas the heterotypic complexes formed by cells that have been suspended with a non-self pheromone for being induced to mate are not internalized and believed to be in some way involved in bridging cells into mating pairs (47).  
 
Pheromone structure/function relationships with other signaling proteins
Evolutionary relationships of proteins are commonly inferred using traditional automatic sequence/structure comparison methods such as PSI-BLAST and Dali. The reliability of these methods is however generally regarded as sound only for globular proteins that are larger than 100 residues, which is not the case of Euplotes pheromones characterized by quite short polypeptide chains. To assess small globular proteins for their evolutionary relationships, an alternative method has been thought to be provided by a comparative analysis of the variety of structural folds that functionally diverse proteins (encompassing growth factors, toxins, enzyme inhibitors and others) have adopted for their small (an average of 57 ± 29 residues) disulfide-rich (an average of 3 ± 1) domains which are either individual or included within larger polypeptide chains, and characterized by global folds that are stabilized primarily by disulfide bonds and, in second place, by secondary structure elements and hydrophobic interactions (48). In this analysis and with the exception of pheromone Er-23 that presents an its unique five-helix globular array, Euplotes pheromones have been classified together with disulfide-rich domains of other four protein families represented by anaphylotoxins (C5a), oncogene-encoded proteins (p8MTCPI), sea anemone toxins K, and cysteine-rich secretory proteins (CRISP) within a fold group identified by a disulfide-bonded three-helix bundle with right-handed connections between the a-helices. While classified within the same fold group, these protein families however show quite disparate functions and mechanisms of action with respect to Euplotes pheromones, and lacking a functional analogy their structural relationships with Euplotes pheromones thus appear to be better interpreted as a result of phenomena of convergent evolution rather than of homology.  
More significant structure/function relationships have been observed between the E. raikovi pheromone family and another family of long-distance water-borne protein pheromones that various species of the common marine mollusk Aplysia freely release from their yolk gland into the environment, and use in phenomena of intra- and inter-specific attraction of sexually mature individuals towards their egg cordons (49). In their active form, these 56_58 amino acid proteins, designated as attractins in relation to their activity, have a compact folded structure (annotated as “a-Hairpin” in the above classification of protein fold groups) that is stabilized by three tightly conserved disulfide bonds and dominated by two anti-parallel helices, the second of which is essentially formed by the heptapeptide Ile-Glu-Glu-Cys-Lys-Thr-Ser (50). This building block is common to all attractins of the different species of Aplysia (which well accounts for attractin inter-specific attractiveness), and finds a close counterpart in the third helix of E. raikovi pheromones in terms of both the conformation of its helical backbone and orientation of its side chains. In addition, it is likely directly involved in receptor-binding interactions, as is shown to be the case for the third helix of E. raikovi pheromones, since the attractin activity is abolished by mutating the charged residues that are exposed on its surface (51).
A last intriguing point on the relationships of Euplotes pheromones with other signaling proteins has been quite fortuitously revealed by experiments in which some prototypic animal growth factors and cytokines were used to competitively inhibit the autocrine cell binding interactions of the E. raikovi pheromone Er-1 (52). Human interleukin-2 (hIL-2) and, to a lesser extent, epidermal growth factor were surprisingly observed to exert strong inhibitory activity. At a 50-fold molar excess, hIL-2 appeared able to inhibit almost completely the pheromone binding to its target cell surface receptors. Considering the disparity in molecular dimensions that separates hIL-2 and Er-1 and precludes reliable comparisons between the two protein families, a possible explanation for this hIL-2 competitive activity has tentatively been identified with some degree of the sequence similarity that hIL-2 and Er-1 bear at the level of two their sequence segments, one extended for five amino acids (Leu56-Gln-Cys-Leu-Glu60 in hIL-2 and Iso8-Gln-Cys-Val-Glu12 in Er-1) and the other for seven (Thr102-Phe-Met-Cys-Glu-Tyr-Ala108 in hIL-2 and Arg25-Thr-Gly-Cys-Tyr-Met-Tyr31 in Er-1). The Cys residue included in each of the two segments is involved in the formation of a disulfide bond known to be essential for the hIL-2 biological activity (53), and the shorter segment represents a substantially conserved motif within both the hIL-2 and Er-1 protein families in addition to being well exposed on the molecular surface.
Cross-reactions between the IL-2 and E. raikovi pheromone systems have supported a possible functional significance of these common sequence motifs. It has initially been shown (54) that Er-1 binds specifically to the IL-2 receptor on the surface of CTLL-2 cells, that are a mouse T lymphocyte line totally dependent on IL-2 for survival and proliferation. The binding involves the a and b subunits of the trimeric IL-2 receptor in particular, and occurs with an appreciable higher affinity to the a subunit that has a leading role in the ligand binding activity of the IL-2 receptor being the first subunit that interacts with extracellular IL-2. More recently it has been observed that the pheromone Er-1 also exerts strong effects on some cellular functions that IL-2 regulates in the human lymphoid T cells of the Jurkat line (55). These cells, suspended in a low-serum containing medium enriched with the addition of Er-1 nano-molar concentrations, were initially reported to significantly increase their proliferation and secretion rates, beside that of IL-2, of a wide spectrum of cytokines including IFN-γ, TNF-α, IL-1β and IL-13 (Fig. 6). They were later observed to undergo a significant decrease in the expression of the a-subunit of the IL-2 receptor due to the reversible and time-dependent endocytosis of this receptor, and activate the transduction pathway that is regulated by the protein kinases Erk1 and Erk2.
 
Conclusions
Significant numbers of molecular structures of ciliate pheromones have been determined from distinct species of Euplotes. This determination has made manifest that these water-borne signaling molecules form species-specific families of small, disulfide-rich globular proteins in full accord with their genetic determination at a single multi-allelic locus. Within each family, the global pheromone structure has been shown to acquire virtually unlimited polymorphisms from variations in shape and geometry that interest individual structural elements. The varying degrees of intra- and inter-specific relationships of structural homology that link Euplotes pheromones to one another closely recall the relationships of homology which exist within and between families of signal proteins of higher life forms. In these families, each component can compete and cross-react with its family members to elicit multiple and context-dependent cell responses. Similarly, Euplotes pheromones have been shown to be fully able to bind cells in mutual competition (56), and to carry out a spectrum of activities that is not exclusively limited to eliciting cell mating as hypothesized for the eccentric case of the structurally unrelated glycoprotein and tryptophan-derivative pheromones of B. japonicum (12). In addition to including a paracrine-like (non-self) activity, this spectrum also includes an autocrine (self) activity which is probably primary and directed to promoting the vegetative proliferation of the same cells from which pheromones are constitutively synthesized and secreted throughout the life cycle (44). The similarities of structure that some Euplotes pheromones have revealed with Aplysia attractins and the mammalian cytokine IL-2 most likely represent fortuitous instances of convergent molecular evolution. Nevertheless, they provide evidence that functionally important sequence motifs may be evolutionarily conserved from ancient to more modern cell signaling proteins, and stimulate investigations on their potential into biotechnological and finalized perspectives.  
 
References
1. Grosberg RK, Strathmann RR. The evolution of multicellularity: a minor major transition? Ann Rev Ecol Evol S 2007; 38:621–54.
2. Waters CM, Bassler BL. Quorum sensing: cell-to-cell communication in bacteria. Ann Rev Cell Dev Biol 2005; 21:319–46.
3. Gibbs KA, Urbanowski ML, Greenberg EP. Genetic determinants of self identity and social recognition in bacteria. Science 2008; 321:256_9.
4. Benabentos R, Hirose S, Sucgang R, Curk T, Katoh M, Ostrowski EA, et al. Polymorphic members of the lag gene family mediate kin discrimination in Dyctiostelium. Curr Biol 2009; 19:567–72.
5. Hirose S, Benabentos R, Ho H-I., Kuspa A, Schaulsky G. Self recognition in social amoebae is mediated by allelic pairs of Tiger genes. Science 2011; 333:467–70.
6. Cooper MD, Alder MN. The evolution of adaptive immune systems. Cell 2006; 124:815–22.
7. Rosengarten RD, Nicotra ML. Model systems of invertebrate allorecognition. Curr Biol 2011; 21:R82–R92.
8. Adl SM, Simpson AGB, Lane CE, Lukes J, Bass D, Bowser SS, et al. The revised classification of eukaryotes. J Eukaryot Microbiol 2012; 59:429–93.
9. Valbonesi A, Ortenzi C, Luporini P. The species problem in a ciliate with a high-multiple mating type system, Euplotes crassus. J Protozool 1992; 39:45–54.
10.  Phadke SS, Zufall RA. Rapid diversification of mating systems in ciliates. Biol J Linnean Soc 2009; 98:187–97.
11. Kimball RF. The nature and inheritance of mating types in Euplotes patella. Genetics 1942; 27:269–85.
12. Miyake A 1996. Fertilization and sexuality in ciliates. In: Hausmann K, Bradbury PC, editors. Ciliates, cells as organisms. Stuttgart: Gustav Fisher; 1996, p. 243–90.
13.  Luporini P, Alimenti C, Vallesi A. Ciliate pheromone structures and activity: a review. It J Zool 2015; 82:3–15.
14. Sugiura M, Harumoto T. Identification, characterization, and complete amino acid sequence of the conjugation-inducing glycoprotein (blepharmone) in the ciliate Blepharisma japonicum. Proc Natl Acad Sci USA 2001; 98:14446–51.
15. Kubota T, Tokoroyama T, Tsukuda J, Koyama H, Miyake A. Isolation and structure determination of blepharismin, a conjugation initiating gamone in the ciliate Blepharisma. Science 1973; 179:400–2.
16. Sugiura M, Shiotani H, Suzaki T, Harumoto T. Behavioural changes induced by the conjugation-inducing pheromones, gamones 1 and 2, in the ciliate Blepharisma japonicum. Eur J Protistol 2010; 46:143–9.
17. Raffioni S, Miceli C, Vallesi A, Chowdhury SK, Chait BT, Luporini P, Bradshaw RA. Primary structures of Euplotes raikovi pheromones from cells with variable mating interactions. Proc Natl Acad Sci USA 1992; 89:2071–5.
18. Brown LR, Mronga S, Bradshaw RA, Ortenzi C, Luporini P, Wüthrich K. Nuclear magnetic resonance solution structure of the pheromone Er-10 from the ciliated protozoan Euplotes raikovi. J Mol Biol 1993; 231:800–16.
19. Mronga S, Luginbühl P, Brown LR, Ortenzi C, Luporini P, Bradshaw RA, Wüthrich K. The NMR solution structure of the pheromone Er-1 from the ciliated protozoan Euplotes raikovi. Prot Sci 1994; 3:1527–36.
20. Ottiger M, Szyperski T, Luginbühl P, Ortenzi C, Luporini P, Bradshaw RA, Wüthrich K. The NMR solution structure of the pheromone Er-2 from the ciliated protozoan Euplotes raikovi. Prot Sci 1994; 3:1515–26.
21. Luginbühl P, Ottiger M, Mronga S, Wüthrich K. Structure comparison of the NMR structures of the pheromones Er-1, Er-10, and Er-2 from Euplotes raikovi. Prot Sci 1994; 3:1537–46.
22. Weiss MS, Anderson DH, Raffioni S, Bradshaw RA, Ortenzi C, Luporini P, Eisenberg D. A cooperative model for ligand recognition and cell adhesion: evidence from the molecular packing in the 1.6 Å crystal structure of the pheromone Er-1 from the ciliate protozoan Euplotes raikovi. Proc Natl Acad Sci USA 1995; 92:10172–6.
23. Liu A, Luginbühl P, Zerbe O, Ortenzi C, Luporini P, Wüthrich K. NMR structure of the pheromone Er-22 from Euplotes raikovi. J Biomol NMR 2001; 19:75–8.
24. Zahn R, Damberger F, Ortenzi C, Luporini P, Wüthrich K. NMR structure of the Euplotes raikovi pheromone Er-23 and identification of its five disulfide bonds. J Mol Biol 2001; 313:923–31.
25. Geralt M, Alimenti C, Vallesi A, Luporini P, Wüthrich K. Thermodynamic stability of psychrophilic and mesophilic pheromones of the protozoan ciliate Euplotes. Biology 2013; 2:142–50.
26. Jiang J, Zhang Q, Warren A, Al-Rasheid KAS, Song S. Morphology and SSU rRNA gene based phylogeny of two marine Euplotes species, E. orientalis spec. nov. and E. raikovi Agamaliev, 1966 (Ciliophora, Euplotida). Eur J Protistol 2010; 46:121–32.
27. Di Giuseppe G, Erra F, Frontini FP, Dini F, Vallesi A, Luporini P. Improved description of the bipolar ciliate, Euplotes petzi, and definition of its basal position in the Euplotes phylogenetic tree. Eur J Protistol 2014; 50:402–11
28. Vallesi A, Alimenti C, Pedrini B, Di Giuseppe G, Dini F, Wüthrich K, Luporini P. Coding genes and molecular structures of the diffusible signalling proteins (pheromones) of the polar ciliate, Euplotes nobilii. Mar Genomics 2012; 8:9–13.
29.  Pedrini B, Placzek WJ, Koculi E, Alimenti C, La Terza A, Luporini P, Wuthrich K. Cold-adaptation in sea-water-borne signal proteins: sequence and NMR structure of the pheromone En-6 from the Antarctic ciliate Euplotes nobilii. J Mol Biol 2007; 372:277–86.
30.  Placzek WJ, Etezady-Esfarjani T, Herrmann T, Pedrini B, Peti W, Alimenti C, Luporini P, Wuthrich K. Cold-adapted signal proteins: NMR structures from the Antarctic ciliate Euplotes nobilii. IUBMB Life 2007; 59:578–85.
31. Di Giuseppe G, Erra F, Dini F, Alimenti C, Vallesi A, Pedrini B, Wüthrich K, Luporini P. Antarctic and Arctic populations of the ciliate Euplotes nobilii show common pheromone-mediated cell-cell signaling and cross-mating. Proc Natl Acad Sci USA 2011; 108:3181–6.
32.  Valbonesi A, Luporini P. Description of two new species of Euplotes and Euplotes rariseta from Antarctica. Polar Biol 1990; 11:47–53.
33. Di Giuseppe G, Barbieri M, Vallesi A, Luporini P, Dini F. Phylogeographical pattern of Euplotes nobilii, a protist ciliate with a bipolar biogeographical distribution. Mol Ecol 2013; 22:4029–37.
34. Alimenti C, Vallesi A, Pedrini B, Wüthrich K, Luporini P. Molecular cold-adaptation: comparative analysis of two homologous families of psychrophilic and mesophilic signal proteins of the protozoan ciliate, Euplotes. IUBMB Life 2009; 61:838–45.
35.  Cazzolli G, Skrbic T, Guella G, Faccioli P. Unfolding thermodynamics of cysteine-rich proteins and molecular thermal-adaptation of marine ciliates. Biomolecules 2013; 3:967_85.
36. Schulze-Dieckhoff H, Freiburg M, Heckmann K. The isolation of gamone 3 and 4 of Euplotes octocarinatus. Eur J Biochem 1987; 168:89–94.
37. Alimenti C, Vallesi A, Federici S, Di Giuseppe G, Dini F, Carratore V, Luporini P. Isolation and structural characterization of two water-borne pheromones from Euplotes crassus, a ciliate commonly known to carry membrane-bound pheromones. J Eukaryot Microbiol 2011; 58:234–41.
38. Brünen-Nieveler C, Weiligmann JC, Hansen B, Kuhlmann HW, Möllenbeck M, Heckmann K. The pheromones and pheromone genes of new stocks of the Euplotes octocarinatus species complex. Eur J Protistol 1998; 34:124–32.
39. Möllenbeck M, Heckmann K. Characterization of two genes encoding a fifth so far unknown pheromone of Euplotes octocarinatus. Eur J Protistol 1999; 35:225–30.
40. Kuhlmann HW, Brünen-Nieveler C, Heckmann K. Pheromones of the ciliate Euplotes octocarinatus not only induce conjugation but also function as chemoattractants. J Exp Zool 1997; 277:38–48.
41. Vallesi A, Alimenti C, Federici S, Di Giuseppe G, Dini F, Guella G, Luporini P. Evidence for gene duplication and allelic codominance (not hierarchical dominance) at the mating type locus of the ciliate, Euplotes crassus. J Eukaryot Microbiol 2014; 61:620–9.
42. Nobili R, Luporini P, Dini F. Breeding system, species relationships and evolutionary trends in some marine species of Euplotidae (Ciliata Hypotrichida). In: Battaglia B, Beardmore J, editors. Marine Organisms: Genetics, Ecology and Evolution. New York: Plenum Press; 1978, p. 591–616.
43. Luporini P, Miceli C. Mating pheromones. In: Gall JG, editor. The Molecular Biology of Ciliated Protozoa. New York: Academic Press; 1986, p. 263–99.
44. Vallesi A, Giuli G, Bradshaw RA, Luporini P. Autocrine mitogenic activity of pheromones produced by the protozoan ciliate Euplotes raikovi. Nature 1995; 376:522–4.
45. Miceli C, La Terza A, Bradshaw RA, Luporini P. Identification and structural characterization of a cDNA clone encoding a membrane-bound form of the polypeptide pheromone Er-1 in the ciliate protozoan Euplotes raikovi. Proc Natl Acad Sci USA 1992; 89:1988–92.
46. Ortenzi C, Alimenti C, Vallesi A, Di Pretoro B, La Terza A, Luporini P. The autocrine mitogenic loop of the ciliate Euplotes raikovi: the pheromone membrane-bound forms are the cell binding sites and potential signaling receptors of soluble pheromones. Mol Biol Cell 2000; 11:1445–55.
47.  Luporini P, Alimenti C, Ortenzi C, Vallesi A. Ciliate mating types and their specific protein pheromones. Acta Protozool  2005; 44:89–101.
48. Cheek S, Krisna SS, Grishin NV. Structural classification of small, disulfide-rich protein domains. J Mol Biol 2006; 359:215–37.
49. Painter SD, Chong MG, Wong MA, Gray A, Cormier JG, Nagle GT. Relative contributions of the egg layer and egg cordon to pheromonal attraction and the induction of mating and egg-laying behavior in Aplysia. Biol Bull 1991; 181:81–94.
50. Schein CH, Nagle GT, Page JS, Sweedler JV, Xu Y, Painter SD, et al. Aplysia attractin : biophysical characterization and modeling of a water-borne pheromone. Biophys J 2001 ; 81 :463–72.
51. Painter SD, Cummuns SF, Nichols AE, Akalal DBG, Schein CH, Braun W, et al. Structural and functional analysis of Aplysia attractins, a family of water-borne protein pheromones with interspecific attractiveness. Proc Natl Acad Sci USA 2004; 101:6929–33.
52. Ortenzi C, Miceli C, Bradshaw RA, Luporini P. Identification and initial characterization of an autocrine pheromone receptor in the protozoan ciliate Euplotes raikovi. J Cell Biol 1990; 111:607–14.
53. Wang X, Rickert M, Garcia KC. Structure of the quaternary complex of interleukin-2 with its alpha, beta, and gamma receptors. Science 2005; 310:1159–63.
54.  Vallesi A, Giuli G, Ghiara P, Scapigliati G, Luporini P. Structure-function relationships of pheromones of the ciliate Euplotes raikovi with mammalian growth factors: cross-reactivity between Er-1 and IL-2 systems. Exp Cell Res 1998; 241: 253–9.
55.  Cervia D, Catalani E, Belardinelli MC, Perrotta C, Picchietti S, Alimenti C, et al. The protein pheromone Er-1 of the ciliate Euplotes raikovi stimulates human T-cell activity: involvement of interleukin-2 system. Exp Cell Res 2013; 319:56–67.
56. Ortenzi C, Luporini P. Competition among homologous polypeptide pheromones of the ciliated Euplotes raikovi for binding to each other’s cell receptor. J Eukaryot Microbiol 1995; 42:242–8.
 
 
Figure Legends
 
Figure 1. Euplotes raikovi pheromone family. (A) Multiple alignment of nine distinct sequences of pheromone precursors, as deduced from their DNA coding sequences. Sequence alignment is based on the Clustal W algorithm and optimized manually by gap insertions. Residues conserved among all sequences, or all but one sequence are shadowed. The positions of the cysteine residues are highlighted in bold. The numbers of residues of the pre, pro and secreted regions are reported on the right of the sequences. (B) Ribbon presentations (as visualized by the program MOLMOL) of the six pheromone structures (side view) that have been determined by NMR analysis of native protein preparations. The Protein Data Bank (PDB) codes are the following ones: 1erc (Er-1), 1erd (Er-2), 1erp (Er-10), 1ery (Er-11), 1hd6 (Er-22), and 1ha8 (Er-23). The amino and carboxyl termini of each molecule are labeled N and C, respectively, and the sulfur atoms involved in the disulfide bonds are indicated as spheres. The three more regular a-helices common to all six molecules are labeled h1 to h3 starting from the N-terminus and shown in dark shading.
 
Figure 2. Euplotes nobilii pheromone family. (A) Multiple alignment of seven distinct sequences of pheromone precursors, as deduced from their DNA coding sequences. (B) Ribbon presentations (as visualized by the program MOLMOL) of the four pheromone structures that have been determined by NMR analysis of native protein preparations. The Protein Data Bank (PDB) codes are the following ones: 2nsv (En-1), 2nsw (En-2), 2jms (En-6), 2kk2 (En-A1). Indications and symbols as in Fig. 1.
 
Figure 3. Euplotes octocarinatus pheromone family. Multiple alignment of nine distinct sequences of pheromone precursors, as deduced from the determination of their DNA coding sequences. Indications as in Fig. 1.
 
Figure 4. Euplotes crassus pheromones. Multiple alignment of four distinct sequences of pheromone precursors, as deduced from their DNA coding sequences. Three sequences (Ec-1, Ec-2 and Ec-4) are cell-type specific and one (Ec-a) is shared identical among different cell types. Indications as in Fig. 1.
 
Figure 5. Intermolecular helix-helix interactions in the crystal structure of the E. raikovi pheromone Er-1. (A) Structure of the symmetrical dimer 1 in which two molecules are related by a twofold axis (). Each monomer involves helices 1 and 2, delimiting face a, in the formation of a four-helix bundle. (B) Structure of the asymmetrical dimer 2 in which two molecules are related by a twofold screw axis (). Each monomer stacks helix 3, and its adjacent faces b and c, in an anti-↑parallel fashion to form a linear structure without symmetrical contacts. The Er-1 molecule packing into the crystal lattice is shown in the boxed diagrammatic presentation. Shadowed molecules mimic ligand binding (pheromone-receptor) molecules oriented with their C-terminus towards the cell outside, while light molecules mimic soluble pheromone molecules oriented with their C-terminus towards the cell inside. After Weiss et al. (22).
 
Figure 6. Effects of E. raikovi pheromones on human lymphocyte Jurkat cells. (A) Viability of Jurkat cells cultured for 48h in the absence (control), or in the presence of increasing concentrations of pheromone Er-1. Histograms represent the mean ± SEM of data from at least four independent experiments. Data are expressed by setting the control samples as 100%. (B) Measurements by quantitative PCR of the mRNA levels of different cytokines in Jurkat cells grown in the presence of 10ng/ml Er-1 for 48h in restrictive culture conditions. Each histogram represents the mean ± SEM of data from at least three independent experiments. Fold changes were determined putting equal to 1 the mRNA levels in cells grown in the absence of Er-1 (control). After Cervia et al. (55)