A genetic approach to the species problem in wild potato.

Wild potatoes are native to the Americas, where they present very wide geographical and ecological distribution. Most are diploid, obligate out-crossers due to a multiallelic gametophytic self-incompatibility (S) locus that prevents self-fertilisation and crossing between individuals carrying identical S-alleles. They have two alternative modes of reproduction: sexual (by seeds) and asexual (by stolons and tubers), which provide, respectively, for genetic flexibility in changing environments and high fitness of adapted genotypes under stable conditions. Since the early twentieth century, their taxonomic classification has been mostly based on morphological phenotypes (Taxonomic Species Concept). More recently, attempts have been made to establish phylogenetic relationships, applying molecular tools in samples of populations (accessions) with a previously assigned specific category. However, neither the reproductive biology and breeding relations among spontaneous populations nor the morphological and genetic variability expected in obligate allogamous populations are considered when the taxonomic species concept is applied. In nature, wild potato populations are isolated through external and internal hybridisation barriers; the latter, which are genetically determined, can be either pre-zygotic (pollen-pistil incompatibility) or post-zygotic (abortion of embryo, endosperm or both tissues, sterility, and hybrid weakness and breakdown in segregating generations). The internal barriers, however, can be incomplete, providing opportunities for hybridisation and introgression within and between populations and ploidy levels in areas of overlap. The widespread occurrence of spontaneous hybrids in nature was recognised in the mid-twentieth century. Using genetic approaches, results have been obtained that provide strong support to the assertion that populations are at different stages of genetic divergence and are not at the end of the evolutionary process, as presupposed by the Taxonomic Species Concept. Furthermore, since wild potatoes have uniparental and biparental overlapping generations, the Biological Species Concept - developed for sexually reproducing biparental organisms - cannot be applied to them. In this paper, morphological, genetic, molecular and taxonomic studies in wild potato are reviewed, considering the genetic consequences of their reproductive biology, in an attempt to shed light on the species problem, because of its relevance in germplasm conservation and breeding.

Meiotic abnormalities underlying pollen sterility in wild potato hybrids and spontaneous populations.

Wild potato species are widely distributed in the Americas, where they spontaneously grow in very diverse habitats. These species - with low chromosome differentiation - form polyploid series with 2n = 2x, 3x, 4x and 6x (x =12). They are isolated in nature by external and internal hybridisation barriers that can be incomplete, allowing hybridisation in areas of sympatry. Nevertheless, most accessions in germplasm banks, regardless of genetic background of the sampled spontaneous populations, have been assigned specific categories based on morphological characters. To further investigate the extent of hybridisation in the group and for comparative purposes, pollen viability was estimated in (i) artificial hybrids between a commercial cultivar (Calén INTA) of the common potato (tetraploid Solanum tuberosum ssp. tuberosum) and the tetraploid cytotype of the related wild species S. gourlayi, and (ii) samples of plants (accessions) and inflorescences of natural populations from Argentina, tentatively classified as 'presumed hybrids' (S. infundibuliforme-S. gourlayi) and 'species' (S. infundibuliforme, S. gourlayi and S. chacoense). Regardless of origin, 98 out of 103 plants analysed had zero to 70% pollen viability (zero to 40% in eight of them). Pollen grains were of variable size and morphology and, in mostly male sterile plants, the only viable pollen grains were 2n and/or 4n. Furthermore, male sterile plants shared various abnormalities in meiosis I and II (unpaired chromosomes, unequal chromosome distribution, precocious/lagging chromosomes, parallel, tripolar, fused and multiple spindles, unequal size nuclei, dyads, triads and pentads in addition to normal tetrads, among others). These results provide novel evidence to support field observations of early potato botanists on the extent of spontaneous hybridisation in wild Argentinian potato populations, which is not reflected in the current taxonomy and has significant consequences for germplasm conservation and breeding.

Interlobe communication in multiple calcium-binding site mutants of Drosophila calmodulin.

We have generated mutants of Drosophila calmodulin in which pairs of calcium-binding sites are mutated so as to prevent calcium binding. In all sites, the mutation involves replacement of the -Z position glutamate residue with glutamine. Mutants inactivated in both N-terminal sites (B12Q) or both C-terminal sites (B34Q), and two mutants with one N- and one C-terminal site inactivated (B13Q and B24Q) were generated. The quadruple mutant with all four sites mutated was also studied. UV-difference spectroscopy and near-UV CD were used to examine the influence of these mutations upon the single tyrosine (Tyr-138) of the protein. These studies uncovered four situations in which Tyr-138 in the C-terminal lobe responds to a change to the calcium-binding properties of the N-terminal lobe. Further, they suggest that N-terminal calcium-binding events contribute strongly to the aberrant behavior of Tyr-138 seen in mutants with a single functional C-terminal calcium-binding site. The data also indicate that loss of calcium binding at site 1 adjusts the aberrant conformation of Tyr-138 produced by mutation of site 3 toward the wild-type structure. However, activation studies for skeletal muscle myosin light chain kinase (SK-MLCK) established that all of the multiple binding site mutants are poor activators of SK-MLCK. Thus, globally, the calcium-induced conformation of B13Q is not closer to wild type than that of either the site 1 or the site 3 mutant. The positioning of Tyr-138 within the crystal structure of calmodulin suggests that effects of the N-terminal lobe on this residue may be mediated via changes to the central linker region of the protein.

Activation of four enzymes by two series of calmodulin mutants with point mutations in individual Ca2+ binding sites.

Activation of four target enzymes by two series of calmodulin Ca2+ binding site mutants has been examined. In each mutant, the conserved bidentate glutamate of one of the Ca2+ binding sites is mutated to glutamine or lysine. The enzymes studied were smooth and skeletal muscle myosin light chain kinases, adenylylcyclase, and plasma membrane Ca(2+)-ATPase. For the first three enzymes, the activation patterns with the two mutant series were very similar: mutation of site 4 was most deleterious, then site 2, site 3, and site 1. This ranking was observed previously in Ca2+ binding and Ca(2+)-induced conformational studies of these mutants. Thus the response of these enzymes is probably determined by the extent to which each mutant's competence to interact with target binding regions has been compromised. In contrast, for Ca(2+)-ATPase, mutants of sites 3 and 4 were much poorer activators than those of sites 1 and 2. Events beyond calmodulin binding and related to enzyme activation probably dictate this unusual activation pattern and also the anomalously poor activation of skeletal muscle myosin light chain kinase by site 1 mutant B1Q. Site 1 mutant B1K showed wild type activation of all four enzymes suggesting that in site 1, the lysine substitution can evoke the conformational changes associated with Ca2+ binding.

Circular dichroism studies on calcium binding to two series of Ca2+ binding site mutants of Drosophila melanogaster calmodulin.

The Ca(2+)-induced structural changes in mutant calmodulins from Drosophila melanogaster have been studied by circular dichroism. The proteins comprise eight site-specific mutants, in which a bidentate glutamic acid (at position 12 in each Ca2+ binding loop) is replaced with either glutamine (BQ series) or lysine (BK series). Previous studies of these proteins indicate that Ca2+ binding at the mutated site is effectively eliminated by each of these substitutions, with additional effects at nonmutated sites. Circular dichroism has now been used to assess Ca(2+)-induced changes in secondary and tertiary structure in these proteins. In the absence of Ca2+, the helical content of these mutant calmodulins is close to that of the wild-type protein. In excess Ca2+, calmodulins with a mutation in the N-terminal sites show Ca(2+)-induced increases in helicity (CD at 222 nm) that are similar to those of the wild-type protein. In contrast, much less additional helix is induced by Ca2+ in calmodulins with mutations in the C-terminal sites, with the two mutations to site IV showing a particularly poor response. Ca(2+)-induced changes to the environment of the single tyrosine of Drosophila calmodulin (Tyr-138 in site IV of the C-terminal domain) have been monitored via CD at 280 nm. The signal from this residue is significantly altered in the Ca(2+)-free form of almost all these mutants, including those in the N-terminal domain. This indicates significant interaction between the N- and C-terminal domains of these mutants.(ABSTRACT TRUNCATED AT 250 WORDS)

Stopped-flow studies of calcium dissociation from calcium-binding-site mutants of Drosophila melanogaster calmodulin.

The kinetics of calcium dissociation from two groups of site-specific mutants of calmodulin from Drosophila melanogaster have been studied by stopped-flow kinetic methods, using the fluorescent calcium chelator 8-amino-2-[(2-amino-5-methylphenoxy)methyl]-6- methoxyquinoline-N,N,N',N'-tetraacetic acid (Quin 2). The BQ series of mutants consists of four proteins in which one of the four bidentate glutamate residues (Glu12 of each of the four calcium binding loops) has been replaced by glutamine. In the BK series of mutants, the corresponding glutamate has been replaced by lysine. Calcium-dissociation kinetics of proteins with a mutation in site I or II (N-terminal domain) are consistent with a model in which the mutation weakens binding at the non-mutated N-terminal partner site and has a small, but significant, effect on the kinetic properties of sites III and IV (C-terminal domain). The proteins with a mutation in site III or IV show a large effect, with decreased Ca2+ dissociation rate from the unmodified N-terminal Ca(2+)-binding sites I and II. A structural interpretation is proposed, based on enhanced interactions between the domains when the affinity of individual sites have been dramatically reduced by mutation. This effect is greatest for the mutations in the C-terminal domain, which appear to destroy the co-operativity of Ca2+ binding at sites III and IV. The results show that site-specific mutation can have surprisingly far-ranging effects on kinetic properties of calmodulin. The kinetic analysis also shows that studies of specifically engineered mutants may in principle help to unmask the values of intrinsic rate constants for the wild-type protein which are not normally observable in the process of Ca2+ dissociation.

Ca2+ binding and conformational change in two series of point mutations to the individual Ca(2+)-binding sites of calmodulin.

Two series of site-directed mutations to the individual Ca(2+)-binding sites of Drosophila melanogaster calmodulin have been generated and studied. In each mutant, a conserved glutamic acid residue at position 12 in all of the Ca(2+)-binding loops has been mutated in one site. In one series the residue is changed to glutamine; in the second series the change is to lysine. The Ca(2+)-binding properties of these mutants and the wild-type protein under pseudo-physiological conditions are presented. In addition, Ca(2+)-induced changes to the environment of the single tyrosine residue (Tyr-138) have been studied for some of the mutants. Ca2+ binding to the wild-type protein is best modeled as two pairs of sites with a higher affinity pair that shows strong cooperativity. For all but one of these eight mutant proteins, only three Ca(2+)-binding events can be detected. In three of the amino-terminal mutants, the three residual sites are (i) a pair of relatively high affinity sites and (ii) a weakened low affinity site. For all four carboxyl-terminal mutations, the residual sites are three relatively low affinity sites. In general, mutations to sites 2 and 4 prove more deleterious than mutations to sites 1 and 3. The Ca(2+)-induced conformational changes in the vicinity of Tyr-138 are relatively undisturbed by mutations of site 1. However, the changes to Tyr-138 in the carboxyl-terminal site mutants indicate that upon disruption of the cooperative binding at the high affinity sites, conformational change in the carboxyl terminus occurs in two phases. It appears that binding of Ca2+ to either carboxyl-terminal site can elicit the first phase of the response but the second phase is almost abolished when site 4 is the mutated site. The final conformations of site 3 and 4 mutants are thus significantly different.

Structure of a recombinant calmodulin from Drosophila melanogaster refined at 2.2-A resolution.

The crystal structure of calmodulin (Mr 16,700, 148 residues) from Drosophila melanogaster as expressed in a bacterial system has been determined and refined at 2.2-A resolution. Starting with the structure of mammalian calmodulin, we produced an extensively refitted and refined model with a conventional crystallographic R value of 0.197 for the 5,239 reflections (F greater than or equal to 2 sigma (F)) within the 10.0-2.2-A resolution range. The model includes 1,164 protein atoms, 4 calcium ions, and 78 water molecules and has root mean square deviations from standard values of 0.018 A for bond lengths and 0.043 A for angle distances. The overall structure is similar to mammalian calmodulin, with a seven-turn central helix connecting the two calcium-binding domains. The "dumb-bell" shaped molecule contains seven alpha-helices and four "EF hand" calcium-binding sites. Although the amino acid sequences of mammalian and Drosophila calmodulins differ by only three conservative amino acid changes, the refined model reveals a number of significant differences between the two structures. Superimposition of the structures yields a root mean square deviation of 1.22 A for the 1,120 equivalent atoms. The calcium-binding domains have a root mean square deviation of 0.85 A for the 353 equivalent atoms. There are also differences in the amino terminus, the bend of the central alpha-helix, and the orientations of some of the side chains.

Structure and sequence of the Drosophila melanogaster calmodulin gene.

A series of phage clones overlapping the single calmodulin gene locus of Drosophila melanogaster has been isolated and the exons of the gene positioned and sequenced within these clones. A calmodulin cDNA clone of the electric eel was used to identify these clones and to position the two major protein-coding exons of the gene. cDNA clones for D. melanogaster calmodulin were then isolated, characterized and used to identify the remaining exons. The gene consists of four exons separated by three introns of 3400 to 4300 bases in length. Exon 1 consists of the 5' untranslated region and the initiator ATG; exon 2 encodes amino acid residues 1 to 58.3; exon 3 encodes residues 58.3 to 139.3; and exon 4 encodes residues 139.3 to 148 and the 3' untranslated region. From the sequence of the 3' untranslated region and the lengths of the cDNA clones, two or three polyadenylation sites are indicated. Sequences potentially involved in the control of transcription of the gene and splicing of the mRNA product have been identified. Comparison of the intron-exon structures of the D. melanogaster calmodulin gene, the chick calmodulin gene, and other genes of the troponin C superfamily reinforces previous hypotheses that these genes arose from a common progenitor and permits identification of four introns that were probably present in the progenitor gene structure. The D. melanogaster calmodulin gene contains three of these introns, and the chick gene contains all four. These gene comparisons also indicate that the region of these genes encoding Ca2+-binding loop 3 is highly variable in structure. The chick and D. melanogaster calmodulin genes differ in this region, the chick gene containing a fifth intron here that is absent from the D. melanogaster gene.