Unidirectional gene conversions in the chloroplast of Chlamydomonas interspecific hybrids

Summary With the goal of studying directly the inheritance and recombination of physically mapped markers on the chloroplast genome, we have recently identified and localized physical differences between the chloroplast DNAs (cpDNAs) of the interfertile algae Chlamydomonas eugametos and C. moewusii. Here we report the inheritance patterns of 24 polymorphic loci mapping throughout the chloroplast genome in hybrids recovered from reciprocal crosses between the two algae. Most polymorphic loci were found to be inherited mainly from the mt ⁺ parent, with no apparent preference for one or the other parental alternatives in reciprocal crosses. Virtually all hybrids, however, inherited exclusively the long alleles of three loci; i.e. an intron in the large subunit ribosomal RNA gene of C. eugametos, a 21 kbp sequence addition in the inverted repeat of the C. moewusii cpDNA and a 5.8 kbp sequence addition in one of the single-copy regions of C. moewusii cpDNA. As these alleles are derived from opposite parental strains, their unidirectional inheritance in hybrids results necessarily from interspecific recombination of cpDNA molecules. We propose that gene conversion events led to the spreading of the long alleles of the three loci.     

Conserved gene clusters in the highly rearranged chloroplast genomes of Chlamydomonas moewusii and Chlamydomonas reinhardtii

We have extended to about 75 the number of genes mapped on the Chlamydomonas moewusii and Chlamydomonas reinhardtii chloroplast DNAs (cpDNAs) by partial sequencing of the very closely related C. eugametos and C. moewusii cpDNAs and by hybridizations with Chlamydomonas chloroplast gene-specific sequences. Only four of these genes (tscA and three reading frames) have not been identified in any other algal cpDNAs and thus may be specific to Chlamydomonas. Although the C. moewusii and C. reinhardtii cpDNAs differ by complex sequence rearrangements, 38 genes scattered throughout the genome define 12 conserved clusters of closely linked loci. Aside from the rRNA operon, four of these gene clusters share similarity to evolutionarily primitive operons found in other cpDNAs, representing in fact remnants of these operons. Our results thus indicate that most of the ancestral bacterial operons that characterize the chloroplast genome organization of land plants and early-diverging photosynthetic eukaryotes have been disrupted before the emergence of the polyphyletic genus Chlamydomonas. All gene rearrangements between the C. moewusii and C. reinhardtii cpDNAs, with the exception of those accounting for the relocations of atpA, psbI and rbcL, occurred within corresponding regions of the genome. One of these rearrangements seems to have led to disruption of the ancestral region containing rpl23, rpl2, rps19, rpl16, rpl14, rpl5, rps8 and the psaA exon 1. This gene cluster, which bears striking similarity to the Escherichia coli S10 and spc operons, spans a continuous DNA segment in C. reinhardtii, while it maps to two separate fragments in C. moewusii.     

Recombination of Chlamydomonas chloroplast DNA occurs more frequently in the large inverted repeat sequence than in the single-copy regions

Summary It is well documented that chloroplast DNA (cpDNA) recombination occurs at a relatively high frequency during sexual reproduction of unicellular green algae from the Chlamydomonas genus. Like the cpDNAs of most land plants, those of Chlamydomonas species are divided into two single-copy regions by a large inverted repeat sequence, part of which encodes the chloroplast rRNA genes. In the present study, we scored the inheritance of polymorphic loci spanning the entire chloroplast genome in hybrids recovered from reciprocal interspecific and F₁ crosses between Chlamydomonas eugametes and C. moewusii, and from these data, estimated the density of recombination junctions within each region of recombinant cpDNAs. Our results indicate that recombination junctions occur at highly variable frequencies across the three main domains of the chloroplast genome. The large inverted repeat sequence was found to exhibit at least a five-fold higher density of recombination junctions compared to one of the singlecopy regions, whereas junctions in the latter region were five-fold more abundant relative to those in the other single-copy region. This marked difference in the densities of recombination junctions implies that the extent of genetic linkage between two given chloroplast loci will depend not only on their physical distance, but also on their locations within the genome.     

A 21 kilobase-pair deletion/addition difference in the inverted repeat sequence of chloroplast DNA from Chlamydomonas eugametos and C. moewusii

Summary Our recent physical mapping of chloroplast DNA (cpDNA) from Chlamydomonas moewusii, a unicellular green alga which is interfertile with Chlamydomonas eugametos, has revealed a two-fold size difference between the inverted repeat sequences of these algae. With a size of 42 kbp, the inverted repeat of C. moewusii is the largest yet identified in any chloroplast genome. Here we have compared the arrangement of conserved sequences within the two algal inverted repeats by hybridizing cloned restriction fragments representing over 90% of these repeats to Southern blots of cpDNA digests from the two algae. We found that the size difference between the two algal inverted repeats is due to the presence of an extra DNA segment of 21 kilobase pairs (kbp) in C. moewusii. Except for this sequence, the C. moewusii inverted repeat is highly homologous to the entire C. eugametos repeat and the arrangement of conserved sequences in the two repeats is identical. Southern hybridizations with specific gene probes revealed that the conserved sequences include the rDNA region and the genes coding for the large subunit of ribulose 1,5 bisphosphate carboxylase-oxygenase (rbcL) and for the 32 kilodalton thylakoid membrane protein (psbA). With respect to the conserved sequences, the extra 21 kbp DNA segment of C. moewusii lies in the region of psbA, most probably slightly downstream from this gene.     

Extensive gene rearrangements in the chloroplast DNAs of Chlamydomonas species featuring multiple dispersed repeats

We have constructed a physical and gene map for the chloroplast DNA (cpDNA) of the unicellular green alga Chlamydomonas gelatinosa, a close relative of Chlamydomonas reinhardtii. At 285 kb, the C. gelatinosa cpDNA is 89 kb larger than its C. reinhardtii counterpart. The alterations in the order of 77 genes on the cpDNAs of these green algae are attributable to nine inversions and one event of expansion/contraction of the inverted repeat. These rearrangements are much more extensive than those previously reported between the cpDNAs of the closely related Chlamydomonas moewusii and Chlamydomonas pitschmannii. Because the divergence level of the C. gelatinosa and C. reinhardtii chloroplast-encoded large subunit rRNA gene sequences is equivalent to that of the corresponding C. moewusii and C. pitschmannii sequences, our results may suggest that, in the same period of time, there have been more numerous rearrangements in the lineage comprising C. gelatinosa and C. reinhardtii than in the lineage comprising C. moewusii and C. pitschmannii. Alternatively, given that substitution rates in chloroplast genes are not necessarily uniform across lineages, the extensive rearrangements between the C. gelatinosa and C. reinhardtii cpDNAs may reflect a longer divergence period for this pair of Chlamydomonas species compared to that for the C. moewusii/C. pitschmannii pair. We have also found that, like its C. reinhardtii homologue but unlike its C. moewusii and C. pitschmannii counterparts, the C. gelatinosa cpDNA features a large number of dispersed repeated sequences that are readily detectable by Southern blot hybridization with homologous fragment probes. Assuming that the two pairs of closely related Chlamydomonas species diverged at about the same time, these data suggest that the susceptibility of Chlamydomonas cpDNAs to rearrangements is correlated with the abundance of repeated sequences. Preliminary characterization of a 345-bp C. gelatinosa cpDNA region containing a repeated sequence by both DNA sequencing and Southern blot analysis has revealed no sequence homology between this region and the cpDNAs of C. reinhardtii and other Chlamydomonas species.      

Physical map of chloroplast DNA in sugi,Cryptomeria japonica

Summary To investigate the evolution of conifer species, we constructed a physical map of the chloroplast DNA of sugi, Cryptomeria japonica, with four restriction endonucleases, PstI, SalI, SacI and XhoI. The chloroplast genome of C. japonica was found to be a circular molecule with a total size of approximately 133 kb. This molecule lacked an inverted repeat. Twenty genes were localized on the physical map of C. japonica cpDNA by Southern hybridization. The chloroplast genome structure of C. japonica showed considerable rearrangements of the standard genome type found in vascular plants and differed markedly from that of tobacco. The difference was explicable by one deletion and five inversions. The chloroplast genome of C. japonica differed too from that of the genus Pinus which also lacks one of the inverted repeats. The results indicate that the conifer group originated monophyletically from an ancient lineage, and diverged independently after loss of an inverted repeat structure.     

Physical map and gene localization on sunflower (Helianthus annuus) chloroplast DNA: evidence for an inversion of a 23.5-kbp segment in the large single copy region

As a first step in the study of chloroplast genome variability in the genus Helianthus, a physical restriction map of sunflower (Helianthus annuus) chloroplast DNA (cpDNA) has been constructed using restriction endonucleases BamH I, Hind III, Pst I, Pvu II and Sac. I. Sunflower circular DNA contains an inverted repeat structure with the two copies (23 kbp each) separated by a large (86 kbp) and a small (20 kbp) single copy region. Its total length is therefore about 152 kbp. Sunflower cpDNA is essentially colinear with that of tobacco with the exception of an inversion of a 23.5-kbp segment in the large single copy region. Gene localization on the sunflower cpDNA and comparison of the gene map with that from tobacco chloroplasts have revealed that the endpoints of the inversion are located between the trnT and trnE genes on the one hand, and between the trnG and trnS genes on the other hand.Analysis of BamH I restriction fragment patterns of H. annuus, H. occidentalis ssp. plantagineus, H. grossesseratus, H. decapetalus, H. giganteus, H. maximiliani and H. tuberosus cpDNAs suggests that structural variations are present in the genus Helianthus.     

Distinctive architecture of the chloroplast genome in the chlorophycean green alga Stigeoclonium helveticum

The chloroplast genome has experienced many architectural changes during the evolution of chlorophyte green algae, with the class Chlorophyceae displaying the lowest degree of ancestral traits. We have previously shown that the completely sequenced chloroplast DNAs (cpDNAs) of Chamydomonas reinhardtii (Chlamydomonadales) and Scenedesmus obliquus (Sphaeropleales) are highly scrambled in gene order relative to one another. Here, we report the complete cpDNA sequence of Stigeoclonium helveticum (Chaetophorales), a member of a third chlorophycean lineage. This genome, which encodes 97 genes and contains 21 introns (including four putatively trans-spliced group II introns inserted at novel sites), is remarkably rich in derived features and extremely rearranged relative to its chlorophycean counterparts. At 223,902 bp, Stigeoclonium cpDNA is the largest chloroplast genome sequenced thus far, and in contrast to those of Chlamydomonas and Scenedesmus, features no large inverted repeat. Interestingly, the pattern of gene distribution between the DNA strands and the bias in base composition along each strand suggest that the Stigeoclonium genome replicates bidirectionally from a single origin. Unlike most known trans-spliced group II introns, those of Stigeoclonium exhibit breaks in domains I and II. By placing our comparative genome analyses in a phylogenetic framework, we inferred an evolutionary scenario of the mutational events that led to changes in genome architecture in the Chlorophyceae.Electronic supplementary material Supplementary material is available in the online version of this article at and is accessible for authorized users.Nucleotide sequence data reported are available in the GenBank database under the accession number DQ630521.     

A linear DNA molecule of 5.9 kilobase-pairs is highly homologous to the chloroplast DNA in the green alga Chlamydomonas moewusii

Summary The unicellular green alga Chlamydomonas moewusii contains small DNA species of unknown cellular location. We report that the most abundant of these DNAs, here designated low-molecular-weight DNA (LMW DNA), is a linear molecule of 5.9 kilobase pairs (kbp). Southern blot hybridization and restriction enzyme analysis revealed that the LMW DNA sequence also exists as an integrated sequence in a discrete region of the chloroplast genome. We have confirmed earlier reports that small DNA species related to the LMW DNA are absent from Chlamydomonas eugametos, an alga which is interfertile with C. moewusii. In the C. eugametos chloroplast genome we found only remnants of the LMW DNA sequence.     

ORGANIZATION OF THE CHLOROPLAST GENOME OF THE FRESHWATER CENTRIC DIATOM CYCLOTELLA MENEGHINIANA1

We constructed a complete physical map and a partial gene map of the chloroplast genome of Cyclotella meneghiniana Kützing clone 1020-1a (Bacillariophyceae). The 128-kb circular molecule contains a 17-kb inverted repeat, which divides the genome into single copy regions of65 kb and 29 kb. This is the largest genome and inverted repeat found in any diatom examined to date. In addition to the 16S and 23S ribosomal RNA genes, the inverted repeat contains both the ndhD gene (as yet unexamined in other diatoms) and the psbA gene (located similarly in one of two other examined diatoms). The Cyclotella chloroplast genome exists as two equimolar populations of inversion isomers that differ in the relative orientation of their single copy sequences. This inversion heterogeneity presumably results from intramolecular recombination within the inverted repeat. For the first time, we map the ndhD, psaC, rpofi, rpoCl, and rpoC2 genes to the chloroplast genome of a chlorophyll c-containing alga. While the Cyclotella chloroplast genome retains some prokaryotic and land plant gene clusters and operons, it contains a highly rearranged gene order in the large and small single copy regions compared to all other examined diatom, algal, and land plant chloroplast genomes.     

Chloroplast DNA differences between two species of Oenothera subsection Munzia: location in the chloroplast genome and relevance to possible interactions between nuclear and plastid genomes

Summary The chloroplast DNAs (cpDNAs) of Oenothera berteriana and Oe. odorata (subsection Munzia) were examined by restriction endonuclease analysis with Sal I, Pvu II, Kpn I, Pst I, Hind III, and Bam HI. The fragment patterns show that these cpDNAs have all 133 restriction sites in common as well as a lot of individual bands. Nevertheless the cpDNAs of the two species can be distinguished by distinct differences in size between a small number of fragments. The 42 cleavage sites produced by Sal I, Pvu II and Kpn I were mapped on the circular cpDNAs. This was achieved by an approach which combined experimental and mathematical procedures. The overall serial order of the fragments was found to be the same for both cpDNAs. The size differences of individual fragments in the Sal I, Pvu II and Kpn I patterns between Oe. berteriana and Oe. odorata cpDNA are located within five regions scattered along the plastid chromosome. Two of these regions have been localized in the larger and one in the smaller of the two single-copy parts of the cpDNA molecule. The remaining two overlap the borders between the large single-copy and each of the duplicated parts of the molecule. The positions of distinct restriction sites are altered among the two Oenothera plastome DNAs by 0.02–0.4 MDa (30–600 base pairs). These alterations probably result from insertions/deletions.Abbreviations cpDNA chloroplast, plastid DNA - Oe. Oenothera - MDa Megadalton - rRNA, rDNA ribosomal RNA, DNA Dedicated to Professor Berthold Schwemmle, Tübingen, on the occasion of his 60th birthday     

Characterization and phylogenetic distribution of a chloroplast DNA rearrangement in theBerberidaceae

Restriction site maps and a clone bank of chloroplast DNA (cpDNA) ofMahonia higginsae (Munz)Ahrendt (Berberidaceae) were constructed. The size ofMahonia cpDNA was about 167 kb. Precise mapping using gene probes revealed that cpDNA ofM. higginsae has an inverted repeat (IR) 11.5 kb larger than the tobacco IR. The expansion of the IR into the large single copy region has resulted in the duplication of at least ten genes includingpsbB. The phylogenetic distribution of the expanded IR was examined in twenty-five species ofBerberis andMahonia, twenty species representing the fifteen remaining genera of theBerberidaceae, and four species from four allied families. Our survey indicates that only the species of the closely related generaBerberis andMahonia share the 11.5kb expansion of IR. This result supports their close phylogenetic relationship, which has been suggested previously by chromosomal, morphological, and serological data.     

Diversity and evolution of chloroplast DNA in Triticum and Aegilops as revealed by restriction fragment analysis

Summary Restriction fragment analysis of chloroplast (cp) DNAs from 35 wheat (Triticum) and Aegilops species, including their 42 accessions, was carried out with the use of 13 restriction enzymes to clarify variation in their cpDNAs. Fourteen fragment size mutations (deletions/insertions) and 33 recognition site changes were detected among 209 restriction sites sampled. Based on these results, the 42 accessions of wheat-Aegilops could be classified into 16 chloroplast genome types. Most polyploids and their related diploids showed identical restriction fragment patterns, indicating the conservatism of the chloroplast genome during speciation, and maternal lineages of most polyploids were disclosed. This classification of cpDNAs was principally in agreement with that of the plasma types assigned according to phenotypes arising from nucleus-cytoplasm interactions. These mutations detected by restriction fragment analysis were mapped on the physical map of common wheat cpDNA, which was constructed with 13 restriction endonucleases. Length mutations were more frequently observed in some regions than in others: in a 16.0 kilo base pairs (kbp) of DNA region, including rbcL and petA genes, 6 of 14 length mutations were concentrated. This indicates that hot spot regions exist for deletions/insertions in chloroplast genome. On the other hand, 33 recognition site mutations seemed to be distributed equally throughout the genome, except in the inverted repeat region where only one recognition site change was observed. Base substitution rate (p) of cpDNA was similar to that of other plants, such as Brassica, pea and Lycopersicon, showing constant base substitution rates among related taxa and slow evolution of cpDNA compared with animal mitochondrial DNA. Phylogenetic relationships among Triticum and Aegilops species were discussed, based on the present data.Contributions no. 45 and no. 490 from the Kihara Institute for Biological Research, Yokohama City University and the Laboratory of Genetics, Faculty of Agriculture, Kyoto University, respectively.     

Chloroplast genome differences between Paspalum dilatatum Poir and the related species P. notatum Flugge

 Chloroplast DNA (cpDNA) of Paspalum dilatatum and P. notatum was digested singly or in combination with the restriction endonucleases PstI, PvuII, SalI, KpnI and XhoI. Data obtained from filter hybridization experiments with barley and wheat cpDNA probes were used to construct restriction site maps of the chloroplast genomes of the Paspalum species. The cpDNA fragments were ordered into a circular configuration of approximately 139.3 kbp that contained two inverted repeat regions of approximately 23 kbp and a small and large single-copy region of approximately 11 kbp and 83 kbp, respectively. The cpDNA maps showed that P. dilatatum and P. notatum shared a large number of restriction sites with the proportion of shared restriction sites S=0.90. No restriction site differences were detected in the KpnI maps. Eight species-specific restriction site differences that could be used to identify the cytoplasm of each Paspalum species were identified in the PstI, PvuII, SalI, and XhoI cleavage maps. The overall structural organization of the Paspalum cpDNAs is rather similar to those of most cpDNAs from other plants. The results presented in this study will be of value for exploring further phylogenetic relationships within the genus Paspalum. Received: 27 February 1997 / Accepted: 7 March 1997     

The Chlamydomonas chloroplast clpP gene contains translated large insertion sequences and is essential for cell growth

Sequence determination of the chloroplast clpP gene from two distantly related Chlamydomonas species (C. reinhardtii and C. eugametos) revealed the presence of translated large insertion sequences (IS1 and IS2) that divide the clpP gene into two or three sequence domains (SDs) and are not found in homologous genes in other organisms. These insertion sequences do not resemble RNA introns, and are not spliced out at the mRNA level. Instead, each insertion sequence forms a continuous open reading frame with its upstream and downstream sequence domains. IS1 specifies a potential polypeptide sequence of 286 and 318 amino acid residues in C. reinhardtii and C. eugametos, respectively. IS2 encodes a 456 amino acid polypeptide and is present only in C. eugametos. The two Chlamydomonas IS1 sequences show substantial similarity; however, there is no significant sequence similarity either between IS1 and IS2 or between these insertion sequences and any other known protein coding sequences. The C. reinhardtii clpP gene was further shown to be essential for cell growth, as demonstrated through targeted gene disruption by particle gun-mediated chloroplast transformation. Only heteroplasmic transformants could be obtained, even under mixotrophic growth conditions. The heteroplasmic transformants were stable only under selection pressure for the disrupted clpP, rapidly segregated into wild-type cells when the selection pressure was removed, and grew significantly more slowly than wildtype cells under phototrophic conditions.     

Chloroplast genome organization of bromegrass,Bromus inermis Leyss

Summary A physical map of the Bromus inermis chloroplast genome was constructed using heterologous probes of barley and wheat chloroplast DNA (cpDNA) to locate restriction sites. The map was aligned from data obtained from filter hybridization experiments on single and double enzyme digests. Cleavage sites for the enzymes PstI, SalI, KpnI, XhoI and PvuII were mapped. The chloroplast genome of B. inermis is similar in physical organization to that of other grasses. The circular cpDNA molecule of B. inermis has the typical small (12.8 kbp) and large (81.3 kbp) single-copy regions separated by a pair of inverted repeat (21 kbp) regions. The cpDNA molecule of B. inermis is collinear in sequence to that of wheat, rye, barley and oats. No structural rearrangements or major deletions were observed, indicating that the cpDNA of Bromus is a useful tool in phylogenetic studies.     

Chloroplast DNA variation and phylogeny of theRanunculaceae

Restriction site mapping of chloroplast DNA from 31 species representing 26 genera of theRanunculaceae was performed using eleven restriction endonucleases. The chloroplast genome varies in length from approximately 152 to 160 kb. Length variants are frequent in theRanunculaceae and range from usually less than 300 bp to rarely 1.5 kb. The inverted repeat is extended into the large single copy (LSC) region by 4–4.5 kb inAnemone, Clematis, Clematopsis, Hepatica, Knowltonia, andPulsatilla. Several inversions are present in the LSC-region of the cpDNA in all these genera and inAdonis. The frequency of restriction site mutations varies within the chloroplast genome in theRanunculaceae between 4 and 32 mutations per kilobase, and is lowest in the inverted repeat and the regions containing the ATPase-genes and the genespsaA, psaB, psbA, rpoB, andrbcL. A total of 547 phylogenetically informative restriction sites was utilized in cladistic analyses of the family using Wagner, Dollo, and weighted parsimony. These three parsimony analyses result in different tree topologies. Four, six, and one equally most parsimonious trees were obtained with Wagner, Dollo, and weighted parsimony, respectively. The amount of support for the monophyletic groups was evaluated using bootstrapping and decay analysis. All three parsimony methods suggest thatHydrastis is the sister group to the remainder of theRanunculaceae, and that theAnemone-Clematis group, which shares several derived cpDNA rearrangements, is monophyletic. Only a few of the traditional groups in theRanunculaceae are supported by cpDNA restriction side data. Only Dollo parsimony provides support for the hypothesis thatThalictroideae andRanunculoideae are monophyletic.     

Vicia faba chloroplast DNA has only one set of ribosomal RNA genes as shown by partial denaturation mapping and R-loop analysis

Summary Broad-bean (Vicia faba) chloroplast DNA (cpDNA) was isolated and characterized. The intact DNA is circular and has a molecular weight of 79.8x 10⁶ dalton. Electron microscopic analysis of self-annealed intact single-strand circles show that it does not have a large double-stranded inverse repeat as seen in spinach chloroplast DNA. Only one ribosomal RNA gene (one set of 16S and 23S rRNA sequences) was found in preparations of R-loops between the Vicia rRNA and cpDNA circles. A restriction enzyme map for SalI and KpnI was derived by comparing the partial denaturation pattern of the fragments with the pattern of the intact circle. The map was confirmed by gel analysis. The ribosomal RNA gene was localized on the SalI fragment 3b by R-loop analysis. SalI fragment 1a although it contains a G-C rich region did not form R-loops with rRNA. Partial denaturation patterns of spinach cpDNA circles and BglI fragments were determined and from this the position of the fragments mapped. This confirmed the reliability of these methods for the arrangement of restriction enzyme fragments along circular molecules. The structures of the two cpDNAs were compared.     

Ultrastructure and properties of the sexual agglutinins of the biflagellate green alga Chlamydomonas moewusii

Summary The mt^– agglutinins of the interfertile species Chlamydomonas moewusii and Chlamydomonas eugametos are very similar fibrous molecules. The mt^– agglutinin of C. moewusii has the same Stokes radius (39 nm) and sedimentation coefficient (9.3 S) as its counterpart in C. eugametos; its length (336 nm) and its ultrastructure, including the position of four kinks are also the same as in C. eugametos. The sugar compositions of both agglutinins are very similar, and they react equally well with the monoclonal antibody Mab 66.3 raised against the mt^– agglutinin of C. eugametos. Finally, they are equally thermoresistant, with half-lives at 100 °C of 50 min (C. moewusii) and 57 min (C. eugametos). The mt⁺ agglutinins of both species are different. Both are fibrous molecules with a terminal head, but the fibrous part of the molecule in C. moewusii is shorter (210 nm compared to 276 nm). The mt⁺ agglutinin of C. moewusii is also significantly more sensitive to heating with a half-life of 6 min at 40 °C compared to the 20 min shown by the mt⁺ agglutinin of C. eugametos. Their sugar compositions are, however, very similar, and they react equally well with Mab 66.3. The mt⁺ agglutinin of C. moewusii is sensitive to denaturing reagents and proteolytic attack, whereas the mt^– agglutinin is highly resistant. It is proposed that the globular head of the mt⁺ agglutinin acts as its recognition domain and interacts with a carbohydrate ligand on the mt^– agglutinin.