Extreme plastomes in holoparasitic Balanophoraceae are not the norm
BMC Genomics volume 24, Article number: 330 (2023)
Balanophoraceae plastomes are known for their highly condensed and re-arranged nature alongside the most extreme nucleotide compositional bias known to date, culminating in two independent reconfigurations of their genetic code. Currently, a large portion of the Balanophoraceae diversity remains unexplored, hindering, among others, evolutionary pattern recognition. Here, we explored newly sequenced plastomes of Sarcophyte sanguinea and Thonningia sanguinea. The reconstructed plastomes were analyzed using various methods of comparative genomics based on a representative taxon sampling.
Sarcophyte, recovered sister to the other sampled Balanophoraceae s. str., has plastomes up to 50% larger than those currently published. Its gene set contains five genes lost in any other species, including matK. Five cis-spliced introns are maintained. In contrast, the Thonningia plastome is similarly reduced to published Balanophoraceae and retains only a single cis-spliced intron. Its protein-coding genes show a more biased codon usage compared to Sarcophyte, with an accumulation of in-frame TAG stop codons. Structural plastome comparison revealed multiple, previously unknown, structural rearrangements within Balanophoraceae.
For the “minimal plastomes” of Thonningia, we propose a genetic code change identical to sister genus Balanophora. Sarcophyte however differs drastically from our current understanding on Balanophoraceae plastomes. With a less-extreme nucleotide composition, there is no evidence for an altered genetic code. Using comparative genomics, we identified a hotspot for plastome reconfiguration in Balanophoraceae. Based on previously published and newly identified structural reconfigurations, we propose an updated model of evolutionary plastome trajectories for Balanophoraceae, illustrating a much greater plastome diversity than previously known.
Holoparasitic Balanophoraceae maintain some of the most extreme plastomes known to date. Considerable structural reconfigurations alongside a highly condensed gene set have identified the family as key lineage for the study of evolutionary processes of so-called “minimal plastomes” [1,2,3,4]. Their nucleotide composition represents the most extreme reported for this organelle so far and is accompanied by two independent modifications of the underlying genetic code [1, 3].
Balanophoraceae (Santalales), consisting of 42 species from 14 genera, is broadly distributed across (sub-) tropical regions worldwide [5,6,7]. Relationships within the family are partially uncertain as molecular data for a large portion of the diversity are currently lacking [3, 8]. Balanophoraceae are estimated to be among the oldest parasitic angiosperm lineages (~ 110 – 115 mya) alongside Cynomoriaceae and Hydnoraceae . However, the limited taxon sampling and the effect of elevated nucleotide substitution rates heavily influence such estimations. Complete plastomes of only four Balanophoraceae genera have been characterized to date, all having lost the quadripartite genome structure and displaying similarly condensed size and gene content as well as nucleotide compositional bias. The Balanophoraceae minimal plastomes range from ~ 17.3 kb – 20.9 kb in genera Lophophytum, Ombrophytum, and Rhopalocnemis [3, 4, 10] and from ~ 14.6 kb – 16 kb in genus Balanophora [1, 2]. In the latter, all cis-spliced introns have been lost and intergenic spacers (IGS) are drastically reduced with various partially overlapping genes [1, 2]. Balanophoraceae plastomes contain a shared set of 13 – 19 genes, comprised of a variable set of rRNAs, ribosomal proteins (rpl and rps), and protein coding genes with other functions (accD, clpP, ycf1, and ycf2), as well as a single tRNA (trnE-UUC) in Balanophora . Adenine and thymine make up to 88.4% of the plastomes of B. reflexa  and Lophophytum leandri , compared to an average of 61 – 66% A + T in autotrophic angiosperms. Additional published Balanophoraceae genera confirm this general trend towards low nucleotide complexity plastomes [2,3,4, 10]. This extreme bias likely resulted in two independent codon reassignments in Balanophoraceae. In Ombrophytum and Lophophytum plastomes, tryptophan is likely encoded by TGA in addition to only TGG , whereas TAG is indicated to additionally encode tryptophan in plastomes of Balanophora . As a result of these reassignments, translation termination is almost exclusively realized using TAA, even in Rhopalocnemis, for which no altered genetic code has been observed.
All previous studies paint a picture of Balanophoraceae plastomes that are exceptional in most categories. However, these conclusions are based on only ~ 30% of the known genus diversity. Here, we report the newly sequenced and assembled plastomes of two additional Balanophoraceae genera, Sarcophyte sanguinea Sparrm. and Thonningia sanguinea Vahl (Fig. 1). We also assessed intraspecific variation by using two and three biological replicates from different origins, respectively. Using comparative approaches, we explore gene and intron losses, structural variation as well as nucleotide compositional bias and changes in genetic code. Based on a fully resolved phylogenetic hypothesis, we propose an expanded model for possible evolutionary trajectories in Balanophoraceae.
Plant materials, DNA extraction and sequencing
Genomic DNAs of Thonningia sanguinea (3 accessions) and Sarcophyte sanguinea (2 accessions) were isolated from silica gel dried plant material and sequenced via next generation sequencing. Two specimens of Thonningia sanguinea were collected in Angola with permits granted by the Instituto Nacional da Biodiversidade e Áreas de Conservação (INBAC, Ministério do Ambiente, N.145/INBAC.MINAMB/2013) and the Governo Provincial de Uíge (52/GD/IP-UNIKIVIY2022) and deposited in the herbarium of the Royal Botanic Garden Edinburgh (Rees 90 and Rees 123, identified by Mathew Rees). The third biological replicate corresponds to a herbarium specimen (Herbarium Dresdense, DR077682, identified by Thea Lautenschläger). Both studied accessions of Sarcophyte sanguinea correspond to herbarium specimens, one in the collection of the Herbarium Dresdense (DR077681, identified by Jay F. Bolin) and the second in the collection of the East African Herbarium (Matheka K. 202101, identified by Kennedy Wambua Matheka). DNA extraction followed the protocol of Doyle and Doyle , modified to include RNAase A (Thermo Scientific, Waltham, MA, USA) treatment (10 mg/ml). DNA concentration and quality were measured using a Qubit 3 Fluorometer (Thermofisher Scientific, Waltham, MA, USA). Sequencing was carried out on an Illumina NovaSeq 6000 system, as 150 bp paired-end reads for 300 cycles, aiming for about 80 million reads per sample.
Data mining from public repositories
For comparative analysis and phylogenetic tree reconstruction, plastomes of Balanophoraceae as well as selected outgroup lineages were mined from the NCBI nucleotide database. The complete list of plastomes used in this study can be found in the Additional file 1. We annotated the plastomes of Balanophora harlandii (MN414177; ) and Balanophora fungosa var. globosa (MN414176; ) following the methods described below, as the published plastomes did not contain any annotations.
Raw data assembly and plastid genome reconstruction
Raw read data were trimmed for sequencing adapters and quality using BBDuk v. 1.0 . Trimmed read data were de novo assembled using two different methods and cross-checked. De novo assembly after filtering for plastid-like reads was done using the GetOrganelle v. 184.108.40.206  pipeline, with standard settings. De novo assembly on all sequence raw data was done using CLC Genomics Workbench (v. 11.0, Qiagen, MD, USA), allowing for automatic calculation of optimal word and bubble sizes for each sample. For each assembly, read-mappings were created using CLC Genomics Workbench (v. 11.0, Qiagen, MD, USA) and assessed using Tablet v. 1.21.02.08 .
Gene annotation of newly sequenced plastomes was done in Geneious (v. 11.1.5, Biomatters, Ltd., New Zealand) using the published plastomes of Balanophoraceae [1,2,3, 10] and autotrophic Erythropalum scandens (NC_036759)  as references. Annotations were manually inspected and adjusted where necessary. Additionally, open reading frames (ORFs) of protein coding genes were investigated using HMMER  to detect homologs using various protein sequence databases. Sequence and secondary structure of tRNAs were predicted using tRNAscan-SE v. 2.0 . Plastomes were visualized using OGDraw .
Single gene alignments for 19 protein-coding genes and 4 ribosomal RNAs were created using MAFFT v. 7.450 [21, 22] and manually adjusted in PhyDE (v 0.9971) and AliView . We excluded rrn5 from the analysis, considering its often highly divergent sequence and uncertain functionality. All other genes were concatenated using SequenceMatrix v. 1.8 . PartitionFinder2  was used to identify the best-fitting nucleotide substitution model (GTR + I + G). Phylogenetic inferences were estimated with RAxML v. 8.2.12 , implemented on Cipres Science Gate , using a gene partitioning approach. Boostrap support (BS) values are based on 1,000 replicates. Tree files were visualized using TreeGraph 2 .
The plastomes of Sarcophyte and Thonningia
Plastome reconstruction resulted in complete, circular molecules for both accessions of Sarcophyte sanguinea and the three accessions of Thonningia sanguinea, all lacking a quadripartite genome structure. The plastomes of Sarcophyte (28,372 – 28,384 bp) are ~ 10 kb larger in size compared to those of Thonningia (18,560 – 19,013 bp) and exhibit a slightly larger nucleotide compositional bias (19.1% GC versus 20.4 – 21.2% GC, respectively, Additional file 1). Both Sarcophyte plastomes contain an identical 27-gene set, comprised of 19 protein coding genes (accD, clpP, matK, rpl, rps, and ycf), four rRNA genes, as well as four tRNAs (trnE, trnI, trnQ, and trnW) (Fig. 2 and Additional file 2). Four genes are found containing cis-spliced introns, namely clpP (2), rpl2 (1), rpl16 (1), and rps12 (1), with the latter additionally containing a trans-spliced intron (Additional file 1). Secondary structure prediction shows typical cloverleaf folds for all identified tRNAs, along with the presentation of the correct anticodon, as predicted by sequence similarity, except for trnW. Here, an UUA anticodon is predicted instead of CCA. Gene order and orientation is identical among the two plastomes of Sarcophyte, accompanied by a low degree of nucleotide diversity. Noteworthy is an indel in a mononucleotide repeat at the rpl2 3’-end, which is supported by the readmappings. The introduced frameshift leads to a gene length increase (10 bp) in accession S01, resulting in a partial overlap (4 bp) with the adjacent rps19 gene.
The plastomes of the three Thonningia sanguinea accessions share an identical 22-gene set, consisting of 17 protein-coding genes (accD, clpP, rpl, rps, and ycf), four rRNA genes, and a single tRNA (trnE). Two introns were identified: a cis-spliced intron in clpP and a trans-spliced intron in rps12 (Fig. 2 and Additional file 2). Gene annotation using the plant plastid code revealed frequent inclusion of premature stop codons (exclusively TAG) in twelve out of 17 protein-coding genes. No TAG stop codons were identified at any 3’-terminus in Thonningia. Nucleotide diversity between accessions T02 and T03 was lower compared with T01; however, gene order and orientation is synonymous among all three. A noteworthy structural distinction unique to accession T02 is the lack of gene overlap for the rpl36 5’-end with the adjacent rpl2 3’-end, which is present in both T01 and T03 (4 bp). The gene length of the T01 rpl16 gene is much shorter (303 bp) compared to the other two accessions (both 405 bp) due to a point mutation altering the start codon. The next in-frame canonical start codon was chosen for the rpl16 annotation of T01.
Gene and intron losses and structural variations in the Balanophoraceae plastomes
Annotation of published Balanophoraceae plastomes using gene features of Sarcophyte allowed for the identification of rrn5 genes (112 – 125 bp) and rrn5 gene fragments (58 – 105 bp) in all accessions. Additionally, we found and annotated an rpl36 gene copy in B. harlandii (Fig. 3, Additional file 2). This annotation is further supported by HMMER prediction and contains a complete ORF. All Balanophoraceae plastomes lack a quadripartite plastome structure and no larger repeat regions (> 30 bp) have been identified. Genome sizes range from ~ 28.4 kb in Sarcophyte to ~ 14.6 kb in Balanophora yakushimensis (Additional file 1). The nucleotide compositional bias heavily favors A and T, culminating in 11.6% GC in both B. reflexa and Lophophytum (Additional file 1). The highest GC content was identified in the plastomes of Thonningia (20.4 – 21.2% GC).
Balanophoraceae plastomes display varying gene and intron contents as well as changes in gene order and orientation. In the plastomes of Sarcophyte, we identified five genes that are missing in all other Balanophoraceae, namely matK, rps8 as well as tRNAs I, Q, and W (Additional file 2). In both B. yakushimensis and B. fungosa, the ycf2 gene has been lost. Balanophora laxiflora, B. reflexa, and B. yakushimensis as well as Lophophytum share the loss of rpl36 (Fig. 3). Furthermore, Lophophytum, Ombrophytum, and Rhopalocnemis are missing rpl14, rps2, rps11, and rrn4.5 compared to the other Balanophoraceae studied here. Lastly, rps4 is missing from the Rhopalocnemis plastomes (Fig. 3 and Additional file 2). In Balanophoraceae, at least five independent intron losses have been identified. Balanophora is yet the only known genus of the family having lost all cis-spliced introns. Thonningia has lost the cis-spliced rps12 and rpl2 introns. The clpP gene, containing two cis-splicing introns in Sarcophyte, Lophophytum, Ombrophytum, and Rhopalocnemis, lacks intron 2 in Thonningia (Additional file 2). We identified a similar pattern of intron loss for rpl16. The gene is missing in Balanophora, Lophopytum, and Ombrophytum, present but without intron in Thonningia and contains a single cis-spliced intron in Rhopalocnemis and Sarcophyte (Additional file 2). The only intron shared among all Balanophoraceae plastomes is the trans-spliced intron of rps12.
Multiple events of structural discordance can be identified among the Balanophoraceae plastomes. Balanophora harlandii is characterized by the only inversion of the rps4 gene yet (Fig. 3). Balanophora and Thonningia share inversions of both rps14 and rps18, whereas ycf2 is inverted in Lophophytum, Ombrophytum, and Rhopalocnemis. Compared to the plastomes of Sarcophyte, the rpl2 – rpl36 (~ 2060—2800 bp) region in Lophophytum, Ombrophytum, and Rhopalocnemis is translocated (Fig. 3). The almost identical region, rpl2 – rps11 (~ 2000—2200 bp), is translocated and inverted in Balanophora and Thonningia, although excluding rpl14 and rpl16.
Considering the plant plastid code, internal TAG stop codons, as identified in the majority of Thonningia protein-coding genes, have additionally been found in genus Balanophora. Internal TGA stop codons, on the other hand, are recognized in Lophophytum and Ombrophytum.
Phylogenetic relationships within Balanophoraceae
Phylogenetic analysis based on the concatenated 22-plastid gene set reveals the monophyly of Balanophoraceae, with Sarcophyte as the first diverged genus among the sampled genera of this family (BS 100, Fig. 4). Lophophytum, Ombrophytum and Rhopalocnemis are resolved as a monophyletic group, and place sister to a clade of Thonningia and Balanophora (BS 100 for both groups, Fig. 4). Within genus Balanophora, B. harlandii diverged first (BS 100). The branches leading to all five genera of Balanophoraceae are considerably longer than those leading to the outgroups, with the longest branch leading to Rhopalocnemis.
Plastome structural changes among Balanophoraceae and close relatives
To determine the extent of plastome reconfiguration in Balanophoraceae, we compared representative plastomes of an autotrophic Santalales (Eryhtropalum)  and hemi-parasitic lineages (Loranthaceae, Opiliaceae, and Schoepfiaceae, Additional file 1). Due to identical gene order and orientation in the selected outgroup taxa, we chose closely related Nuytsia (NC_058869)  for pairwise comparisons with Balanophoraceae. For better visualization, we created three versions of the Nuytsia plastome, containing only the gene content present in a) Sarcophyte, b) Balanophora and Thonningia, and c) Lophophytum, Ombrophytum, and Rhopalocnemis (Additional file 3). The pairwise comparisons revealed two translocations combined with inversions between Sarcophyte and Nuytsia (matK-trnQ and rps11-rpl2). The gene order of Thonningia and Balanophora can be explained by two independent inversions (rps14, rps18) along with a single translocation + inversion (rpl14-rpl16). Lastly, the plastome structure of Lophophytum, Ombrophytum, and Rhopalocnemis differs in an inversion (rpl36-rpl2) and a translocation combined with an inversion (ycf2) from the Nuytsia plastome (Additional file 3).
Relative synonymous codon usage analysis in Balanophoraceae
To illuminate potential patterns of bias in synonymous codon usage, we estimated the RSCU for the protein-coding genes of the Balanophoraceae plastomes. In all accessions, A/T-rich codons are greatly preferred to codons composed of C/G. Several codons are not used at all, mostly in Balanophora, followed by Ombrophytum, Lophophytum, and Rhopalocnemis (Additional file 4). Noteworthy is that the sole codon for tryptophan (UGG) is not being used in one accession of Rhopalocnemis (MZ269413)  and several Balanophora species. Thonningia and Sarcophyte show a less extreme, but still highly biased RSCU. Little differences in RSCU are identified among the different accessions of Thonningia and Sarcophyte, as well as within Balanophora.
Out of the three canonical stop codons, TAA is by far the most frequently used in the protein-coding genes of Balanophoraceae, with the exception of Thonningia and Ombrophytum, which show TGA and TAG instead, respectively (Additional file 4).
The Balanophoraceae plastomes published to date are considered some of the most exceptional with genome sizes of only ~ 15 – 21 kb, the most extreme nucleotide compositional bias (up to 88.4% AT) and the only two described modifications of the plant plastid translational code among land plants [1,2,3]. The newly sequenced plastomes of Thonningia are no exception to this. Assembled genomes of all three Thonningia accessions are to be considered “minimal plastomes”, with lengths ranging from ~ 18.6 – 19 kb and an A/T content of ~ 80% (Additional file 1). Gene annotation and codon analysis indicate an alteration of its genetic code, identical to that of its sister genus Balanophora (Additional file 4). In the latter, transcriptome analysis confirmed the change of a terminal codon (TAG) to a tryptophan codon, resulting in translation extension rather than termination .
The newly assembled plastomes of Sarcophyte, on the other hand, are distinct from other Balanophoraceae in various aspects. With genome sizes of ~ 28 kb, they are roughly 33 – 50% larger than their family members (Additional file 1) and classify in the upper size range of what can be considered a minimal plastome, comparable to various holoparasitic Hydnoraceae (~ 28 kb)  and certain species of holo-mycoheterotrophic Epipogium (~ 30 kb) . Tree reconstruction recovered Sarcophyte sister to the remaining Balanophoraceae, comprised of the clades Balanophora + Thonningia and Ombrophytum + Lophophytum + Rhopalocnemis (Fig. 4). This result is congruent with the reconstructions estimated by Chen et al.  and Ceriotti et al. , although based on a different taxon sampling. The recovery of Sarcophyte as sister to all other sampled Balanophoraceae in our study (BS 100, Fig. 4) is however incongruent with its placement recovered by Su et al. . In their study, Sarcophyte places sister to Lophophytum + Ombrophytum and Corynaea + Helosis (the latter two not included in our study) with low bootstrap support.
Dynamic gene and intron losses in Balanophoraceae plastomes
Mapping individual events of gene and intron loss on the Balanophoraceae phylogeny reveals a highly dynamic picture of microstructural evolution (Fig. 5), in line with many other heterotrophic lineages [33,34,35]. Sister to the remaining Balanophoraceae, Sarcophyte maintains the largest plastid gene set (27, Additional file 2) and shows no intron loss in its plastid genes compared to the autotrophic outgroup (Erythropalum, NC_036759). Among the five genes uniquely retained in Sarcophyte are three tRNAs (I, Q, and W) and a protein-coding gene encoding for the maturase (matK, Additional file 2). The majority of additional gene losses are attributed to the branch leading to Ombrophytum, Lophophytum, and Rhopalocnemis, with a total of five (rpl14, rps2, rps11, rrn4.5, and trnE) (Fig. 5). No further genes have been lost on the branch leading to Balanophora and Thonningia. Here, plastome condensation mostly concerns the loss of group II introns. It is uncertain, whether the rpl16 intron was lost prior to the complete gene loss in Balanophora (Fig. 5). In Balanophoraceae, we identified three genes, which have each been independently lost twice. Namely, rpl16 in Lophophytum + Ombrophytum and Balanophora, rpl36 in B. reflexa and B. laxiflora, and ycf2 in B. fungosa and B. yakushimensis (Fig. 5). All retained cis-spliced introns in the Balanophoraceae plastomes (Additional file 2) belong to group II [36, 37] and require a maturase for mRNA splicing prior to translation. MatK is the only plastid-encoded maturase and therefore considered essential for protein synthesis of group II intron-containing genes , with the exception of the clpP intron 2 . Considering this, only Sarcophyte (presence of matK and group II introns) and Balanophora (loss of all cis-spliced introns) could facilitate translation for all retained protein-coding genes. However, it seems unlikely that multiple intron-containing genes could be retained (and evolving under purifying selection;  in matK-lacking genera without an alternative means of splicing. A plausible solution is the import of a nuclear-encoded matK-like protein from the cytosol. In Physcomitrella such a protein has been shown to splice clpP pre-mRNA . Complementary transcriptome data are highly desirable as additional evidence to verify successful splicing of intron-containing genes in Balanophoraceae.
All protein-coding genes in Thonningia and Sarcophyte are assumed functional based on their intact ORFs. Due to extreme sequence divergence, no further comments on the functionality of the newly identified rrn5 genes in the published plastomes can be made. Secondary structure prediction estimates typical cloverleaf folds for all annotated tRNAs and displays the correct anticodons as predicted by sequence similarity. TrnW of Sarcophyte is the only exception to this; it displays TTA (a stop codon) instead of the expected CCA anticodon. Tryptophan, with the exception of Balanophora, Lophophytum, Ombrophytum, and Thonningia, is encoded by only a single codon.
Comparative analysis on genetic code reassignment, RSCU and tRNA loss within Balanophoraceae
Codon analysis of Sarcophyte, unearths various differences to most of the extremes observed in the family [1,2,3]. Similar to Rhopalocnemis, canonical stop codons are solely found at terminal position of the protein-coding genes. There is no evidence for Sarcophyte to carry any genetic code reassignments, which independently occurred twice in other groups of Balanophoraceae (Fig. 5) [1, 3]. Nevertheless, both Sarcophyte and Rhopalocnemis almost exclusively show use of the TAA stop codon (Additional file 4), similar to those Balanophoraceae plastomes with genetic code reassignments [1, 3]. RSCU analysis signals a correlation between overall plastome A/T content and codon usage bias. With the highest G/C content in Balanophoraceae, Sarcophyte and Thonningia (19.1 and 21.2% G/C, respectively) generally use a more diverse set of codons, resulting in RSCU values not as extreme as their relatives (Additional file 4). This lower bias in Sarcophyte and Thonningia can be best explained by their nucleotide composition due to neutral mutation/drift rather than selection efficiency in codon-usage [1, 41, 42]. In short, our data provide compelling evidence for a reassignment of genetic code in Thonningia, identical to the one identified in Balanophora (Fig. 5).
Codons corresponding to tRNAs retained in the plastomes of Sarcophyte, Thonningia, and Balanophora show no remarkable differences in their use compared to plastomes from which they are missing (Additional file 4). Considering that none of the retained tRNA gene sets are sufficient for protein synthesis, the import of cytosolic, plastid-targeted tRNAs likely exists in Balanophoraceae, in line with many other heterotrophic lineages. It is hypothesized for such an import mechanism to be able to also compensate for the presumed recent pseudogenization of trnE in Balanophora , as well as for the likely pseudogenized trnW in Sarcophyte. Retention of trnE, despite its pseudogene-like character, is presumed due to its additional function in heme biosynthesis [43,44,45]. Whether the same can be called for trnW in Sarcophyte is uncertain. However, in the plastome of Prosopanche americana, a trnW pseudogene-like copy with similarly altered anticodon still persists . It could be hypothesized that the tRNAs retained in the Balanophoraceae plastomes may play less or no roles in protein synthesis anymore.
Likely evolutionary trajectories of rearrangements in Balanophoraceae plastomes
Plastome structure, generally highly conserved , is often altered in holo-heterotrophic plants [31, 32, 35, 48,49,50]. Understanding the evolutionary direction of structural rearrangements is essential to study underlying mechanisms. In Balanophoraceae plastomes, we identified multiple independent events of structural reconfigurations (Fig. 3). In combination with an improved taxon sampling compared to a previous study  and including closely-related outgroups, we propose an updated model for structural evolution. Considering the identical plastome structures among autotrophic Erythropalum and hemi-parasitic Champeria, Nuytsia, and Schoepfia (Santalales outgroups), likely representing the ancestral state, seven independent events of rearrangement have happened in Balanophoraceae (Additional file 3). Based on these rearrangements, Balanophoraceae can be subdivided into three monophyletic groups, which share identical gene order and orientation among them: A) Sarcophyte, B) Thonningia and Balanophora (with the exception of an inversion unique to B. harlandii (rps4, Figs. 2 and 4)), and C) Ombrophytum, Lophophytum, and Rhopalocnemis (Fig. 5, Additional file 3). The plastome structure of Sarcophyte can be explained by two events of sequence translocation and inversion. Three steps are required to explain the plastomes of Thonningia and Balanophora, one inversion and one translocation + inversion. Lastly, two steps are required for the conversion of the outgroup plastome structure to that present in Ombrophytum, Lophophytum, and Rhopalocnemis (Fig. 5, Additional file 3). None of the abovementioned rearrangements are shared among all Balanophoraceae, therefore having to be considered independent events. Noteworthy however is the plastome region containing the bulk of the retained ribosomal protein-coding genes (rpl, rps). Parts of this region are affected by three different rearrangement events (Fig. 3 and Additional file 3), identifying it as hotspot for plastome reconfiguration in Balanophoraceae.
Our results on plastome diversity, genetic code change and the highly dynamic nature of gene and intron losses in combination with their age identify holoparasitic Balanophoraceae as pivotal to study the general evolutionary processes acting on minimal plastomes (such as condensation), in line with mycoheterotrophic Burmanniaceae [51, 52], Orchidaceae , and Thismiaceae [54, 55]. Derived models will allow for estimations on putative past and future evolutionary trajectories and will further our understanding of universal aspects of plastome evolution . Confident model establishment however requires more data than currently available, which is especially true for Balanophoraceae. Our study highlights, that the available data for a snippet of the known diversity of a lineage are not necessarily representative for it as a whole. Screening the NCBI database (https://www.ncbi.nlm.nih.gov/) highlights data deficiency for many holoparasitic lineages, as for only half of the approximately 40 genera of holoparasitic plants, a complete plastome reconstruction exists to date.
This study compliments and expands our knowledge on plastome evolution in Balanophoraceae and holo-heterotrophic plants as a whole and highlights, for the first time, less extreme and less condensed plastomes within the family. These results indicate that Balanophoraceae plastomes show a much greater diversity than previously known and it is essential to increase sampling efforts for currently missing genera and species. Based on our results, we propose Balanophoraceae as an ideal holoparasitic lineage to study the processes of minimal plastome evolution.
Availability of data and materials
The newly assembled plastomes of Sarcophyte and Thonningia are available in the NCBI nucleotide database under accession numbers OQ810027—OQ810031.
Su HJ, Barkman TJ, Hao W, Jones SS, Naumann J, Skippington E, Wafula EK, Hu JM, Palmer JD, dePamphilis CW. Novel genetic code and record-setting AT-richness in the highly reduced plastid genome of the holoparasitic plant Balanophora. Proc Natl Acad Sci. 2019;116(3):934–43. https://doi.org/10.1073/pnas.1816822116.
Chen X, Fang D, Wu C, Liu B, Liu Y, Sahu SK, Song B, Yang S, Yang T, Wei J, Wang X, Zhang W, Xu Q, Wang H, Yuan L, Liao X, Chen L, Chen Z, Yuan F, Chang Y, Lu L, Yang H, Wang J, Xu X, Liu X, Wicke S, Liu H. Comparative plastome analysis of root- and stem-feeding parasites of Santalales untangle the footprints of feeding mode and lifestyle transitions. Genome Biol Evol. 2020;12(1):3663–76. https://doi.org/10.1093/gbe/evz271.
Ceriotti LF, Roulet ME, Sanchez-Puerta MV. Plastomes in the holoparasitic family Balanophoraceae: extremely high AT content, severe gene content reduction, and two independent genetic code changes. Mol Phylogenet Evol. 2021;162:107208.
Yu R, Sun C, Zhong Y, Liu Y, Sanchez-Puerta MV, Mower JP, Zhou R. The minicircular and extremely heteroplasmic mitogenome of the holoparasitic plant Rhopalocnemis phalloides. Curr Biol. 2022;32(2):470-479.e5. https://doi.org/10.1016/j.cub.2021.11.053.
Hansen B. Balanophoraceae. Flora Neotropica; 1980;23:1-80.
Su HJ, Murata J, Hu JM. Morphology and phylogenetics of two holoparasitic plants, Balanophora japonica and Balanophora yakushimensis (Balanophoraceae), and their hosts in Taiwan and Japan. J Plant Res. 2012;125(3):317–26. https://doi.org/10.1007/s10265-011-0447-5.
Yu RX, Zhou SY, Li YQ, Ma KN, Li YY, Huang LS, Zhou R, Liu Y. Two new species of Balanophora (Balanophoraceae) From South China. Syst Bot. 2021;46(4):1053–66. https://doi.org/10.1600/036364421X16370110179500.
Su HJ, Hu JM, Anderson FE, Der JP, Nickrent DL. Phylogenetic relationships of Santalales with insights into the origins of holoparasitic Balanophoraceae. Taxon. 2015;64(3):491–506.
Naumann J, Salomo K, Der JP, Wafula EK, Bolin JF, Maass E, Frenzke L, Samain MS, Neinhuis C, dePamphilis CW, Wanke S. Single-copy nuclear genes place haustorial Hydnoraceae within Piperales and reveal a Cretaceous origin of multiple parasitic Angiosperm lineages. PLOS ONE. 2013;8(11):e79204.
Schelkunov MI, Nuraliev MS, Logacheva MD. Rhopalocnemis phalloides has one of the most reduced and mutated plastid genomes known. PeerJ. 2019;7:e7500.
Doyle JJ, Doyle JL. A rapid DNA isolation procedure for small quantities of fresh leaf tissue. Phytochem Bull. 1987;19:11–5.
Bushnell B. BBTools software package. 2014. Available online: http://sourceforge.net/projects/bbmap. Accessed 13 Mar 2023.
Jin JJ, Yu WB, Yang JB, Song Y, dePamphilis CW, Yi TS, Li DZ. GetOrganelle: a fast and versatile toolkit for accurate de novo assembly of organelle genomes. Genome Biol. 2020;21(1):241. https://doi.org/10.1186/s13059-020-02154-5.
Milne I, Stephen G, Bayer M, Cock PJA, Pritchard L, Cardle L, Shaw PD, Marshall D. Using tablet for visual exploration of second-generation sequencing data. Brief Bioinform. 2013;14(2):193–202. https://doi.org/10.1093/bib/bbs012.
Zhu ZX, Wang JH, Cai YC, Zhao KK, Moore MJ, Wang HF. Complete plastome sequence of Erythropalum scandens (Erythropalaceae), an edible and medicinally important liana in China. Mitochondrial DNA Part B. 2018;3(1):139–40. https://doi.org/10.1080/23802359.2017.1413435.
Potter SC, Luciani A, Eddy SR, Park Y, Lopez R, Finn RD. HMMER web server: 2018 update. Nucleic Acids Res. 2018;46(W1):W200–4. https://doi.org/10.1093/nar/gky448.
Chan PP, Lin BY, Mak AJ, Lowe TM. tRNAscan-SE 2.0: improved detection and functional classification of transfer RNA genes. Nucleic Acids Research. 2021;49(16):9077–96. https://doi.org/10.1093/nar/gkab688.
Greiner S, Lehwark P, Bock R. OrganellarGenomeDRAW (OGDRAW) version 1.3.1:expanded toolkit for the graphical visualization of organellar genomes. Nucleic Acids Research. 2019;47(1):59–64. https://doi.org/10.1093/nar/gkz238.
Sharp PM, Tuohy TM, Mosurski KR. Codon usage in yeast: cluster analysis clearly differentiates highly and lowly expressed genes. Nucleic Acids Res. 1986;14(13):5125–43. https://doi.org/10.1093/nar/14.13.5125.
Tamura K, Stecher G, Kumar S. MEGA11: Molecular Evolutionary Genetics Analysis Version 11. Mol Biol Evol. 2021;38(7):3022–7. https://doi.org/10.1093/molbev/msab120.
Katoh K, Misawa K, Kuma K, Miyata T. MAFFT: a novel method for rapid multiple sequence alignment based on fast Fourier transform. Nucleic Acids Res. 2002;30(14):3059–66. https://doi.org/10.1093/nar/gkf436.
Katoh K, Standley DM. MAFFT multiple sequence alignment software version 7: improvements in performance and usability. Mol Biol Evol. 2013;30(4):772–80. https://doi.org/10.1093/molbev/mst010.
Larsson A. AliView: a fast and lightweight alignment viewer and editor for large datasets. Bioinformatics. 2014;30(22):3276–8. https://doi.org/10.1093/bioinformatics/btu531.
Vaidya G, Lohman DJ, Meier R. SequenceMatrix: concatenation software for the fast assembly of multi-gene datasets with character set and codon information. Cladistics. 2011;27(2):171–80. https://doi.org/10.1111/j.1096-0031.2010.00329.x.
Lanfear R, Frandsen PB, Wright AM, Senfeld T, Calcott B. PartitionFinder 2: New methods for selecting partitioned models of evolution for molecular and morphological phylogenetic analyses. Mol Biol Evol. 2017;34(3):772-773. https://doi.org/10.1093/molbev/msw260.
Stamatakis A. RAxML version 8: a tool for phylogenetic analysis and post-analysis of large phylogenies. Bioinformatics. 2014;30(9):1312–3. https://doi.org/10.1093/bioinformatics/btu033.
Miller MA, Pfeiffer W, Schwartz T. Creating the CIPRES Science Gateway for inference of large phylogenetic trees. IEEE. 2010.
Stöver BC, Müller KF. TreeGraph 2: combining and visualizing evidence from different phylogenetic analyses. BMC Bioinformatics. 2010;11(1):7. https://doi.org/10.1186/1471-2105-11-7.
Nickrent DL. Parasitic angiosperms: How often and how many? Taxon. 2020;69(1):5–27. https://doi.org/10.1002/tax.12195.
Nickrent DL, Su HJ, Lin RZ, Devkota MP, Hu JM, Glatzel G. Examining the needle in the haystack: Evolutionary relationships in the mistletoe genus Loranthus (Loranthaceae). Syst Bot. 2021;46(2):403–15. https://doi.org/10.1600/036364421X16231785234748.
Jost M, Naumann J, Bolin JF, Martel C, Rocamundi N, Cocucci AA, Lupton D, Neinhuis C, Wanke S. Structural plastome evolution in holoparasitic Hydnoraceae with special focus on inverted and direct repeats. Genome Biol Evol. 2022;14(6):evac077. https://doi.org/10.1093/gbe/evac077.
Schelkunov MI, Shtratnikova VY, Nuraliev MS, Selosse MA, Penin AA, Logacheva MD. Exploring the limits for reduction of plastid genomes: a case study of the mycoheterotrophic orchids Epipogium aphyllum and Epipogium roseum. Genome Biol Evol. 2015;7(4):1179–91. https://doi.org/10.1093/gbe/evv019.
Daniell H, Lin CS, Yu M, Chang WJ. Chloroplast genomes: diversity, evolution, and applications in genetic engineering. Genome Biol. 2016;17(1):134. https://doi.org/10.1186/s13059-016-1004-2.
Mower JP, Vickrey TL. Structural diversity among plastid genomes of land plants. Adv Bot Res. 2018;85:263–92. https://doi.org/10.1016/bs.abr.2017.11.013.
Wicke S, Naumann J. Molecular evolution of plastid genomes in parasitic flowering plants. Adv Bot Res. 2018;85:315–47. https://doi.org/10.1016/bs.abr.2017.11.014.
Kelchner SA. Group II introns as phylogenetic tools: structure, function, and evolutionary constraints. Am J Bot. 2002;89(10):1651–69. https://doi.org/10.3732/ajb.89.10.1651.
Daniell H, Wurdack KJ, Kanagaraj A, Lee SB, Saski C, Jansen RK. The complete nucleotide sequence of the cassava (Manihot esculenta) chloroplast genome and the evolution of atpF in Malpighiales: RNA editing and multiple losses of a group II intron. Theor Appl Genet. 2008;116(5):723–37. https://doi.org/10.1007/s00122-007-0706-y.
Schmitz-Linneweber C, Barkan A. RNA splicing and RNA editing in chloroplasts. Cell and Molecular Biology of Plastids. Springer. 2007:213–48. https://doi.org/10.1007/4735_2007_0233.
Zoschke R, Nakamura M, Liere K, Sugiura M, Börner T, Schmitz-Linneweber C. An organellar maturase associates with multiple group II introns. Proc Natl Acad Sci U S A. 2010;107(7):3245–50. https://doi.org/10.1073/pnas.0909400107.
Hattori M, Miyake H, Sugita M. A pentatricopeptide repeat protein is required for RNA processing of clpP pre-mRNA in moss chloroplasts. J Biol Chem. 2007;282(14):10773–82. https://doi.org/10.1074/jbc.M608034200.
Plotkin JB, Kudla G. Synonymous but not the same: the causes and consequences of codon bias. Nat Rev Genet. 2011;12(1):32–42. https://doi.org/10.1038/nrg2899.
Suzuki H, Morton BR. Codon adaptation of plastid genes. PLoS ONE. 2016;11(5):e0154306.
Howe CJ, Smith A. Plants without chlorophyll. Nature. 1991;349(6305):109–109. https://doi.org/10.1038/349109c0.
Barbrook AC, Howe CJ, Purton S. Why are plastid genomes retained in non-photosynthetic organisms? Trends Plant Sci. 2006;11(2):101–8. https://doi.org/10.1016/j.tplants.2005.12.004.
Agrawal S, Karcher D, Ruf S, Bock R. The functions of chloroplast glutamyl-tRNA in translation and tetrapyrrole biosynthesis. Plant Physiol. 2020;183(1):263–76. https://doi.org/10.1104/pp.20.00009.
Jost M, Naumann J, Rocamundi N, Cocucci AA, Wanke S. The first plastid genome of the holoparasitic genus Prosopanche (Hydnoraceae). Plants. 2020;9(3):306. https://doi.org/10.3390/plants9030306.
Palmer JD. Comparative organization of chloroplast genomes. Annu Rev Genet. 1985;19(1):325–54.
Bellot S, Renner SS. The plastomes of two species in the endoparasite genus Pilostyles (Apodanthaceae) each retain just five or six possibly functional genes. Genome Biol Evol. 2016;8(1):189–201. https://doi.org/10.1093/gbe/evv251.
Lim GS, Barrett CF, Pang CC, Davis JI. Drastic reduction of plastome size in the mycoheterotrophic Thismia tentaculata relative to that of its autotrophic relative Tacca chantrieri. Am J Bot. 2016;103(6):1129–37. https://doi.org/10.3732/ajb.1600042.
Schneider AC, Braukmann T, Banerjee A, Stefanović S. Convergent plastome evolution and gene loss in holoparasitic Lennoaceae. Genome Biol Evol. 2018;10(10):2663–70. https://doi.org/10.1093/gbe/evy190.
Lam VKY. Phylogenomics and comparative plastome analysis of mycoheterotrophic plants. Ph.D. Thesis. University of British Columbia, Vancouver, Canada. 2016. https://doi.org/10.14288/1.0305829.
Li X, Qian X, Yao G, Zhao Z, Zhang D. Plastome of mycoheterotrophic Burmannia itoana Mak. (Burmanniaceae) exhibits extensive degradation and distinct rearrangements. PeerJ. 2019;7:e7787.
Kim YK, Jo S, Cheon SH, Joo MJ, Hong JR, Kwak M, Kim KJ. Plastome evolution and phylogeny of Orchidaceae, with 24 new sequences. Front Plant Sci. 2020;11:22. https://doi.org/10.3389/fpls.2020.00022.
Yudina SV, Schelkunov MI, Nauheimer L, Crayn D, Chantanaorrapint S, Hroneš M, Sochor M, Dančák M, Mar SS, Luu HT, Nuraliev MS, Logacheva MD. Comparative analysis of plastid genomes in the non-photosynthetic genus Thismia reveals ongoing gene set reduction. Front Plant Sci. 2021;12:602598.
Garrett N. Plastid phylogenomics and molecular evolution of Thismiaceae (Dioscoreales). M.Sc. Thesis. University of British Columbia, Vancouver, Canada. 2022. https://doi.org/10.14288/1.0413200.
The Kimpa Vita University was an essential base for our fieldwork in northern Angola and provided logistical support. We like to thank the authorities of Angola (Angolan research permit N.145/INBAC.MINAMB/2013 granted by the Instituto Nacional da Biodiversidade e Áreas de Conservação – INBAC, Ministério do Ambiente and permit 52/GD/IP-UNIKIVIY2022 granted by the Governo Provincial de Uíge) and Kennedy Wambua Matheka (National Museum of Kenya) for their contribution. We also like to thank the Herbarium Dresdense.
Open Access funding enabled and organized by Projekt DEAL. We appreciate the financial support from ERASMUS for exchange with Kenya, provided by the International Office and the Leonardo Office of TU Dresden. The fieldwork in Angola was supported by a travel fund from the German Academic Exchange Service (DAAD) and the program ‘Strategic Partnerships’ of the TU Dresden.
Ethics approval and consent to participate
The authors confirm that all the experimental research and field studies on plants (either cultivated or wild), including the collection of plant material, were carried out in accordance with relevant institutional, national, and international guidelines and legislation.
Consent for publication
We would like to thank Daniel Nickrent and Barbara Ditsch for providing us with additional photos for Sarcophyte and Thonningia and for giving their consent for these photos to be published.
The authors declare no competing interests.
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
About this article
Cite this article
Kim, W., Lautenschläger, T., Bolin, J.F. et al. Extreme plastomes in holoparasitic Balanophoraceae are not the norm. BMC Genomics 24, 330 (2023). https://doi.org/10.1186/s12864-023-09422-1