Hidden in plain sight: Lindauera gen. nov.: a new genus of Dictyotales from New Zealand

Keywords: Australia, Dictyota papenfussii, Dictyotales, Lindauera gen. nov., Lobospira bicuspidata, New Zealand, Phaeophyceae, phylogenetic analyses.

The brown alga Dictyota papenfussii Lindauer was described in 1949, with additional anatomical and morphological information provided in the later flora treatment (Lindauer 1949; Lindauer et al. 1961). The type material was collected from Pihama, Taranaki (North Island, New Zealand). Lindauer designated specimen VWL 4110 (AK295737) as the ‘Type’ (Fig. 1), and distributed specimens of D. papenfussii labelled ‘Co-types’, as specimen number 187, in Fascicle VIII, of the Algae Nova-Zelandicae Exsiccatae (Nelson and Phillips 1996).

Fig. 1.  Holotype specimen of Dictyota papenfussii Lindauer, AK295737.

The species was reported to occur sporadically in shallow water in the upper subtidal, growing in dense tufts on stones. Lindauer reported that it had a seasonal distribution from late summer through autumn (February to May). Adams (1994) gave the distribution of D. papenfussii as Taranaki, the Cook Strait region, the southern South Island in Fiordland, and Stewart Island (Fig. 2). In both the original description and the flora treatment (Lindauer 1949; Lindauer et al. 1961), there is reference to ‘an identical specimen’ reported by Valerie May Jones from Sydney. However, in their treatment of brown algae from New South Wales, Millar and Kraft (1994) considered the record of Dictyota papenfussii from Sydney to be Dictyota dichotoma (Huds.) J.V.Lamour.

Fig. 2.  Map showing regions of New Zealand where Lindauera papenfussii has been recorded and sites of specimen records confirmed with sequence data.

Dictyota papenfussii was described as having prostrate primary fronds, with dark brown, flat, and linear erect fronds 7–19 cm in height, a percurrent main axis, sometimes forked below, and axes up to 4 mm wide, a decompound holdfast with a stupose stipe, and basally stupose axes. The axes were reported to be spirally twisted, narrowing upward, with subdichotomous-pinnately decompound alternate branches, and very short, regularly alternate ultimate tooth-like ramuli.

The order Dictyotales is diverse and speciose (Bringloe et al. 2020), with peak diversity found in the tropical Indo-Pacific, and Australasia (Küpper et al. 2019). There is a single family and two tribes recognised in the order; members of the Dictyoteae have a single apical cell, whereas members of the Zonarieae are characterised by having many localised or marginal apical cells. As noted by Bogaert et al. (2020), the morphology of species of Dictyota J.V.Lamour. is highly variable, and many species also display considerable intra-specific morphological variability. Bogaert et al. (2020, p. 4) noted that given this morphological plasticity, ‘molecular techniques are indispensable for species identification’. The most common markers used include psaA, psbA, rbcL and cox1.

As part of a wider survey of the Dictyotales, sequence data have been obtained from specimens collected throughout the New Zealand region. Analyses have shown that specimens with the morphology and anatomy of Dictyota papenfussii form a clade that is distinct from Dictyota, both from New Zealand and elsewhere. In these analyses, a psbA sequence in GenBank attributed to Lobospira bicuspidata Aresch., was also placed in this clade. This result contrasts with the analyses conducted by Lee and Bae (2002) using rbcL and 18S rDNA data, in which L. bicuspidata was found to group with other members of the Zonarieae. Lobospira is a monotypic genus endemic to Australia, with a very distinctive thallus apex. As observed by Edelstein and Womersley (1975) the occurrence of a marginal row of apical cells in both the adult axes and juvenile stages indicates that Lobospira is correctly placed in Zonarieae and not the Dictyoteae. Dictyota papenfussii from New Zealand has a single apical cell typical of the Dictyoteae.

In this paper, we describe a new genus for Dictyota papenfussii and provide new data on its geographic, seasonal and ecological distribution, and compare this taxon with sequence data from newly collected material of Lobospira bicuspidata from Australia.

Specimens of Dictyota papenfussii have been collected from a wide variety of habitats throughout the New Zealand region over the past 15 years, pressed as herbarium specimens and subsampled at the time of collection, with material being placed in silica gel for genetic analyses. This study is based on both this recent material, as well as observations of additional specimens in the herbaria of Tāmaki Paenga Hira Auckland Museum (AK) and Te Papa Tongarewa Museum of New Zealand (WELT) (herbarium codes follow Index Herbariorum, see http://sweetgum.nybg.org/ih/, accessed January 2022).

There are six psbA sequences in GenBank ascribed to L. bicuspidata, four (KU249198.1, MW225728.1 MW225729.1, MW225731.1) collected at Jervis Bay, New South Wales (NSW), Australia and two (MW225732.1, MW225733.1) collected at Port MacDonnell in South Australia (Vieira et al. 2016, 2021). The vouchers of material for these sequences were unable to be located in either Australia or Belgium. Two specimens of Lobospira bicuspidata collected from Cape Peron, Western Australia, were borrowed from the Western Australian Herbarium (PERTH 09316523, J.Huisman, 15 February 2021; PERTH 09316639, J.Huisman, 16 February 2021) (Fig. 3), including subsamples dried in silica gel for molecular analysis.

Fig. 3.  Lobospira bicuspidata, PERTH 09316639.

DNA was extracted using the CTAB-Proteinase K method of Zuccarello and Lokhorst (2005). Collection locations and voucher numbers are given in Table 1. We attempted to amplify the following three markers from each specimen: psaA, psbA and rbcL as in Nelson et al. (2018). Polymerase chain-reaction (PCR) products were purified using ExoSAP-IT (Affymetrix, Santa Clara, CA USA), and sequenced at Macrogen Inc. (Seoul, South Korea). Sequences were trimmed and aligned using Geneious Prime (ver. 2023.0.4, Biomatters Ltd, Auckland, New Zealand, see https://www.geneious.com). All attempts to sequence the holotype and isotype material of D. papenfussii were unsuccessful.

Table 1.  Specimen voucher details and GenBank accession numbers of taxa in the phylogenetic analysis.

We used BLAST searches to identify GenBank sequences with high homology to ours, confirming the placement of our specimens within Dictyotales, and assembled a concatenated alignment of our data with sequences from representative Dictyotales taxa from GenBank, including including Canistrocarpus De Paula & De Clerck, Dictyota, Dictyotopsis Sond., Dictyopteris J.V.Lamour., Dilophus J.Agardh, Lobophora J.Agardh, Newhousia Kraft, G.W.Saunders, I.A.Abbott & Haroun, Padina Adans., Rugulopteryx De Clerck & Coppejans, Scoresbyella Womersley, Spatoglossum Kütz., Stoechospermum Kütz., Stypopodium Kütz., and Zonaria C.Agardh (Table 1). We included sequences from Cutleria adspersa (Mert. Ex Roth) De Not., and C. multifida (Turner) Grev. (Tilopteridales), Onslowia endophytica Searles and Verosphacela ebrachia E.C.Henry (Onslowiales), Heribaudiella fluviatilis (Aresch.) Sved., Sphacelaria cirrosa (Roth) C.Agardh and Stypocaulon scoparium (L.) Kütz. (Sphacelariales) as outgroups. GenBank accession numbers of taxa in the phylogenetic dataset are given in Table 1.

The dataset was initially partitioned by marker and by codon (i.e. 9 partitions) and the most appropriate partitioning strategy and models of sequence evolution were identified using IQ-TREE2 (ver. 2.1.2, see https://github.com/iqtree/iqtree2; Kalyaanamoorthy et al. 2017; Minh et al. 2020). We used IQ-TREE2 to estimate the maximum likelihood (ML) tree under this partitioning strategy and models, along with support under both the approximate likelihood ration test (aLRT, Anisimova and Gascuel 2006) and 1000 ML bootstrap replicates.

We used MrBAYES (ver. 3.2.6, see https://github.com/NBISweden/MrBayes/; Ronquist et al. 2012) to implement a Bayesian analysis. Because MrBAYES uses a more restricted suite of models than does IQ-TREE2, we used a second IQ-TREE2 analysis to identify appropriate models and an appropriate partitioning strategy for the Bayesian analysis. Partitioning strategies and models for all analyses are listed in Table 2. For the Bayesian analysis we ran four MCMC chains for 5 million generations, sampling every 1000 generations. Burn-in values were assessed through inspection of log-likelihood plots and average parameter values using Tracer (ver. 1.7.2, see https://github.com/beast-dev/tracer/releases/tag/v1.7.2; Rambaut et al. 2018) and were confirmed by inspection of potential scale reduction factor (PSRF) values calculated in MrBAYES. Trees were visualised using FigTree (ver. 1.4.4, see http://tree.bio.ed.uk/software/figtree).

Table 2.  Partitions and models of sequence evolution used in the concatenated analyses.

Sequence data from specimens identified as Dictyota papenfussii were distinctly different from those of other Dictyota taxa; percentage homology to sequence data from any Dictyota taxon was less than 88% for psaA, 93% for psbA and 92% for rbcL. The phylogenetic matrix contained 62 taxa and 2558 characters. In the phylogenetic analysis (Fig. 4), the clade containing specimens identified as Dictyota papenfussii was clearly distinct from Dictyota, and was resolved in a well-supported monophyletic clade with Rugulopteryx taxa (98/97/1). Maximum pairwise percentage sequence identities between Dictyota papenfussii and Rugulopteryx sequences were 93% (psbA) and 93.5% (rbcL).

Fig. 4.  Maximum-likelihood phylogram estimated from concatenated psaA, psbA and rbcL sequence data. Support values are shown on each branch: approximate likelihood ratio test (aLRT,%) and ML bootstrap (%) values above, and Bayesian posterior probability (PP) values below. Only values greater than 80% (aLRT), 80% (bootstrap) and 0.9 (PP) are shown; all support values are shown if two support methods for a clade reach the cutoff value. Thick lines lead to clades receiving support of 100/100/1 (alrt/bootstrap/PP).

Two psbA sequences from Australia, ascribed to Lobospira bicuspidata, were resolved with the Dictyota papenfussii sequences. The sequence of specimen HV2179 (MW225728.1), from NSW, differed from that of WELT A034505 by one substitution over 766 bp. The sequence of LT0243 from South Australia differed from the NSW sequence by 16 substitutions over 708 bp, and differed from that of WELT A034505 by 17 substitutions over 732 bp. Four further sequences of Australian material ascribed to Lobospira bicuspidata from NSW and one from South Australia were identical to that of HV2179.

We obtained two rbcL sequences from the specimens of Lobospira bicuspidata collected from Western Australia but were unable to sequence either psaA or psbA from these specimens. The two rbcL sequences were identical, and this sequence was resolved with Zonarieae taxa in our analysis, in an unsupported (90.9/58/0.72) clade with Spatoglossum crassum and S. asperum.

On the basis of the level of differentiation of ‘Dictyota’ papenfussii specimens from Dictyota and support for this clade, we propose a new genus.

Type: Lindauera papenfussii (Lindauer) W.A.Nelson & J.E.Sutherl. (=Dictyota papenfussii Lindauer)

Dark brown, flat, erect fronds, narrowing upwards, with subdichotomous to alternate branching. Axes frequently spirally twisted. Attached to solid substrates by primary, prostrate, dichotomously branched axes.

The genus is named to recognise the contributions of Victor W. Lindauer (1888–1964), a pioneering New Zealand phycologist, and, in particular, his work documenting the brown algal flora.

(Fig. 1.)

Dictyota papenfussii Lindauer, Trans. & Proc. Roy. Soc. New Zealand 77: 390, p. 38: fig. 1–3; p. 39: fig. 17 (1949). Type: New Zealand, North Island, South Taranaki, Pihama, Campbell’s beach, 25 Feb. 1944, V.W.Lindauer 4110 (holo: AK (AK295737); iso: distributed as specimen number 187, in Fascicle VIII, of the Algae Nova-Zelandicae Exsiccatae (Nelson and Phillips 1996) e.g. WELT A000987, A020630).

Fronds, erect, flat, 2–4 mm wide, (3)7–15(20) cm in height; lower axes mid- to dark brown, upper axes golden brown. Axes frequently spirally twisted, narrowing upwards, with subdichotomous-pinnately decompound alternate branches. Apices vary in shape from pointed (Fig. 5a) to rounded (Fig. 5b). Basal axes stupose. Attached to solid substrates by primary, prostrate dichotomously branched axes, up to 2 cm long (Fig. 6a, b), arising directly from the lower main axis. Thalli with 3 cell layers – a single medullary layer of large rectangular cells, and a unilayered cortex of small cuboidal cells on each surface. Sporangia, large, solitary, scattered on upper part of thallus on both surfaces, with single sporangial stalk cell. Oogonia and antheridia infrequently seen.

Fig. 5.  Specimens showing morphological variation present in Lindauera papenfussii. (a) WELT A026094, Doubtful Sound, Fiordland; (b) WELT A032190, Te Miko Reef, Bay of Islands.

Fig. 6.  Attachment to shells with prostrate axes (a, b). WELT A032191.

N. M. Adams, Seaweeds of New Zealand 105 (1994); W. A. Nelson, New Zealand Seaweeds 61 (2020).

NEW ZEALAND: Manawatāwhi/Three Kings Islands: Princes Islands, Rosemary Island, Cattons Cave, 13 Apr. 2013, R.D’Archino, AK384069. North Island: Bay of Islands, Deep Water Cove, 30 Nov. 2009, R.D’Archino, WELT A030593; Bay of Islands Te Miko Reef, 10 Feb. 2010, S.Miller & R.D’Archino, WELT A032190; Bay of Islands, Te Miko Reef, 11 Feb. 2010, R.Stewart & B.Crocker, WELT A032191; Cavalli islands, Motokawanui, 9 Apr. 2013, R.D’Archino, AK384068. South Island: Marlborough Sounds, Queen Charlotte Sound mouth, White Rocks, 20 Nov. 2008, R.D’Archino, WELT A034492; Queen Charlotte Sound mouth, D’Urville I., 27 Mar. 2017, beam trawl, WELT A034505; Arnott Point, 10 Dec. 200, N.Shears, WELT A026837; Fiordland, Doubtful Sound, Secretary Island, 22 Jan. 2000, C.Duffy, WELT A026094; Shelter Islands, 27 Jan. 2008, F.Thomas, C.Hepburn, & D.Richards, WELT A034491; Fiordland, Breaksea Sound, 19 Dec. 2019, R.D’Archino, WELT A034502; Fiordland, Dusky Sound, Five-Fingers Peninsula, 14 Nov. 1984, P.Brotherson, WELT A016265. Rakiura/Stewart Island: Port Pegasus, Hell’s Gates, 4 Mar. 2009, C.Hepburn, AK384065; Port Pegasus, Pearl Island outer, 6 Mar. 2009, R.D’Archino, AK384067, and Pearl Island, 3 Dec. 2018, M.Desmond & A.Kluibenschedl, WELT A034498.

On the basis of the results of phylogenetic analyses and examination of herbarium specimens, the distribution of Lindauera papenfussii is now understood to extend in New Zealand from Manawatāwhi/Three Kings Islands, and the northern North Island through to the southern South Island (particularly Fiordland), and Stewart Island (Fig. 2), and in Australia it has been found in both New South Wales and South Australia.

Lindauera papenfussii is found from the upper subtidal zone through to depths of at least 20 m, usually in areas of moderate exposure. It is found growing on rocky reefs as well as on stable shell hash and cobbles. It can be locally abundant, forming dense patches.

Specimens have been collected throughout the year, with fertile material recorded from November through to February (late spring through summer).

Nelson et al. (2019) categorised Dictyota papenfussii as ‘data deficient’ on the basis of the information available at the time of assessment.

The morphology of L. papenfussii is quite variable, having pointed to rounded tips, and in the height that thalli attain. In northern areas, thalli are typically smaller than those found in the south and generally have more rounded and diverging apices than specimens collected from Taranaki south.

Despite sharing many features with Dictyota, Lindauera differs from other genera in the Dictyoteae in that it does not have the rhizoidal attachment typical of the tribe, but rather possesses prostrate, primary fronds. In the emended description of Dictyota (De Clerck et al. 2006), and as summarised by Bogaert et al. (2020), most Dictyota species attach to substrates by way of multicellular, uniseriate, basal or marginal branching rhizoids, although a few species attach by stoloniferous holdfasts. Rugulopteryx, the most closely related genus to Lindauera in our analyses, possesses both basal rhizoids, and stoloniferous holdfasts, whereas in Canistrocarpus attachment to substrates is by basal or marginal rhizoids and for Dilophus by rhizoids or stolons (Womersley 1987; De Clerck et al. 2006).

De Clerck et al. (2006) described Rugulopteryx and Canistrocarpus by using a combination of sequence data and anatomical and morphological features (e.g. number of medullary cell layers, undulate thalli, presence of uni- or multi-cellular antheridial paraphyses, number of sporangial stalk cells, presence or absence on sporangial involucre). Although sporangial material of Lindauera has been found, no recent material has been collected displaying antheridia or oogonia, nor any herbarium specimens that have been examined. In the original description Lindauer (1949) stated ‘sexual organs not seen’, but in Lindauer et al. (1961) there is a description of pyriform oogonia and antheridial sori, neither of which were illustrated. The predominance of sporophytic thalli is not unusual in the Dictyoteae (e.g. Phillips et al. 1990; Hwang et al. 2009; García-Gómez et al. 2021). Although research on gametogenesis in members of the order is limited, diurnal and lunar periodicity have been shown to influence gametogenesis and release of gametes (Müller 1962; Phillips et al. 1990; Bogaert et al. 2020). Further work is required to understand reproduction in Lindauera, including the influence of seasonality, and whether diurnal and lunar periodicity influence the production and release of gametes and spores.

We consider that the sequences in GenBank from specimens collected in Jervis Bay, New South Wales, and Port MacDonnell, South Australia, have been incorrectly assigned to Lobospira bicuspidata Aresch. These sequence data confirm the presence of Lindauera in Australia and support Lindauer’s observation about material collected from Sydney being ‘identical to D. papenfussii’. On the basis of morphology, Lobospira belongs in the Zonarieae, possessing a distinctive row of 8–12 apical cells in a terminal invagination. The sequence data from newly collected specimens of Lobospira bicuspidata from Western Australia, that displayed the very distinctive apex of this genus, were placed in our analyses within the Zonarieae. Further work and field investigations are required to understand the geographic and seasonal distribution of Lindauera in Australia.

We have documented a more extensive geographic range in New Zealand for L. papenfussii than formerly understood, but there remain large gaps in its recorded distribution that are likely to reflect collection gaps. In addition, further work is required in New Zealand to understand the seasonality and ecology of this species.

The data that support this study are available in the article, specimens accessible in two publicly accessible herbaria (AK, WELT), and sequence data submitted in GenBank.

The authors declare that they have no conflicts of interest.

Research was supported by NIWA SSIF funding in the Oceans National Centre.