Spore germination and young gametophyte development of the endemic Brazilian hornwort Notothylas vitalii Udar & Singh ( Notothyladaceae-Anthocerotophyta ) , with insights into sporeling evolution

Notothylas vitalii is an endemic Brazilian hornwort species, easily identifi ed by the absence of pseudoelaters and columella, and the presence of yellow spores. Plant material was collected in Recife, Brazil, and the spores were sown onto Knop’s medium, germinating after thirty days only with the presence of light. Germination occurred outside the exospore, and only after the walls had separated into three or four sections did a globose sporeling initiate its development. Following longitudinal and transversal divisions, the initial loose mass of cells became a thalloid gametophyte, subsequently developing into a rosette-like juvenile thallus with fl attened lobes. Additional information concerning sporeling types in key genera of hornworts, such as Folioceros and Phymatoceros, will be crucial for inferring the possible ancestral type and the evolution of this trait among hornworts. Our study supports the necessity of supplementary studies on sporeling development, combined with morphological and phylogenetic investigations, to help elucidate the evolution of the Anthocerotophyta and their distribution patterns.

Notothylas is a genus of ephemeral plants that grow on open soils as pioneer plants (Shaw & Renzaglia 2004).They are very distinct hornworts, with reduced and horizontal sporophytes enclosed within involucres (Udar & Singh 1981;Cargill et al. 2005;Chantanaorrapint 2015).Three Notothylas species are recognized in Brazil (N.breutelii, N. orbicularis, and N. vitalii); Notothylas vitalii is endemic to Brazil and has been recorded in the central-western, northern, northeastern, and southeastern regions of that country (Yano et al. 2011).
Notothylas vitalii is commonly encountered growing on garden soils and near plantations, generally appearing during rainy months and disappearing during the dry season (Martins et al. 1990).There are no dehiscence lines on the capsules of N. vitalii (Fig. 1A, B), and Udar & Singh (1980) noted that its spores are usually dispersed by a transversal separation of the apical portion of the sporangium; pseudoelaters and columella are also absent (Udar & Singh 1980;Martins et al. 1990;Gradstein & Costa 2003).Benitez et al. (2012) reported N. vitalii for Ecuador as being characterized by the presence of pseudoelaters and longitudinal sporangium opening -but this suggests a misidentification, as those traits are descriptions of Notothylas orbicularis.This type of error often occurs because the descriptions of N. vitalii and N. orbicularis were inverted by Gradstein & Costa (2003).
Even though N. vitalii is considered endemic to Brazil, it has characters that resemble Notothylas javanica, including the presence of sporangium without a special dehiscence line, the absence of pseudoelaters, a non-persistent/absent columella, and yellow spores (Stieperaere & Matcham 2007;Chantanaorrapint 2015).Taxonomic studies will need to be undertaken, whoever, to confirm if N. vitalii might be N. javanica in South America.
To determine if the sporeling pattern observed in other species of Notothylas is maintained in the rather peculiar species N. vitalii, we examined its spore germination, sporeling development, and juvenile thallus.Our study likewise contributes to a better understanding of sporeling evolution among hornworts based on recent phylogenies and data on sporeling types available in the literature.
Plant material for the study was collected in September 2005 on the campus of the Federal University of Pernambuco (Recife, Pernambuco State, Brazil; 8°03'09.3"S,34°57'11.7"W).In September 2005, fresh, mature sporophytes of Notothylas vitalii Udar & Sigh were detached from the gametophytes, sterilized in 1.5 % sodium hypochlorite for 5 min, and subsequently rinsed with distilled water (Duckett et al. 2004).Forty sporangia were opened and the spores removed, homogenized, and spread onto 10 mL of Knop´s nutrient solution (ca. 100 spores per mL; Nehira 1988) plus nystatin fungicide (100 U mL -1 ) solidified with 0.4 % bacteriological agar in Petri dishes (5 cm diameter) (Silva et al. 2006;2009).Five replicates were used per treatment (under both continuous light and continuous darkness), randomly arranged and then rearranged on a weekly basis.Plastic materials utilized in these experiments were immersed in 70 % ethanol for over 48 h, and then dried under laminar air flow.Petri dishes and nutrient solution were autoclaved for 20 min at 120 ºC, at 1 Kg cm -2 .Culture procedures were performed in a laminar air flow chamber in a growth room under controlled conditions (ca. 25 °C, irradiance ca.20-40 µmol m -2 s -1 , 12 h photoperiod, or continuous darkness -wrapped in black plastic bags).
Daily observations were made (using a dissecting microscope) to detect spore germination and to analyze the early phases of gametophyte development; the study materials were photographed using a digital camera coupled to an optical microscope.
The spores of Notothylas vitalii were observed (using an optical microscope) to be golden colored, unicellular, with smooth surfaces and a trilete scar, and to contain oil droplets (Figs.1C, 2A).Spore germination occurred after approximately thirty days of laboratory culture, but only when exposed to light.Late germination is common among some hornwort species, such as Phymatoceros bulbiculosus (varying between 11 and 40 days, depending on temperature and photoperiod conditions) (Herguido & Ron 1989) and Anthoceros punctatus, A. erectus, and Notothylas levieri (10, 24 and 35 days respectively) (Mehra & Kachroo 1962); in contrast, the spores of N. javanica germinated after only two days (Mehra & Kachroo 1962).A light requirement for germination (i.e., positive photoblastism), combined with the long latent period for spore germination, suggest that N. vitalii can establish soil spore banks.This would explain why this species (similar to other hornworts in Brazil) appears in ephemeral and annual populations one or two months after a rainy period and then quickly disappear (O Yano unpubl.res.).Anthoceros agrestis and Phaeoceros carolinianus, for example, develop spore banks in cultivated fields in Switzerland, and their populations are favored by the soil management techniques used for crop cultivation (Bisang 1995;Cailliau & Price 2007).
A small globose sporeling develops in N. vitalii, with each of its cells having a single large chloroplast (Fig. 2D); although its gametophytes have pyrenoids (Villarreal & Renner 2012), none were detected in the early stages of sporeling development.The cells continue dividing into several groups of four cells, forming a loose mass, and, subsequently, the thallus.In contrast, species of Notothylas, Phaeoceros erectus (as Anthoceros erectus), and Anthoceros punctatus produce a germ tube that develops into a column of cells that later divides to form a thallus (Mehra & Kachroo  1962).The sporeling of P. bulbiculosus starts as a globose and loose mass of cells, later becoming a cylindrical columnar structure, with peripheral cells that are larger than its apical cells, which continues to divide and dichotomize the thallus (Herguido & Ron 1989).In Megaceros tosanus, a primary cell divides continuously to form a globose and compact mass of cells (Nehira 1983;Renzaglia 1978), while in Dendroceros and N. renzagliensis the mature spore (still inside the capsule) develops into a compact protonema (Nehira 1983).
The wedge-shaped apical cells of N. vitalii eventually divide longitudinally and transversely into smaller cells (Fig. 2E), with subsequent cell divisions producing a thalloid gametophyte (Fig. 2F) whose terminal lobes become flattened over time.This pattern has been observed in many other hornwort species, such as Phaeocerus erectus (as Anthoceros erectus), Notothylas javanica, and Megaceros tosanus (Mehra & Kachroo 1962;Renzaglia 1978;Nehira 1983).Rosette-like juvenile thalli were observed with N. vitalii after ca.fifty days of culture under laboratory conditions (Fig. 2G); the early gametophytes resembled those of N. orbicularis (Renzaglia 1978), with a globose multicellular mass formed by wedge-shaped apical cells growing into a rosette-shaped thalloid gametophyte.The sporelings of N. vitalii differ from those of N. levieri and N. javanica (which are globose-cylindrical multicellular clumps that develop into a cylindrical mass of cells) (Mehra & Kachroo 1962); the results for N. levieri and N. javanica should be carefully considered, however, as the authors used only a weak light source to grow the sporelings.
We were not able to infer the possible ancestral sporeling type of the hornworts as there is no information available concerning the sporeling types of some key genera, such as Folioceros.If Folioceros has a sporeling type similar to Leiosporoceros (club-shaped), it would indicate that a clubshaped mass of cells was the ancestral type in hornworts (Fig. 3).We suggest, however, that the Phaeoceros-Notothylastype (a cylindrical/globose mass of cells) appeared in a branch that diverged into three families -Notothyladaceae, Phymatocerotaceae and Dendrocerotaceae.Additional studies will still need to be carried out, however, to confirm the sporeling type of Paraphymatoceros (Notothyladaceae).Finally, it is unknown when large masses of cells and endospory appeared in the other genera of Dendrocerotaceae (Fig. 3) -especially as Villarreal et al. (2012) suggested that endospory arose independently in Nothoceros renzagliensis and in species of Dendroceros.This data shows the imperative of investigating the sporeling types of the missing hornwort genera to fully understand the evolution of that character.
In conclusion, the germination pattern of Notothylas vitalii, an endemic Brazilian species, adds to our knowledge of the Anthocerotophyta.Unfortunately, as discussed by Renzaglia et al. (2007), this group is still the most neglected and poorly studied major group of land plants, with few publications focusing on their sporelings.This character is still unknown for some key genera, such as Folioceros and Paraphymatoceros.Supplementary studies on sporeling development will therefore be necessary to further elucidate the evolution and distribution patterns of hornworts, and it will be important to combine morphological and phylogenetic studies to further our understanding of the evolution of this group.

Figure 1 .
Figure 1.Notothylas vitalii Udar & Sigh.A -Sporangium inside the involucre; B -Apical region of the sporangium, with no indication of a true dehiscence line; C -Yellow spores with trilete marks.

Figure 3 .
Figure 3. Sporeling types in an optimized hornwort phylogeny, simplified from Villarreal & Renner (2012; 2013).The Leiosporocerostype, with a club-shaped mass of cells; the Anthoceros-type, with a germ-tube shaped sporeling; the Notothylas-Phaeoceros-type, with a mass of cells; the Nothoceros-Megaceros-type, with a large globose mass of cells; and the Nothoceros-Dendroceros-type, showing endosporic germination.Please see the PDF version for color reference.