The Methuselah of plant diaspores: Sphagnum spores can survive in nature for centuries

A persistent sexual propagule bank (seeds and spores) is important for plant population regeneration (Saatkamp et al., 2014) and for the maintenance of genetic diversity (Hock et al., 2008). Many phanerogams produce long-lived seeds that may remain viable for centuries, with the oldest reliable records for seed longevity demonstrated in the Sacred Lotus Nelumbo nucifera Gaertn. buried in a peatland for 1300 yr (Shen-Miller et al., 1995), the Date Phoenix dactylifera L. stored in a desert climate for 2000 yr (Sallon et al., 2008), and fruit tissue of Silene stenophylla Ledeb. buried in Siberian permafrost for 30 000 yr (Yashina et al., 2012). Reports about long-lived bryophytes have recently been published, with the most extreme case in bryophyte material from the Antarctic, which regrew after being buried for > 1500 yr (Roads et al., 2014). In Sphagnum (peat moss) – the dominant plant group in northern peatlands – Clymo & Duckett (1986) observed that shoots regenerated from 25- to 60-yr-old peat slices, and some of them probably originated from spores. In fairly well-dated layers of forest soil, some bryophyte diaspores buried for more than a century were still viable (Dyer & Lindsay, 1992). A 3-yr burial experiment revealed that many Sphagnum spores survived and the rate of viability decline suggested a potential longevity of several decades or even centuries (Sundberg & Rydin, 2000). However, there is no direct evidence that bryophyte spores might survive for centuries in the field. In this study, we tested Sphagnum spore viability in peat cores from two peatland micro-habitats down to a depth of 0.5 m, representing up to 680-yr-old peat. The peat cores were obtained from Hani Peatland (42°13′N, 126°31′E), northeast China, a transitional mire of 17 km2. It is located in the temperate zone with −15.7 and 22.5°C mean temperatures for January and July, respectively, and a mean annual precipitation of 765 mm yr−1. The peatland has a 16-kyr development history (Zhou et al., 2010). The vegetation is differentiated into forested and nonforested parts. Sphagnum palustre L., S. magellanicum Brid., S. affine Ren. & Card., S. fuscum (Schimp.) Klinggr. and S. capillifolium (Ehrh.) Hedw. are the most common peat mosses. Of these, S. capillifolium produces spores most frequently and is distributed mainly in the forested part, where Larix olgensis Henry, Rhododendron tomentosum Harmaja, Vaccinium uliginosum L. and Smilacina japonica A. Gray dominate. A macrofossil analysis by Stebich et al. (2011) and carbon-14 (14C) dating by Zhou et al. (2010) showed that during the Holocene, the peatland had experienced several cold and wet, to warm and dry, climate cycles. Sphagnum started to re-dominate the peatland around two centuries ago after having lost dominance c. 3800 yr before present (bp). In July 2009, we took four 50-cm-deep, 5 cm × 5 cm peat cores from the centre of a S. capillifolium-dominated hummock in the forest margin of Hani Peatland. In September 2012, we took four 50-cm deep, 5-cm diameter semi-circlular cores from the centre of a S. capillifolium-dominated hollow close to the hummock. The surface of the hummock and the hollow were c. 35–45 cm above, and from 5 cm above to 5 cm below, the average water table, respectively. In the laboratory, all the cores were cut into 50 1-cm-thick slices. Three cores were used for spore extraction and one core for bulk density measurements and dating. For spore extraction, the outer 1 cm and 0.5 cm for hummock and hollow slices, respectively, was discarded to remove potential contaminants, and the rest was air-dried and then ground into a powder. Sieves with a 38.5-μm mesh width were used to primarily remove larger remains of Sphagnum gametophytes and other particulate matter. The filtered solution was filtered again in sieves with a 10-μm mesh width. Sphagnum spores (and other remains) left in the sieve were diluted in 3 ml of distilled water to produce a spore suspension. We added culture medium made from 0.2 g agar and 10 ml nutrient solution (Rudolph et al., 1988) to a 6-cm-diameter Petri dish. One millilitre of spore suspension was poured evenly onto the medium for cultivation and testing of spore viability in a growth chamber (Harbin Donglian Electronic & Technology Development Co. Ltd, Harbin, China). The cultivation conditions were 16 h, 27°C : 8 h, 20°C (day : night), light 60 : 0 μmol m−2 s−1. Most viable Sphagnum spores began to absorb water and swell within 1 wk, and germinated within 3 wk. After 3 wk of cultivation, thalloid protonemata began to obscure the agar surface, and no new germinating spores were seen in the uncovered agar. At this time, the proportion of viable (germinating) spores was assayed using a microscope at ×200 magnification. Spores that had developed a filamentous protonema were defined as viable. The chronological construction of the peat cores was conducted by three dating methods: caesium-137 (137Cs), lead-210 (210Pb) and 14C. Most peat layers were dated by 137Cs and 210Pb techniques. These radionuclides were screened by direct gamma assay at the Paleoecology Laboratory, Institute for Peat and Mire Research, Northeast Normal University for the hummock cores, and at the State Key Laboratory of Lake Science and Environment, Nanjing Institute of Geography and Limnology, Chinese Academy of Sciences for the hollow cores. Although the 137Cs profiles and 210Pb profiles of the peat cores can be used for estimating mean linear accumulation rates of the peat cores (Ali et al., 2008), the estimated rates may contain large uncertainties, especially for the hollow cores. Thus, we used the results from different dating methods to date each core. The initial burial time (IBT) for the uppermost layer was the coring year 2009 and 2012 for the hummock and the hollow, respectively. IBT of the layer with the 137Cs peak was regarded as c. ad 1963 (Ali et al., 2008). The IBT of the first layer with zero excess 210Pb should be c. 170 ± 10 yr ago since excess 210Pb will fully decay after 7.5 half-lives of 210Pb (T1/2 = 22.3 yr) (22.3 yr × 7.5 ≈ 170 yr) (Appleby & Oldfield, 1992; Ali et al., 2008). We further used plant residues separated from the bottom peat layer of the hollow core for accelerator mass spectrometry (AMS) 14C measurements at the laboratory of Beta Analytic Inc. (Miami, FL, USA). Hence, for the hollow core, an AMS 14C date of 680 ± 30 yr at 50-cm depth was used. The relationship between depth and age of the hummock and hollow cores is shown in Fig. 1. In total, 45 080 and 18 178 spores were observed in the full hummock and hollow profiles, respectively. In cores from both habitats, spores were relatively few in the uppermost layers (Fig. 2a), which was likely the result of downwash of spores from the young and loose surface peat to denser, more saturated layers. This is in line with the results of Clymo & Mackay (1987) who experimentally demonstrated that pollen grains and spores were displaced by a median of 1.5 cm, but occasionally up to 10 cm, along the vertical water flow direction in unsaturated, acrotelm peat and living S. papillosum shoots. The realized vertical displacement will then decrease along with peat decomposition and compaction, such that an original displacement of 10 cm at the top of the incipient peat column may correspond to a 1-cm displacement at 50 cm below the peat surface (cf. Fig. 1). Maximum spore numbers were observed in the 37th and the 6th layers of the hummock and the hollow cores, respectively (Fig. 2b), where the difference in depth probably reflects the faster and greater degree of compaction in the hollow peat (bulk density of dried peat of the upper 20 layers (mean ± 1 SE): 0.96 ± 0.055 g cm−3 and 0.56 ± 0.036 g cm−3 in hollow and hummock cores, respectively; F1,38 = 3.377, P = 0.074). The decline of spore numbers at greater depths (despite that these represent longer periods of time) may reflect the difficulty in extracting spores, which were attached to larger peat particles, or the decomposition of spores. Germinating spores were found in all the peat layers of both habitats. The germination and protonema dynamics were similar to in 1- to 3-yr-old spores stored in the laboratory, except that quite a few thalloid protonema developed in the former (Fig. 3). Germination percentage (GP) was 10.5 ± 2.4% (mean ± 1 SE) and 9.1 ± 0.6% in the hummock (170 ± 10 yr) and hollow bottom layers (680 ± 30 yr), respectively (Fig. 4). GP of spores from different depths is reported in Supporting Information Table S1. Overall, GP in all the 50 layers was greater in the hollow (32 ± 2.3%) than in the hummock (25 ± 3.4%; U50,50 = 840, P = 0.005). Burial time significantly affected GP (hummock: df = 49, χ2 = 128.471, P < 0.001; hollow: df = 49, χ2 = 122.351, P < 0.001). GP in relation to burial time showed a decrease according to an exponential function (hummock: y = 0.3779e−0.0155x, R2 = 0.618, P < 0.001; hollow: y = 0.4722e−0.0027x, R2 = 0.837, P < 0.001) (Fig. 4). Most layers (68%) of hummock spores showed GP below 25%, while only 26% of the hollow layers were found within this range. However, spores of the uppermost layers in the hummock showed a higher germinability than in corresponding layers of the hollow. Apparently, the anoxic environment of hollows rapidly reduces germinability of a majority of the spores but promotes longevity in the remaining spores. The sharp decrease in GP for hummock spores may be the result of drought and dry-freezing before the spores enter into water-saturated peat. The longevity of Sphagnum spores, defined as the point when viability falls below 1% (Sundberg & Rydin, 2000), was predicted to be 230 and 1400 yr (according to extrapolation of the fitted models), while half-life was 18.3 and 21.2 yr in hummocks and hollows, respectively. In this study, a high percentage of the c. 170- and 680-yr-old spores from the cores of a hummock and a hollow, respectively, were still able to germinate, and this shows that Sphagnum can form an extremely long-term persistent spore bank. The models indicate that the spores may survive for a millennium or more in nature. This is the greatest longevity demonstrated for bryophyte spores. Exceptional longevity of the Sphagnum spores and lotus seeds may be attributed to the peatland environment, with water-logged, anoxic, acid and nutrient-deficient conditions, resulting in a weak metabolic activity of both spores and decomposing microorganisms (Bragazza et al., 2006). It appears that peatlands as well as ice/permafrost and dry (desert) habitats lead to environments that retard metabolic rates and may thus lead to very long-lived plant diaspores (Shen-Miller et al., 1995; Sallon et al., 2008; Yashina et al., 2012; La Farge et al., 2013; Roads et al., 2014). Long-lived spores may be important for Sphagnum in colonizing new sites since they can await decent germination conditions in a changing and fluctuating environment. They can be a reserve that is added to the current spore production, thus increasing the effective spore numbers by up to two orders of magnitude, and swamping the effect of variable spore production among years (Sundberg, 2002). Sphagnum spores may disperse far, often several hundred kilometres and even transcontinentally or intercontinentally (Sundberg, 2013), but they may suffer from UV radiation, drought and dry-freezing (van Zanten, 1978). Great longevity in buried spores may thus be a side-effect during the evolution of spores, enabling them to resist harsh conditions during long-distance transport. This study was funded by the National Key Research and Development Project (no. 2016YFA0602301), the National Natural Science Foundation of China (no. 41471043 and no. 41371103), Department of Human Resources and Social Security of Jilin Province. The authors thank Yuan Tang and Li-Hong Jiang for their work on a pilot study, Gao-Lin Zhao, Jin-Ze Ma, Jing Zeng for peat core sampling and processing, Xiao-Yuan Zhong, Ya-Min Feng, Cong Xu, Qi-Min Gui and Ai-Qin Sun for help with the germination experiment, and Dicky Clymo and three anonymous reviewers for valuable comments on the manuscript. Z-J.B. conceived and designed the experiment. L.F. performed the germination experiment. H-K.L. and H-Y.Z. carried out radiometric dating. Z-J.B. and H-C.L. analysed the data. Z-J.B. and S.S. wrote the manuscript. Please note: Wiley Blackwell are not responsible for the content or functionality of any Supporting Information supplied by the authors. Any queries (other than missing material) should be directed to the New Phytologist Central Office. Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.