Making Eden

by David Beerling

  tions began with a liverwort species by the name of Marchantia paleacea, which originated from soils beneath the temperate cloud forests of Veracruz, Mexico,

  and the cells and tissues of which are colonized heavily by arbuscular mycor-

  rhizal fungi.26 Our small breakthrough came when we discovered how to

  manipulate the fungal colonization of young plants nurtured from tiny vegetative

  propagules (gemmae) sitting in splash cups on the surface of the thallus. Once

  displaced from splash cups, gemmae develop into small plants, known rather

  charmingly as gemmalings.27 These can be cultivated either in sterilized soil or

  adjacent to mature mycorrhizal adult plants to transfer fungal colonization.

  By means of this simple procedure, we could establish gemmalings of relatively

  uniform size and shape either with or without their symbiotic fungal partners,

  and this set the stage for some much-needed experiments. Before this, the

  roadblock to investigating the Pirozynski and Malloch hypothesis had always

  been that mycorrhizal fungi resisted all attempts at being grown and manipulated

  under laboratory conditions. The solution to the problem was to turn it around

  and manipulate the fungal colonization of the host plants.

  Evidence gathered from experiments conducted with the gemmalings soon

  made clear what the fossil record could not tell us: liverworts gained substantial benefits from entering into a symbiotic alliance with mycorrhizal fungi.28 When

  conducted in a carbon dioxide-rich atmosphere like that of the Palaeozoic, some

  400 million years ago, liverworts colonized by mycorrhizal fungi grew three

  times larger than their counterparts lacking any mycorrhizal fungal associations, and had a dramatically higher reproductive output. This is what evolutionary

  132 a Ancestr Al AlliAnces

  biologists like to call the ‘fitness’ of an organism, and gemmae production jumped four-fold in liverworts with mycorrhizal fungal partners, compared to those

  plants without them. If such benefits accrued to the pioneers on land, we can

  imagine that the carbon dioxide-rich atmosphere of the Palaeozoic accelerated

  the greening of the land by fostering the prodigious production of tiny green

  emissaries, for establishing satellite populations. Supplying the mycorrhizal fungi with a radioactively labelled source of phosphate demonstrated the transfer of

  this essential nutrient to the tissues of the host plants. Conversely, supplying liverworts with radioactive carbon dioxide demonstrated that host plants reciprocated

  by provisioning organic carbon to support fungal growth and proliferation, to

  produce in the fungus a staggering absorptive area for nutrient uptake from soil, equivalent to that of a tennis court.

  The exciting conclusion drawn from these experiments is that partnerships

  between liverworts and mycorrhizal fungi seem to operate in a ‘mutualistic

  symbiosis’ comparable to that found in trees and grasses. At last, Pirozynski and Malloch’s hypothesis appeared to have gained crucial experimental verification.

  Extrapolated to the past, these findings support the notion of proto-bryophyte-type early land plants teaming up with fungi to obtain scarce nourishment from the

  soil.

  Satisfying though all this may be, the story does not end here. In 2011, Martin

  Bidartondo of the Royal Botanic Gardens, Kew, and his colleagues, made a sen-

  sational discovery.29 Bidartondo is a leading expert in the use of DNA finger-

  printing to identify the fungal partners of land plants. Applying his DNA-based

  probes to strange liverworts collected from New Zealand, Bidartondo and his

  team, to their surprise, did not find the arbuscular mycorrhizal fungi they

  expected. Instead, these liverworts turned out to harbour a very ancient and

  partially saprotrophic fungal lineage (saprotrophs, recall, obtain energy by

  feeding on decaying matter). This peculiar and primitive fungal lineage is an

  early-evolving group that includes pea truffles (order Endogonales) and is pos-

  sibly older than the arbuscular mycorrhizal fungi (order Glomerales).30 A cau-

  tionary note is warranted here because it is difficult to date accurately the early diversification of fungi. Nevertheless, reconstructions based on the best available evidence support the origin of Endogonales fungi as pre-dating that of arbuscular mycorrhizal fungi. What Bidartondo and his team had discovered was that

  Ancestr Al AlliAnces a 133

  one of the oldest phyla of fungi colonized the oldest lineage of liverworts.

  Classified within the taxonomic class of Haplomitriopsida, these primitive liver-

  worts represent living relatives of early land plants that descended from fresh-

  water algae by adapting to life out of the water hundreds of millions of years

  ago. Until the taxonomic confusion about the evolutionary relationships within

  bryophytes (liverworts, hornworts, and mosses) and vascular plants is resolved,31

  it is difficult to be sure how to interpret the plant–fungus relationship, but it raises the question: could this be the ancestral symbiosis that operated early in plants’

  tenure on dry land?

  Here’s the rub, though. If we want to make the argument that this weird part-

  nership between primitive liverworts and fungi facilitated the arrival of terrestrial plant life, we need to know about the functional status of the association. Just

  because the two groups of organisms live in close association with each other, we cannot assume a harmonious relationship. Tip the balance in favour of the plant

  and it becomes a commensal relationship, i.e., one organism benefiting without

  adversely affecting the other. Tip it in favour of the fungus, on the other hand, and the fungus drains resources from the plant and becomes a parasite. Observations

  of fungi and their intimate connections with the interior of cells of an extant

  lineage of ancient liverworts are not enough to conclusively establish the status of the association, and the challenge is undertaking experiments with living specimens to find out.

  Confined largely to the Southern Hemisphere landmasses, the Haplomitriopsida

  liverworts that Bidartondo and his colleagues studied have a centre of diversity

  in Australasia. This geographical distribution suggests a pattern reflecting the

  far-off days of the Palaeozoic, when the group was widely distributed across

  the southern supercontinent of Gondwana, which later fragmented into Africa,

  South America, India, Antarctica, Australia–New Guinea, New Zealand, and New

  Caledonia. We collected specimens for experiments by visiting New Zealand’s

  South Island, a strange paradise marked out by one of the richest assemblages

  of evolutionarily ancient land plants to be found anywhere in the world.

  Consequently, it offers the best modern analogue we have for the earliest stages of the colonization of the land by multicellular plant life. Bryophytes heavily colonize massive exposed rock faces, coating them with a green veneer of photosynthetic

  life, drip-fed with water draining from the thin infertile soils above. In a scene

  134 a Ancestr Al AlliAnces

  played out endlessly over millions of years, liverworts, hornworts, and mosses

  jostle side-by-side for light and what dissolved nutrients there are to be found in the films of water coating the fine-grained rocks.

  Our plant-collecting trip to the west coast of South Island was in December,

  when the keening of the blackbirds in the evenings reminded us it was summer-

  time in the Southern Hemisphere, where the climate, at a
latitude of 42°S, is

  mild and wet. Settlers introduced the blackbirds and other songbirds from

  Great Britain, as sentimental reminders of home. The rainfall that soaks the

  western coastline of South Island is generated by the rocky spine of the Southern Alps, which intercepts the westerly winds bringing in moisture-laden clouds

  from the Tasman Sea. This feeds glaciers, lakes, and wild temperate rainforests,

  creating ideal habitats for bryophytes to thrive and form close-cropped chloro-

  phyllous carpets in this quiet, unspoilt corner of the world.

  English bryologist and retired professor of botany, Jeff Duckett, of the Natural

  History Museum, London, guided us to collecting localities here and further

  north, in the pristine, moss-draped, old-growth southern beech ( Nothofagus) forests. Duckett is usually found dwelling in the Natural History Museum’s

  attic, but for that fortnight he was gasping fresh air as part of the team hunting antipodean bryophytes. In companionable mode, Duckett kept us entertained en

  route to the collecting sites with tales of botanists getting married underwater, not to mention the curious incident of the three distinguished research professors who holidayed in Iceland, only to be stuck in their hotel after discovering that one was unable to drive, the other would not drive on the right-hand side of the road, and the third only drove on tarmac road surfaces.

  The fascinating carpet of bryophytes beneath our feet soon transfixed us, and

  before long we discovered patches of liverworts belonging to the genus Haplomitrium (Haplomitrioposida) (Figure 20). These plants have a strange appearance, quite unlike that of their later-evolving liverwort cousins. They grow with sickly yellow subterranean axes that creep through the soil and leaf litter on the forest floor. Occasionally an erect shoot pushes upwards, reaching only a few

  millimetres or so above the soil surface, wrapped in a series of small, flat, glossy green leaf-like structures.32 Below ground, the creeping axes are naked, devoid

  of rhizoids that typically anchor later-evolving liverwort lineages into the soil.

  Instead, the outer cells of the axis secrete a strange mucous substance. The copious mucilage might act as a water reservoir for the plants to draw on during dry

  Ancestr Al AlliAnces a 135

  Figure 20 Haplomitrium gibbsiae (upper picture, between arrows) and Treubia pygmaeaone (lower picture), two of the most primitive liverworts on Earth.

  136 a Ancestr Al AlliAnces

  spells, or perhaps provision the symbiotic fungus with carbohydrates to sus-

  tain it along with the sugars produced by the plant. Inside the tissues of the

  subterranean stems are distinctive swollen finger-like projections and filamentous coils of pea truffle fungi that extend within cells. Similar fungal projections

  and coils penetrate the cells of the mucilage-coated underground stems of the

  other ancient antipodean liverwort lineage, classified to the primordial genus

  Treubia.33

  Back in the UK, and cultured under environmental conditions mimicking

  those in New Zealand, our collected specimens of Haplomitrium and Treubia thrived, making it possible for us to conduct experiments designed to answer

  key questions about the nature of their alliance with pea truffle fungi.34 We fed the plants radioactive carbon dioxide that was converted subsequently into

  radioactively labelled organic compounds via photosynthesis. This meant we

  could determine whether the plants delivered these sugars to the fungi. Fungi

  were, in turn, supplied with specific sources of phosphorus and organic nitrogen, each labelled with radioactive forms to allow traceability. From this work, we

  discovered the fungi were adept at capturing and transferring nutrients from

  soil to their host liverworts. When we conducted the experiments under the

  simulated carbon dioxide-rich atmosphere of the Palaeozoic, the fungi gained

  greater supplies of carbon from their hosts. Could this be a clue to explaining

  why they switched from a saprotrophic lifestyle, involving feeding on dead tis-

  sue, to a mutualistic relationship with living plants? By forming associations

  with plants, they could obtain abundant and reliable supplies of photosynthate

  in return for assisting the uptake by the plant of growth-limiting nutrients.

  Complicated experiments like these ultimately provided evidence that

  Haplomitriopsida liverworts and pea truffle fungi are engaged in a mutually

  beneficial symbiosis.

  Still, mycologists are a difficult bunch to satisfy and demand a high burden of

  proof if you want to demonstrate that the two groups of organisms function in a

  symbiosis. For them, a prerequisite to establishing the functional basis of any relationship between a fungus (and indeed any microorganism) and its host plant is

  that the putative partners be grown separately, then brought together to deter-

  mine whether the naturally occurring symptoms and structures of the association

  are reproduced. This procedure is one of Koch’s Postulates, after the German

  Ancestr Al AlliAnces a 137

  physician Robert Koch (1843–1910), who first proposed what has become central

  to the diagnosis of pathogenesis and the aetiology of disease. Working with

  specimens collected from New Zealand, we successfully cultured Haplomitrium

  and Treubia and the pea truffle fungi separately in sterile petri dishes laced with sugary agar-gel. Without their fungal associates, liverworts failed to develop

  normal leafless axes and failed to produce copious mucilage, suggesting fungal

  partners induce this feature of their biology. Without their plant partners providing them with a supply of photosynthate energy, the fungi grow miserably (if at all)

  relying on their saprotrophic capabilities to obtain energy. Re-introducing

  them to each other induces substantial anatomical changes in the host plants.

  Liverworts in combination with their fungal partners reproduced the typical

  symptoms of colonization seen in the specimens collected from field sites on

  South Island.35

  How might these findings from laboratory experiments fit in with remote

  events back in the Ordovician? A curious feature of the biology of pea truffle

  fungi is that some of them can live a dual lifestyle, functioning either as free-living saprotrophs or as symbionts teamed up with plants. Given this versatility, it is

  possible these fungi lived on the Ordovician landscape long before terrestrial

  plants arrived on the scene. Functioning in saprotrophic mode, they could have

  made a living by obtaining nutrients from decomposing organic matter formed

  from algal and cyanobacteria remains on the margins of shallow freshwater

  ponds and lagoons. When land plants appeared, the fungi could have seized the

  ecological opportunity to establish a new symbiotic alliance with a wonderful,

  new, reliable energy source for doing what organisms like to do: grow and repro-

  duce. Plant life benefited from the arrangement by securing access to soil nutri-

  ents from the fungi, including inorganic nutrients like phosphorus and organic

  nitrogen from decaying litter and detritus. We can see now that bringing the pea

  truffle fungi into the evolutionary picture seems to offer a solution to the unlikely coincidence of two distinct groups of organisms (plants and fungi) making land

  more or less simultaneously.

  Notice that only an experimental approach to fundamental questions raised by

  the fossil record can establish essential functional evidence supporting or refuting the long-standing conjecture of the presumed symbiotic
origins of our land floras.

  Adopting such experimental approaches to unpicking symbioses has its roots in

  138 a Ancestr Al AlliAnces

  the philosophy of Anton de Bary (1831–1888). De Bary is widely credited with coining the term symbiosis, and strongly urged his fellow mycologists to undertake

  experiments to determine the functional nature of symbioses between plants and

  fungi. Some might argue that experiments like the ones we have been discussing

  for early land-plant lineages necessarily employ contemporary fungal strains, and might question the wisdom of drawing inferences from such studies. Genomics

  provides the counter-argument. Fungal genes controlling functions of the sym-

  biosis are ancient and highly conserved, retaining similarities to fungi in their far more distant ancestors from which they diverged hundreds of millions of years

  ago. Similarly, the genetic toolbox plants required for the establishment and functioning of the symbiosis dates back hundreds of millions of years to a common

  mycorrhizal ancestor.36 On this basis at least, experimental findings offer insights into the past, the product of a marriage between contemporary ecology and palaeontology.

  Not surprisingly, a resurgence of interest in fossil fungi followed, with palaeontologists re-examining thousands of thin section slides cut from the Rhynie

  Chert fossils in search of evidence for early symbiotic encounters.37 One reward

  of this effort has been the discovery of fungal symbioses in the early Devonian

  fossil plant Horneophyton lignieri, which has cells with features suggestive of both arbuscular mycorrhizal and pea truffle fungi.38 The evidence is far from clear-cut, but the intriguing suggestion is that the plant benefited from a dual fungal symbiosis. This suggestion fits with recent studies showing examples of liverworts,

  hornworts, lycophytes, and ferns entering into dual fungal partnerships, although the functional purpose of such a strategy remains unclear. Did early land-plant

  lineages, like their living descendants, associate with a wider and more versatile suite of fungal symbionts than previously assumed, interacting in ways still to be discovered?

  A case in point arose from the genetic fingerprinting of the fungal partners of

  surviving early vascular land-plant lineages, which offered a glimpse into what