| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
,
* School of Biological Sciences, University of Wales Bangor, Bangor, UK;
Florida Museum of Natural History, University of Florida, Gainesville, Florida, USA; and
Zoology Department, University of Florida, Gainesville, Florida, USA
1Correspondence: School of Biological Sciences, University of Wales Bangor, Bangor LL57 2UW, UK. E-mail: h.braig{at}bangor.ac.uk
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONREFERENCES |
|---|
-Proteobacteria. Its closest sequenced relative is Arsenophonus nasoniae, from which it differs by more than 10%. A. nasoniae is a male-killing endosymbiont of jewel wasps. Using microdissection and multiphoton confocal microscopy, we show the remarkable interaction of this bacterium with its host. This endosymbiont is unique because it occupies sequentially four different mycetomes during the development of its host, undergoes three cycles of proliferation, changes in length from 2–4 µm to more than 100 µm, and has two extracellular migrations, during one of which the endosymbionts have to outrun its host’s immune cells. The host and its symbiont have evolved one of the most complex interactions: two provisional or transitory mycetomes, a main mycetome and a paired filial mycetome. Despite the close relatedness of body and head lice, differences are present in the mycetomic provisioning and the immunological response.—Perotti, M. A., Allen, J. M., Reed, D. L., Braig, H. R. Host-symbiont interactions of the primary endosymbiont of human head and body lice.
Key Words: host-pathogen interaction • host-symbiont evolution • organ-forming bacteria • immune evasion • bacteriotome
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONREFERENCES |
|---|
, so have human lice and their primary endosymbiont. The first finding of a mycetome (bacteriome, bacteriotome) dates back to 1664 when Robert Hooke studied the physiology of the human louse (2)
. Mycetomes are organs of mainly invertebrates that host obligatory bacterial or fungal symbionts. The lice that Hooke worked on were probably body lice, Pediculus h. humanus. Humans also carry head lice, P. h. capitis and pubic lice, Phthirus pubis. Lacking any precedence, Hooke still considered the organ to be the liver. But soon afterward, in 1669 Jan Swammerdam started working on lice and gave a detailed description of granules representing the endosymbionts in the peritoneal gland, which later became known as the stomach disc, the main mycetome of the human louse (3)
. Since then, many attempts have been made over the years to isolate, culture or identify this first mycetomic bacterium; however, it always remained elusive. Many of these experiments have in quite unique ways demonstrated the obligate nature of the symbiont and its mycetome. Microsurgery and the centrifugation of eggs and living embryos for several days to displace the symbiont or the mycetome revealed the dependence of the host on this bacterium for reproduction (4
5
6
7)
. The identification of the endosymbiont allows us to investigate the evolution of an extraordinary variety of unique and intricate host-endosymbiont interactions and adaptations. |
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONREFERENCES |
|---|
, and from Lice Solutions in West Palm Beach, FL. Head lice for host-symbiont interactions were from La Rioja and also from donations of volunteers in Mar del Plata, province of Buenos Aires, Argentina and London, United Kingdom. One hundred and forty head lice were studied by fluroescence in situ hybridization (FISH). The body lice were from the Culpepper strain maintained in Baltimore, Maryland, USA.
Microdissections
All lice used were originally fixed in 96% ethanol. Head and body lice were dissected for light and fluorescence microscopy. Before dissecting, lice were washed for 1 min in washing buffer (0.7% sodium chloride, 0.05% Triton X-100) and 20 s in an ultrasonic bath and then rinsed in sterile distilled water. The lice were rehydrated in PBTA (phosphate buffer with Triton X-100 plus sodium azide) in three consecutive steps: 30, 60, and 100% PBTA. The dissections were performed under a Leica stereoscope with a magnification of x100 using tungsten tips and special carbon steel blades. A maximum dissecting resolution of 10–15 µm was obtained. The dissected bodies were fixed inside 1.5-ml tubes for further preparations.
Primers
The fragment of the 16S gene of around 1,515 bp was amplified from dissected, clean mycetomes of 2nd instar nymphs of P. h. capitis at UWB and from the ovaries of adult females at FLMNH. Universal 16S rRNA primers for bacteria, 47F-1387R (9)
, were used for polymerase chain reaction (PCR), the products obtained were ligated into QIAGEN pDrive cloning vector, and 15 positive colonies per ligation were sequenced in both directions. From the dissected mycetomes (University of Wales Bangor), 90% of the clones derived from PCR contained the new bacterium, the remaining 10% only empty vector; no other bacterial sequences were obtained from the cleaned stomach discs. These sequences were identical to the sequences obtained at the FLNMH from the adult female ovaries.
Phylogenetic interference
Individual louse specimens were sequenced after dissection. Per specimen, 10 clones were sequenced in both directions
Sequences obtained from GenBank: Arsenophonus insecticola DQ115536, A. nasoniae ARSRR16S, Drosophila paulistorum U20277, Providencia rettgeri AM040492, P. stuartii AY803746, Raoultella planticola AF181574, Kluyvera ascorbata AJ627203, Citrobacter freundii AF025365, Enterobacter hormaechei AJ853889, Escherichia coli NC_000913, Tatumella ptyseos AJ233437, Cinara cedri AY620432, Serratia symbiotica AY822594, S. odorifera AJ233432, S. entomophila AJ233427, Pectobacterium chrysanthemi AF373199, Pantoea agglomerans AY924375, Brenneria quercina AJ233416, Myrmeleon monilis DQ068837, Sitophilus oryzae AF548137, Sodalis glossinidus AY861703, Blochmannia floridanus BX248586, Wigglesworthia glossinida NC_004344, Baumannia cicadellinicola AF465797, Hamiltonella defensa AF293622, Regiella insecticola AF293627, Buchnera aphidicola AY620431, and Wolbachia pipientis AY833061.
Bayesian interference was analyzed under the GTR + I + 
model using version 3.1 of MrBayes. The analysis was run for 1 million generations and samples taken every 100 generations. The first 2,500 trees were discarded.
Light- and fluorescent microscopy
Light microscopy was done on a Leitz microscope; fluorescent microscopy was done on a Nikon Eclipse CFI60. Confocal analyses were made with a Zeiss LSM510 microscope with Coherent Multiphoton laser. Length measurements were done with the help of the software Zeiss LSM Image Examiner Version 2.80.1123. Counting was done manually in selected confocal planes.
FISH
Fluorescence in situ hybridization was performed using an endosymbiont-specific probe.
Fixation
Specimens were washed as for dissections and selected under the microscope on a slide with a drop of water and immediately fixed. The fixative used was ethanol-glacial acetic acid (3:1) for 2 h. They were kept overnight in xylene:ethanol (1:1) at 4°C, were then transferred to xylene:ethanol (1:2) for 30 min and to ethanol for 30 min, washed with –20°C 80% acetone for 20 min and dehydrated in ethanol.
Rehydration
Changes at 20-min intervals of ethanol:PBTA were performed at the following ratios: 2:1, 1:1, and 1:2 and finally only PBTA for 30 min. Everything was kept in pure PBTA at 4°C until use. For fluorescence with 4',6'-diamidino-2-phenylidole (DAPI), whole specimens were incubated overnight and then mounted.
Hybridization
Before hybridization, PBTA samples were incubated with PBTA-HB 1:1 (Hybridization buffer: TRIS-HCl 0.02M, sodium chloride 0.09 M, SDS 0.01%, formamide 35%, Denhardt’s solution 15%) for 20 min followed by only hybridization buffer (around 500 µl/tube), sonicated in an ultrasonic bath for 20 s, and then probes and corresponding helpers added (final concentration of 100 pmol each). Samples were incubated at 47°C in total darkness for 16 h, then washed for 1 h with the same HB without probe/helpers at 47°C, then changed to HB:PBTA, 1:1, at room temperature and lastly to PBTA. Samples were mounted with PBTA:glycerol mounting medium.
Probe
The endosymbiont-specific probe 2-SD-Cy5 5'-Cy5-GAG ATT GTT GCC TAG GTG-3' and the 3 helpers, Helper 1-SD 5'-ACC TCA CCT ACT AGC TAA TCT C-3', Helper 2-SD 5'-GTA TGG GCT CAT CTA AAG-3', and Helper 3-SD 5'-TTT AGG TAG ATY CCC ATA T-3' were based on the endosymbiont sequence obtained from human head louse endosymbiont. (The sequences of the endosymbionts have been deposited in GenBank (accession numbers EF110569–EF110572). No-probe and competition suppression controls using excess unlabeled probe were performed.
|
RESULTS AND DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONREFERENCES |
|---|
-Proteobacteria. Its closest relatives are species in the genus Arsenophonus (Fig. 1
). The new bacterium shows the highest sequence identity with Arsenophonus nasoniae from which it diverges in the 16S sequence by more than 10%. A. nasonia is a reproductive parasite that kills selectively male embryos of its host, the parasitoid and jewel wasp, Nasonia vitripennis (Hymenoptera) (10)
. However, male-killing bacteria are found in a wide range of bacterial lineages (11
, 12)
. Arsenophonus species are widespread as secondary endosymbionts of whiteflies, psyllids and aphids (Hemiptera) where they undergo extensive horizontal transfer on an evolutionary timescale (13
14
15
16
17)
. The assassin bug and Chagas vector Triatoma infectans carries an intracellular Arsenophonus species that can be cultured axenically (18)
. Arsenophonus also have recently been detected in bat and louse flies (Diptera) (19
, 20)
. A related bacterium is reported from the dog tick, Dermacentor variabilis (21)
. Arsenophonus does not show any obligatory relationship with their hosts. The large divergence between Arsenophonus and the new bacterium in lice suggests an independent evolutionary history. While many bacterial isolates and sequences have been obtained over time from human and animal lice, all studies have failed to prove the identity with the primary endosymbiont. Using a highly specific fluorescent probe, we have localized the new bacterial lineage in its various mycetomes and traced the path of the primary symbiont during the host’s development. The only other bacterium present was the secondary symbiont, Wolbachia pipientis, which has been described previously (8
, 22)
.
|
Sequential host adaptations and organ formations during development
A schematic overview of the fate of the primary endosymbiont during louse development is given in Fig. 2
.
|
At the beginning of oogenesis, the symbionts fill the posterior pole of the oocyte. The bacteria are free in the periplasm separated by a membrane from the yolk of the oocyte. A single ovial basket mycetome is formed by diploid cleavage cells surrounding the endosymbionts at the beginning of the invagination of the germ band (Fig. 3
). This is the first transport mycetome. The bacteria remain extracellular in the original periplasm while inside the basket mycetome. The invagination pushes the basket mycetome to the anterior pole and transports the endosymbionts into the interior of the embryo. The complete movement takes less than 12 hours. Primordial germ cells are moved from the posterior to the anterior pole by a similar process. The concentration of the endosymbionts at the posterior end of the egg forms an obstruction for the nuclei migrating to the egg surface and leads to a temporary interruption of the blastoderm development.
|
At the anterior pole, a single primary embryonic mycetome is formed in body lice. It is assembled from 12 to 14 embryonic, yolk-free premycetocytes, each with four nuclei that are 16-ploid. The premycetocytes are free of bacteria. The primary embryonic mycetome constitutes the second transport mycetome. The cells of the embryonic mycetome connect through a monaster with the interior of the basket mycetome and the endosymbionts end up in vacuoles of the new mycetomic cells. The bacteria are now intracellular. The basket mycetome disintegrates. The new embryonic mycetome is situated above the germ tube and prevents it from closing. The mycetome transports the endosymbionts inside the embryonal gut lumen to a place in the midgut, where a single definitive mycetome, the stomach disc, will be formed.
Head lice construct a similar embryonic mycetome. However, at least four additional large individual, yolk-free mycetocytes that are heavily infected are dispersed over the embryo (Fig. 4
). The bacteria in these individual yolk-free mycetocytes may account for one-third of all symbionts. All individual yolk-free mycetocytes hold four nuclei as well. The yolk-reach area of head lice embryo, where later the midgut will form also contains large amounts of extracellular bacteria (Fig. 4
, top insert). The highly infected individual mycetocytes and the free symbionts have not been observed before.
|
The embryonic mycetome pushes the gut epithelium to form a sac-like abscission of the intestinal epithelium. Cells with 16-ploid nuclei originating from a mesodermal plate form the envelope of the main mycetome. Diploid cells originating from the same mesodermal plate migrate into the center of the new mycetome. Cells, which are derived from the endodermal intestinal saclet form a syncytium that will house the endosymbionts. The nuclei are destroyed. More diploid cells originating from the same mesodermal plate are installed between the envelope cells and the future chambers of the endosymbionts so that the syncytial anucleate mycetocyte is surrounded and transversed by striated muscle bands. It is the only known mycetome surrounded by muscle cells. The mycetome is capable of muscular contraction. From the intestinal wall, septa form 10–16 radially arranged anucleate chambers. The endosymbionts of the embryonic mycetome form a single huge vacuole in each chamber (Fig. 5
; Supplemental Movie 1. In the center between the chambers, a slightly vacuolized cell mass remains, which does not contain any endosymbionts. The envelope cells become syncytial and more fibrous. The final stomach disc has no connection with the gut lumen. The embryonic mycetome disintegrates. The two provisional transport mycetomes and the definitive mycetome develop before the first nymphal instar hatches.
|
In late second and early third instar female nymphs, all endosymbionts leave the stomach disc and glide on the ventral side of the stomach to the lateral oviducts where the bacteria penetrate and enter individual vacuoles of new mycetocytes (Fig. 6
). Around a thousand mycetocytes originating from the oviduct coalesce into paired ovarial ampullae, which are the filial mycetomes with the sole function of endosymbiont transmission (Fig. 7
). The ampullae are delimitated by a tunica to the body cavity. Basal cells originating anteriorly from the egg follicles are free of bacteria. The mycetocytes of the ampullae are recruited from the oviduct epithelium. Cells coming posteriorly from the oviduct form an inner lining of the ampullae. The stomach disc of males simply degenerates over time with all endosymbionts still inside.
|
|
The colonization of the oocyte occurs through hydropyles of the egg shell (Fig. 3
, bacteria not shown). Hydropyles are parallel channels of 2–3 µm diameter that penetrate the whole thickness of the chorion (23)
. They are specific for Phthiraptera. The function remained elusive, although Hinton suggested that they might serve in the uptake of water (24)
. We propose that phthirapteran hydropyles represent another functional host adaptation that allows the uptake of the primary endosymbiont by the egg. We assume that with very few exceptions, practically all sucking (anopluran) and biting (mallophagan) lice species carry obligate endosymbionts and therefore have hydropyles.
The stomach disc and the ovarial ampullae are both endosymbiont-specific organs built from different cell types (6
, 25
, 26)
. The ovarial ampullae are unique to lice. During early embryonic development, the louse host provides sequentially two different transport mycetomes for the endosymbiont. Then the bacteria are housed in a mycetome to fulfill their obligate physiological role. Transovarial transmission to the next generation is secured by the host provision of a transmission mycetome. The identification of the primary louse symbiont will facilitate the study of the evolution in the interaction of host and symbiont in the development of mycetomes (27
28
29)
.
Changes of the symbiont
Three cycles of proliferation are known, one early during the formation of the stomach disc. Before leaving the stomach disc, another increase in numbers is observed, and yet a third after arriving in the ampullae. It is here in the ampullae where replication leads to a maximum of 
5,000 symbionts. The low numbers of endosymbionts in the primary mycetome might be one of the reasons why this endosymbiont has escaped identification for so long. Indeed, a secondary endosymbiont, the reproductive parasite Wolbachia pipientis, has been identified previously in lice (8
, 22)
.
The symbionts start out as 2- to 4-µm-long rodlets in the periplasm of the oocyte. During development, these rodlets become more and more filamentous. In the stomach disc of 1st instar nymphs, the filaments start to measure 2 x 25–30 µm. In early 2nd instar nymphs, the filaments reach 50 µm in length. In female late 2nd and early 3rd instar nymphs before leaving the stomach disc, the lengths already decreases, whereas in male 3rd instar nymphs, filaments continue to increase to reach a length of 100 µm. In male adults, some filaments surpass the 100 µm before degeneration of the bacterial mass makes measurements impossible. In many male specimens, degeneration of the symbionts already starts during 3rd instar. The biggest increase in size occurs in the stomach disc, while the biggest decrease occurs during entering the periplasm of the oocyte. The maximum size difference for this new endosymbiont is 20-fold in females and 50-fold in males.
Historically, changes in shape have been characteristic for Rickettsia. The obligate and mycetomic Rickettsia of booklice (Psocoptera) change in length roughly twofold during development in their host (30)
. Pronounced changes in structure have been reported for the a and t symbionts of the leafhopper Euscelis plebejus (31
, 32)
. A tenfold difference in length has been observed for the symbiont of the sawtoothed grain beetle Oryzaephilus surinamensis. The bacteria achieve their maximum length in pupae with 60–70 µm, while the bacteria are only 3–6 µm in length in the mycetome of the adult female just before transovarial transmission (33)
. During transovarial transmission most endosymbionts adopt the smallest size and most regular shape than at other stages of the life cycle.
Alternations between intracellular and extracellular location
The endosymbionts are always surrounded by a host vacuolar membrane while intracellular. At the beginning, the bacteria are extracellular in the periplasm and the basket mycetome. Later, they become extracellular again during migration from the stomach disc to the ovarial ampullae and during migration from the ovarial ampullae to the individual oocytes. In head lice, the bacteria are also extracellular in the yolk during early development. The bacteria in the chambers of the stomach disc mycetome might be considered extracellular as well. The migration in the body cavity from the stomach disc to the tunica of the ovaries is of very short duration. The endosymbionts literally have to outrun the patrolling host immune cells in head lice. During migration, hemocytes engulf the endosymbionts (Fig. 6)
and concentrate on the surface of the tunica. The hemocytes try to follow the penetrating endosymbionts, which leads to a picture where, for a moment, each hole in the tunica is covered by an immune cell. After the bacteria have successfully crossed over to the inside of the ovaries, the hemocytes detach and leave behind the empty holes. The tunica is eventually restored by underlying basal cells. This might look like a stage representing little adaptation of the host to its obligatory symbiont, yet at the same time, the tunica, which normally is smooth and impenetrable for the bacteria, roughens and develops a mesh to help orient the bacteria perpendicular to the surface (25
, 34)
. This happens just before the bacteria start to migrate and disappears again afterward. The attack of the primary endosymbiont by the host’s immune system suggests that for host-symbiont interaction, morphological or structural adaptations might evolve much easier or faster than functional adaptations in the immunological defense system.
The host-symbiont interaction in human lice suggests that the immune system might be the most limiting factor in the evolution of these associations. Lice have shown that the insect host can provide the most elaborate morphological or structural adaptations. Immunological adaptations of the host might be difficult to achieve because of the lack of a specific immune system in insects and invertebrates. Many host provisions to their symbionts might have been shaped by evading an unadaptable immune system. Interestingly, organisms with an adaptable, specific immune system do not have comparable symbiotic associations.
Response of hemocytes in head and body lice to symbiont migration
The migration of the symbiont in head and body lice are comparable. In head lice, the hemocytes attack the symbionts almost instantly when they leave the stomach disc mycetome. Many hemocytes are seen engulfing the symbionts (Fig. 6)
. In body lice, the hemocyte response is delayed by several hours. Hemocytes with engulfed symbionts were rarely observed. Hemocytes started covering the holes of the tunica after most of the bacteria had already crossed over into the oviduct. Human body lice are important vectors for Borrelia recurrentis (louse-borne relapsing fever), Bartonella quintana ((urban) trench fever), and Rickettsia prowazekii (epidemic typhus). Experimentally infected body lice are also capable of transmitting R. rickettsii (Rocky Mountain spotted fever) and R. conorii (Mediterranean spotted fever, Indian tick typhus) to rabbits (35)
. On the other hand, head lice have never been implicated as vectors of any human disease. It is tempting to speculate that this marked difference in vector capacity might be related to differences in the host’s immune response to bacteria.
Head and body lice
Human body lice evolved from head lice around 100,000 yr ago (36
, 37)
. This small time period already shows emerging differences in the provision of mycetomic structures, especially in the embryonic mycetome and mycetocytes. It also shows functional differences in the hemocyte response to the endosymbionts.
Deciphering the evolution of body and head lice might provide far-reaching insights not only on lice themselves but also on the recent evolutionary history of humans. The last three years have seen a steadily increasing number of studies on the genetic differences of human body and head lice (36
, 38
39
40
41
42
43)
. Already Levene and Dobzhansky have speculated that the differences between human head and body lice might be explainable by differences in the proportion of two hypothetical endosymbionts (44)
. Two endosymbionts of human lice have now finally been identified.
Human body lice depend on clothing. The divergence of body and head lice provides an estimate for a minimum age of human body cover for which there is no fossil record. Kittler et al. propose a minimum of a 107,000 yr for the origin of human clothing (36
, 37)
. However, human head lice are characterized by two lineages that have segregated long before head and body lice diverged from each other (39
, 42
, 43)
. The lineages also show differences in their cosmopolitan distribution—a worldwide clade and another that is restricted to parts of the New World, Europe, and Australia. The 16S sequence of the primary endosymbiont is too conserved to resolve population differences of their host. Further work will require the identification of a phylogenetically reliable, fast evolving gene in this new bacterium. Reed et al. proposed that the two lineages of head lice separated around 1.18 million years ago at the same time when Homo erectus split off the hominid lineage that eventually led to the present day H. sapiens. One of the lineages underwent a bottleneck along with its emerging host H. sapiens circa 100,000 yr ago. A host switch during direct physical contact between H. sapiens and H. erectus might explain both head lice on modern humans (42)
. The obligate symbiont of human lice will underpin these studies. But not only living but even dead bacteria in lice are now being investigated to deduce the diseases transmitted to soldiers of Napoleon’s Grand Army, an example of the growing contribution of insects to the new discipline of palaeomicrobiology (45
, 46)
. An obligate symbiont will be equally important for the determination of the origin of parasitism in lice. Phylogenetic reconstruction now suggests that parasitism in lice originated at least twice and that free-living booklice (Liposcelidae, Psocoptera) may have derived from parasitic lice (47
48
49
50)
.
Although human lice have been one of the first insects to be studied microscopically, basic biological characteristics like obligate endosymbionts or chromosome loss in male lice through transmission ratio distortion are just being discovered and described (51)
.
On the applied side, lice infestation in school children in the developed world is still on the rise, while resistance against topical insecticides has reached a point where now oral insecticides such as ivermectin are increasingly used against both, head and body lice emphasizing the strong need for new targets (52
, 53)
.
|
ACKNOWLEDGMENTS |
|---|
Received for publication September 6, 2006. Accepted for publication November 28, 2006.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTS AND DISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  T. Fukatsu, T. Hosokawa, R. Koga, N. Nikoh, T. Kato, S.-i. Hayama, H. Takefushi, and I. Tanaka Intestinal Endocellular Symbiotic Bacterium of the Macaque Louse Pedicinus obtusus: Distinct Endosymbiont Origins in Anthropoid Primate Lice and the Old World Monkey Louse Appl. Envir. Microbiol., June 1, 2009; 75(11): 3796 - 3799. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. M. Allen, D. L. Reed, M. A. Perotti, and H. R. Braig Evolutionary Relationships of "Candidatus Riesia spp.," Endosymbiotic Enterobacteriaceae Living within Hematophagous Primate Lice Appl. Envir. Microbiol., March 1, 2007; 73(5): 1659 - 1664. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
-Proteobacteria, as an outgroup. Pediculus humanus and Drosophila paulistorum are the hosts for bacteria that have yet to be named. The probability of the nodes is given for all bifurcations.
