Succession of the microbiota in the gut of reproductives of Macrotermes subhyalinus (Termitidae) at colony foundation gives insights into symbionts transmission

Abstract

Termites have co-evolved with a complex gut microbiota consisting mostly of exclusive resident taxa, but key forces sustaining this exclusive partnership are still poorly understood. The potential for primary reproductives to vertically transmit their gut microbiota (mycobiome and bacteriome) to offspring was investigated using colony foundations from field-derived swarming alates of Macrotermes subhyalinus. Metabarcoding based on the fungal internal transcribed spacer (ITS) region and the bacterial 16S rRNA gene was used to characterize the reproductives mycobiome and bacteriome over the colony foundation time. The mycobiome of swarming alates differed from that of workers of Macrotermitinae and changed randomly within and between sampling time points, highlighting no close link with the gut habitat. The fungal ectosymbiont Termitomyces was lost early from the gut of reproductives, confirming the absence of vertical transmission to offspring. Unlike fungi, the bacteriome of alates mirrored that of workers of Macroterminae. Key genera and core OTUs inherited from the mother colony mostly persisted in the gut of reproductive until the emergence of workers, enabling their vertical transmission and explaining why they were found in offspring workers. These findings demonstrate that the parental transmission may greatly contribute to the maintenance of the bacteriome and its co-evolution with termite hosts at short time scales.

Full text

Frontiers in Ecology and Evolution 01 frontiersin.org
Succession of the microbiota in
the gut of reproductives of
Macrotermes subhyalinus
(Termitidae) at colony
foundation gives insights into
symbionts transmission
MichelDiouf
1
*, VincentHervé
2, SophieFréchault
1,
JosieLambourdière
3,4, AbdoulayeBaïlaNdiaye
5,
EdouardMiambi
1, AméliaBourceret
3, My DungJusselme
1,
Marc-AndréSelosse
3,6 and CorinneRouland-Lefèvre
1
1 Département ECOEVO, Institut d'Ecologie et des Sciences de l'Environnement de Paris (iEES,
Paris), Université Paris Est Créteil, Créteil Cedex, France, 2 INRAE, AgroParisTech, UMR SayFood,
Université Paris-Saclay, Palaiseau, France, 3 UMR 7205-ISYEB (Institut de Systématique, Evolution,
Biodiversité, CNRS, MNHN, UPMC, EPHE), Muséum National d’Histoire Naturelle, Sorbonne
Universités, Paris, France, 4 UMR BOREA (MNHN, CNRS-7208, IRD-207, Sorbonne Université, UCN,
UA), Université des Antilles, Pointe-à-Pitre, France, 5 Laboratoire de Zoologie des Invertébrés
Terrestres, Institut Fondamental d’Afrique Noire Cheikh A. Diop (IFAN Ch. Anta Diop), Université
Cheikh Anta Diop de Dakar (UCAD), Dakar, Sénégal, 6 Department of Plant Taxonomy and Nature
Conservation, University of Gdańsk, Gdańsk, Poland
Termites have co-evolved with a complex gut microbiota consisting mostly of
exclusive resident taxa, but key forces sustaining this exclusive partnership are still
poorly understood. The potential for primary reproductives to vertically transmit
their gut microbiota (mycobiome and bacteriome) to ospring was investigated
using colony foundations from field-derived swarming alates of Macrotermes
subhyalinus. Metabarcoding based on the fungal internal transcribed spacer (ITS)
region and the bacterial 16S rRNA gene was used to characterize the reproductives
mycobiome and bacteriome over the colony foundation time. The mycobiome of
swarming alates diered from that of workers of Macrotermitinae and changed
randomly within and between sampling time points, highlighting no close link with
the gut habitat. The fungal ectosymbiont Termitomyces was lost early from the
gut of reproductives, confirming the absence of vertical transmission to ospring.
Unlike fungi, the bacteriome of alates mirrored that of workers of Macroterminae.
Key genera and core OTUs inherited from the mother colony mostly persisted
in the gut of reproductive until the emergence of workers, enabling their vertical
transmission and explaining why they were found in ospring workers. These
findings demonstrate that the parental transmission may greatly contribute to the
maintenance of the bacteriome and its co-evolution with termite hosts at short
time scales.
KEYWORDS
termite reproductives, Macrotermitinae, gut microbiota, Termitomyces,
transmission, colony foundation
TYPE Original Research
PUBLISHED 09 January 2023
DOI 10.3389/fevo.2022.1055382
OPEN ACCESS
EDITED BY
Aram Mikaelyan,
North Carolina State University,
UnitedStates
REVIEWED BY
Thomas Chouvenc,
University of Florida,
UnitedStates
Jane Oja,
University of Tartu,
Estonia
*CORRESPONDENCE
Michel Diouf
michel.diouf@u-pec.fr
SPECIALTY SECTION
This article was submitted to
Social Evolution,
a section of the journal
Frontiers in Ecology and Evolution
RECEIVED 27 September 2022
ACCEPTED 20 December 2022
PUBLISHED 09 January 2023
CITATION
Diouf M, Hervé V, Fréchault S,
Lambourdière J, Ndiaye AB, Miambi E,
Bourceret A, Jusselme MD, Selosse M-A
and Rouland-Lefèvre C (2023) Succession
of the microbiota in the gut of
reproductives of Macrotermes subhyalinus
(Termitidae) at colony foundation gives
insights into symbionts transmission.
Front. Ecol. Evol. 10:1055382.
doi: 10.3389/fevo.2022.1055382
COPYRIGHT
© 2023 Diouf, Hervé, Fréchault,
Lambourdière, Ndiaye, Miambi, Bourceret,
Jusselme, Selosse and Rouland-Lefèvre.
This is an open-access article distributed
under the terms of the Creative Commons
Attribution License (CC BY). The use,
distribution or reproduction in other
forums is permitted, provided the original
author(s) and the copyright owner(s) are
credited and that the original publication in
this journal is cited, in accordance with
accepted academic practice. No use,
distribution or reproduction is permitted
which does not comply with these terms.

Diouf et al. 10.3389/fevo.2022.1055382
Frontiers in Ecology and Evolution 02 frontiersin.org
Introduction
Termites have established very close mutualistic interactions
with microorganisms. e gut microbiota of termites mostly
consists of bacteria and archaea for the large family Termitidae
(also known as “higher termites”), and with the same
microorganisms plus protozoa for the paraphyletic “lower
termites.” ese symbionts crucially contribute to the exceptional
ability of termites to digest lignocellulose compounds and exploit
diverse resources, from wood to soil organic matter (Brune, 2014).
e rich literature about termite symbiosis (Bignell, 2016) has
demonstrated, in particular, the uniqueness of the gut microbiota
and its co-evolution with hosts, likely facilitated by the social
organization of termites, as in social bees (Kwong etal. 2017; Su
etal. 2021).
Previous studies have considered how micro-environmental
gut conditions and diet shape the gut microbiota (Mikaelyan etal.,
2015, 2017). Meanwhile, other studies have emphasized the
importance of vertical inheritance, thus linking microbiota and
host phylogenies (Hongoh etal., 2005; Noda etal., 2007; Abdul
Rahman etal., 2015; Diouf etal., 2015; Michaud etal., 2020; Arora
etal., 2022). e most comprehensive study in terms of host taxa
revealed that termite microbiota is primarily governed by a mixed
mode of transmission that combines vertical transmission from
parents to ospring and frequent colony-to-colony horizontal
transfers (Bourguignon etal., 2018). e importance given to
horizontal transfers, which are one-time events, is justied by
their cumulative eect over the long evolutionary history of
termites (around 150 million years; Engel etal., 2009; Bourguignon
etal., 2014; Bucek etal., 2019). erefore, assessing the actual
transfers of symbionts at each generation should precise their
signicance and the microevolution of termite microbiota at
shorter time scales (da Costa and Poulsen, 2018).
e symbiosis in Macrotermitinae, a sub-family of Termitidae
also known as fungus-growing termites, has been widely studied,
but the transmission of the gut symbionts (at the individual level)
and fungal ectosymbionts (at the colony level) remains poorly
documented. Macrotermitinae have an ectosymbiotic relationship
with plant-degrading fungi of the genus Termitomyces
(Agaricomycotina; Rouland-Lefèvre, 2000). Each generation of
ospring workers oen acquires fungi from the environment,
resulting in frequent host switchings (Aanen etal., 2002; Diouf
and Rouland-Lefevre, 2018). From an evolutionary viewpoint,
some analogies could bedrawn between the theorized mixed
mode of transmission of the gut symbionts (Bourguignon etal.,
2018) and that of the fungi in Macrotermitinae. However, while
the spores of Termitomyces can survive in natura for long periods,
the survival of bacterial symbionts, including many anaerobic
lineages outside the digestive tract, may limit horizontal transfers.
Understanding the establishment of the gut microbiota in
fungus-growing termites during the foundation of incipient
colonies at each generation may provide valuable information on
the colony-to-ospring transmission and, thus, the persistence of
symbionts over the microevolution of these termites. e
foundation of new colonies starts with the swarming ight of
alates, which are mature reproductives with complete wings
(Figure 1). Aer forming couples, alates shed their wings
(becoming dealates), nd a suitable substratum, and begin mating
aer digging out a subterranean initial colony chamber called
copularium where they are sheltered with the brood. e rst
batch of eggs is laid between 4 and 15 days aer colony initiation,
and the rst ospring larvae and workers emerge between 21 and
46 days and between 52 and 80 days, respectively (Grassé and
Noirot, 1955; Okot-Kotber, 1981; Harit etal., 2016; Mitchell,
2020). Previous studies on reproductives of Termitidae found that
alates swarmed from the mother colonies with a complete set of
the host microbiota (Hongoh et al., 2006; Diouf etal., 2018).
However, it is still unclear whether the gut-specic lineages persist
and overlap enough with ospring development to bereliably
transferred. First, the long lag time between the initiation of
colony foundation and the emergence of ospring may be a
potential barrier to this eective transfer. Second, the feeding
behavior of reproductives during this long period is unfavorable.
Indeed, unlike reproductives of lower termites, which continue to
feed during colony foundation, those of Termitidae cease feeding
(Nutting, 1969; Han and Bordereau, 1982; Grassé, 1984) and draw
the required energy from the resorption of their fat body and of
their useless alar muscles. Given the crucial role of the diet on the
microbiota (Mikaelyan etal., 2015), fasting may result in the loss
of the initial microbiota.
Apart from alates, the microbiota of the reproductive caste of
Termitidae has only been addressed in aged kings and queens
(Poulsen etal., 2014; Otani et al., 2019). At this late stage, the
microbiota is marked by the dominance of a single or very few
major bacterial lineages, preventing any extrapolation of the role
of reproductives in the transmission and the persistence of
symbionts. In contrast to the gut microbiota, the transmission of
FIGURE1
Simplified developmental cycle of Macrotermes subhyalinus. Red
arrows indicate the developmental pathway for the reproductive
caste. The chronology of the emergence of the ospring stages
is indicated in bold from our own observations and in italic for
Macrotermes sp. from literature.

Diouf et al. 10.3389/fevo.2022.1055382
Frontiers in Ecology and Evolution 03 frontiersin.org
the fungal ectosymbiont of Macrotermitinae has received more
attention. One approach was to check if the fungal asexual spores
(conidia) ingested from the fungus comb of the mother colony by
one sex of reproductive persisted in their gut and allowed eective
inoculation of primordial fungus combs in lab-founded colonies
(Sands, 1960; Johnson, 1981; Johnson etal., 1981; Sieber, 1983). In
several Macrotermes species, the initiation of the primordial
combs occurs around 90–95 days (Grassé and Noirot, 1955;
Collins, 1977; Lepage and Darlington, 2000), once the rst
dierentiated workers can gather plant matter, which is long
enough to jeopardize the viability of fungal propagules. Recent
studies have instead addressed the issue indirectly through host-
fungus co-cladogenesis (review by Diouf and Rouland-Lefevre,
2018). Finally, except for a few species mainly from the genus
Microtermes and the species Macrotermes bellicosus, the fungal
symbionts are considered acquired horizontally from the
surrounding environment (Aanen etal., 2009).
e objective of our study was to assess the potential for
reproductives to ensure the actual transfer of their microbiota to
ospring. In this respect, we focused on the succession of
reproductives microbiota (mycobiome and bacteriome) during
the foundation of incipient colonies while paying attention to their
gut morphology as a proxy for feeding behavior. e fungus-
growing termite Macrotermes subhyalinus was used as a model to
study the succession of the digestive bacteriome and mycobiome,
with a special focus on Termitomyces which has been reported to
betransmitted horizontally in this species (Johnson etal., 1981;
Vesala etal., 2017). Deep amplicon sequencing and taxon-specic
PCR for Termitomyces were used to monitor both the mycobiome
and the bacteriome in the gut of reproductives from the swarming
ight to the emergence of rst dierentiated workers and in
the copularium.
Materials and methods
Termite sampling and colony foundation
Termites were collected near Diamniadio, located 30 km
East of Dakar (Senegal). Reproductives were captured during
the swarming flight from two distinct colonies of Macrotermes
subhyalinus (colony 1 and colony 2) and one colony of the
sympatric species Microtermes sp. that swarmed at the same
time. Alates of Microtermes sp., and specially females that are
known to carry condidia of Termitomyces for the inoculation
of the comb were considered as positive control. This
sampling time, considered as time zero is referred as “alates”
in this manuscript. For both wild colonies of M. subhyalinus,
workers were also collected to compare their bacteriome with
the bacetriome of alates. As in most of Macrotermtinae
species, the worker caste of M. subhyalinus encompassed two
distinct morphotypes with different tasks and feeding
behaviors: major workers (larger workers that develop from
male larvae) and minor workers (smaller workers that
develop from female larvae) (Badertscher etal., 1983). At the
laboratory (IRD-ISRA Research Center of Bel-Air, Dakar),
alates were allowed to form pairs. Then they broke off their
wings, and each pair was isolated and transferred into an
individual LAB1 box (60x45x50 mm, VWR International,
LLC) partially filled with moistened non-sterile soil collected
in the research center. Individuals from the colony
foundations are referred as “dealates” in the manuscript. All
colony foundations were set up using reproductives from
colony 1 of M. subhyalinus. Two hundred and thirty colonies
were set up and kept in a tropicalized termite-rearing room.
The moisture of the soil was regularly checked, and adjusted
when necessary by wetting the cotton balls laid on the soil
surface with sterile distilled water. No offspring sterile caste
(workers or soldiers) was found in colonies withdrawn from
0 to 60 days, and no foraging trace was observed in
non-processed colonies during this period. Distinct offspring
castes (workers and soldiers) appeared in almost all active
foundations between days 60 and 90, and the foraging activity
of workers was visible through foraging tunnels. From day 90,
colonies were transferred into LAB2 rearing boxes
(90 × 60 × 50 mm) more appropriate for that population size
and fed with non-sterilized wheat straws.
For molecular analyses of the microbiota, three distinct
colonies were withdrawn at 0 (alates), 10, 30, 45, 60, and 120 days.
e copularoum samples on day 120 were analyzed in parallel to
ensure that digestive symbionts defecated by reproductives in their
surroundings did not thrive.
For the measurements of the gut size, 9 colonies were
withdrawn at each sampling time point, i.e., at 0, 45, 60, and
75 days, leading to 6–9 properly dissected guts for each sex, except
for the sampling time point 60 days where fewer colonies were
considered. e sex of the collected individuals was always
checked based on the conguration of the terminal sternites of the
abdomen (Weesner, 1969). Very few false pairings, i.e., same sex
couples, occurred and were discarded from analyses.
Dissection of termite reproductives and
workers from mother colonies
Alates were dewinged, and then rinsed three times in sterile
distilled water. e surface of dealates from colony foundations
was cleaned similarly. en, under a biosafety cabinet, the
whole gut was removed aseptically using ne, sterile scissors
and placed individually into a 1.5-mL sterile microtube. As
indicated earlier, for each sex, three replicates from three
distinct colony foundations were considered at each sampling
time point for molecular analyses. For major and minor workers
that were collected at the same time as swarming alates from the
same colonies of M. subhyalinus, the whole gut of workers was
removed aseptically using ne sterile forceps. For each
morphotype, DNA extraction was performed from pools of
10 guts.

Diouf et al. 10.3389/fevo.2022.1055382
Frontiers in Ecology and Evolution 04 frontiersin.org
Determination of the size of the gut
during colony foundation
For assessing the variation in gut size over the time of colony
foundation, female and male reproductives were dissected as
described above and photographed using the Zoom Stereo
Microscope SMZ1000 of Nikon calibrated at the same scale. e
images were transferred to a connected computer via a camera
(Digital Sight, DS-U3, Nikon), and treated with the soware
NIS-Element (Basic Research, Version 4.0) from Nikon (Nikon
Instruments Inc.) to measure the length of the whole gut, the areas
of the most representative gut sections (the hindgut section and
the paunch sub-section). Wealso used the method of Shimada
etal. (2013) to measure the width of the most enlarged part of the
paunch (Supplementary Figure S1).
DNA extraction
Individual termite gut for alates and dealates, or pools of 10
guts for worker morphotypes from the mother colonies were
crushed with a sterile polypropylene pestle in a 1.5-mL microtube
containing the lysis buer of the DNA extraction kit (NucleoSpin®
DNA Insect Kit, Macherey-Nagel GmbH & Co. KG). en, the gut
homogenate was transferred to NucleoSpin® Bead tube for an
additional lysis step by bead-beating with the FastPrep®-24
homogenizer (MP Biomedicals). e subsequent DNA extraction
steps were performed following the manufacturer’s instructions.
e yield and purity of DNA were analyzed with a ND-1000
Spectrophotometer (NanoDrop products, Wilmington,
UnitedStates). e same DNA extraction protocol was applied to
the copularium samples collected at the end of the experiment
(day 120). DNA-based analyses were performed on aliquots
adjusted at 10 ng μL−1 for all samples.
Taxon-specific PCR targeting the
symbiotic fungus Termitomyces in the
gut of reproductives
We rst compared the performance of several polymerase
enzymes for PCR-amplication of the Internal Transcribed Spacer
(ITS) of the ribosomal DNA with the Termitomyces-specic primer
ITS1FT (Aanen etal., 2007) combined with ITS4R (White etal.,
1990). e endpoint PCR reactions were run in 25 μl containing
0.5 μM for each primer, according to conditions and the thermal
cycle dened by Aanen etal. (2007). e Illustra PureTaq (PCR
beads, GE Healthcare) was the most sensitive, producing detectable
amplicons at concentrations below 20 pg.μL−1 with DNA extracts
from pure mycelium of Termitomyces. Following PCR reactions, the
amplicons were checked by electrophoresis onto a TAE 0.5X buered
agarose gel (1.5% (w:v) containing the DNA staining dye GelRed
(GelRed® Nucleic Acid Gel Stain, Biotium, VWR International
LLC.). A UV trans-illuminator (GenoSmart, VWR) allowed the
visualization of the positive PCR-b ands. To check the correspondence
of the amplicons with the ITS of Termitomyces, some bands were
randomly excised and sequenced. e corresponding sequences are
referenced in GenBank under the accession numbers
MN400299-MN400306. For each sampling time point, 12 to 16
individuals of M. subhyalinus were tested for each sex. e results are
displayed as a percentage of positive essays at each time point.
Comparative analysis of the gut
bacteriome of alates and workers from
wild colonies using Illumina MiSeq
metabarcoding
DNA samples from the guts of minor and major workers,
and alates were readjusted to 3.5 ng/μL, and then puried by
Ampure XP beads (Agencourt, Beckman Coulter, UnitedStates).
PCR amplications were performed in triplicate for each sample
with the primer pair 28F/338R (Cheung etal., 2018) tagged with
combinations of nucleotides specic to each sample. ese
primers target the V1-V2 hyper variable region of the bacterial
16S rRNA gene. PCR reactions were performed in 25 μl of
mixture as following: 10 ng of DNA template, 1X incomplete
buer, 0.3% bovine serum albumin, 2 mM MgCl2, 200 mM
dNTPs, 300 nM of forward and reverse primer and 2 U of
DFS-Taq DNA polymerase (Bioron, Ludwigshafen, Germany).
e following program was used: 95°C during 5 min, then
30 cycles of 94°C for 1 min, 55°C for 1 min, 72°C for 1 min 30 s,
and to nish 10 min at 72°C. Amplicons from triplicates were
pooled, checked on agarose gels (5 μl, 2%, 100 V, 20 min), and
then puried twice by Ampure XP beads. e DNA
concentration in puried amplicons was determined, and an
equal amount of 25 ng for each sample was pooled. is nal
pool was puried again and reconcentrated twice by Ampure XP
beads before the last quantication by Qubit ® dsDNA HS Assay
Kit (Invitrogen, UnitedStates). MetaFast library preparation and
sequencing were performed on an Illumina 2 × 250 MiSeq
platform by Fasteris SA (Switzerland).
Subsequently, the 16S rRNA gene data were analyzed with
mothur version 1.39.5 (Schloss etal., 2009), as described in Hervé
and Lopez (2020). Briey, the 16S rRNA gene reads were denoised
by removing any reads having ambiguous bases, homopolymers
longer than 8 bp, and having a length inferior to 280 bp or superior
to 320 bp. en sequences were aligned to the SILVA reference
database 132 (Quast etal., 2012) and preclustered (pre.cluster,
dis = 1). Chimeras were detected and removed using UCHIME
(Edgar et al., 2011). Sequences were classied with a naïve
Bayesian classier (Wang etal., 2007) and the SILVA database 132.
Non-bacterial and unknown sequences were excluded. Sequences
were clustered into operational taxonomic units (OTUs) using the
OptiClust algorithm (Westcott and Schloss, 2017) with a 97%
sequence similarity threshold. Finally, singletons were excluded,
and each sample was rareed to the smallest library size (21,601
reads) by random subsampling (McKnight etal., 2019). e raw

Diouf et al. 10.3389/fevo.2022.1055382
Frontiers in Ecology and Evolution 05 frontiersin.org
sequence reads were deposited as Sequence Read Archive (SRA)
in GenBank under the Bioproject accession PRJNA347254 with
the BioSample accession number SAMN30890275.
Metabarcoding of reproductives’ gut
mycobiome and bacteriome over colony
foundation
Based on the preliminary results of the ITS-based detection
of Termitomyces, the sequencing of the gut mycobiome was
limited to 60 days, stage at which only eggs and larvae were
found in brood. However, even absent from the gut of
reproductives, fungal propagules especially those of
Termitomyces could bekept in their immediate environment.
e copularium samples were therefore collected at the end of
the experiment and analyzed to check this possibility. e whole
mycobiome was assessed by PCR amplifying the ITS2 region
with the primer pair ITS86/ITS4 targeting a wide range of fungal
lineages (Turenne etal., 1999; Waud etal., 2014), according to
the PCR cycling conditions described by Schneider-Maunoury
etal. (2018). Tagged primers unique for each sample were used
in a second PCR step that was performed in the same conditions.
e subsequent steps of the sequencing were those described by
Schneider-Maunoury et al. (2018), using the Ion Torrent
sequencer (Life technology, Carlsbad, UnitedStates).
For bacteria, DNA fragments spanning the V1-V2
hypervariable region of the 16S rRNA gene were amplied using
the primers 28F and 338R (Cheung etal., 2018). e following
thermal conditions were applied: 94°C for 5 min, followed by
25 cycles of 94°C for 30 s, 56°C for 30 s, 72°C for 60 s, and a nal
elongation step of 72°C for 5 min. e subsequent library
preparation and sequencing steps were the same as those for fungi.
For the processing of sequencing data, demultiplexed raw data
were rst converted from bam format to fastq format using
BEDtools version 2.25 (Quinlan and Hall, 2010). Subsequently,
ITS and 16S rRNA gene data were analyzed independently with
mothur version 1.39.5 (Schloss etal., 2009), following a similar
procedure as the one described above.
Bacterial 16S rRNA gene amplicons were denoised by
removing any reads having ambiguous bases, homopolymers
longer than 8 bp, having Phred quality score < 25, and having a
length inferior to 200 bp or superior to 320 bp. en primers were
removed, and sequences were aligned to the SILVA reference
database 132 (Quast etal., 2012) and preclustered (pre.cluster,
dis = 1). Chimeras were detected and removed using UCHIME
(Edgar et al., 2011). Sequences were classied with a naïve
Bayesian classier (Wang etal., 2007) and the SILVA database 132.
Non-bacterial and unknown sequences were excluded. Sequences
were clustered into operational taxonomic units (OTUs) using the
OptiClust algorithm (Westcott and Schloss, 2017) with a 97%
sequence similarity threshold. Finally, singletons were excluded,
and each sample was rareed to the smallest library size (2,220
reads) by random subsampling (McKnight etal., 2019).
Fungal ITS reads were denoised using the same quality
parameters described, and each sequence that passed quality
ltering was truncated to a 200-bp length aer removing primer
sequences (Brown etal., 2013) and then preclustered (pre.cluster,
dis = 1). Chimeras were detected and removed using UCHIME
(Edgar et al., 2011). Sequences were classied with a naïve
Bayesian classier (Wang etal., 2007) and the Unite database v7.1
(Nilsson etal., 2018). Non-fungal and unknown sequences were
excluded. Fungal ITS sequences were pairwise aligned to generate
a distance matrix that was used to compute OTUs using the
OptiClust algorithm (Westcott and Schloss, 2017) with a 97%
sequence similarity threshold. Finally, singletons were excluded,
and each sample was rareed to the smallest library size (1,282
reads) by random subsampling (McKnight etal., 2019).
e raw sequence reads for both microbial communities were
deposited as Sequence Read Archive (SRA) in GenBank under the
Bioproject accession number PRJNA557705.
A particular focus has been placed on the persistence of the
most representative fungal and bacterial taxa. e 40 genera
displayed were the most abundant in the entire data set. Core
OTUs were dened as any OTU that persisted in the gut samples
at all the sampling time points for fungi and at all or at least at 5
of the 6 sampling time points from days 0 to 120 for bacteria.
For the specic case of Termitomyces (Lyophyllaceae family), the
classication of the OTUs was improved by a phylogenetic approach.
is consisted in combining all the OTUs aliated to Lyophyllaceae
from our data set with all the ITS sequences of Lyophyllaceae
referenced in the Unite database. ITS sequences of the close genus
Rhodocybe (Entolomataceae) were also included as an outgroup of
Lyophyllaceae. All these sequences were aligned with MAFFT v7.427
and the L-INS-i method (Katoh and Standley, 2013). e resulting
alignment was manually curated, and Smart Model Selection (Lefort
etal., 2017) was used to determine the best model of nucleic acid
evolution (GTR + G). Subsequently, a maximum likelihood
phylogenetic tree was built with PhyML 3.0 (Guindon etal., 2010).
Branch supports were calculated using a Chi2-based parametric
approximate likelihood-ratio test (Anisimova and Gascuel, 2006).
Statistical analyses
e eect of time and sex on the measurements of the gut size
was evaluated by Kruskall-Wallis tests. Alpha diversity was
investigated using the R package hilldiv (Alberdi and Gilbert, 2019)
based on abundance-based Hill numbers (qD) that integrate
commonly used alpha-diversity indices according to the values of
q. Hence, q = 0 refers to OTUs richness, and q = 1 or 2 refers to the
converted indices of Shannon (exponential of the Shannon index)
and Gini-Simpson (the multiplicative inverse of the Gini-Simpson
index), respectively (Chao etal., 2016). e dierences in alpha-
diversity indices were analyzed by One-way ANOVA with the
Fisher’s LSD Post Hoc Test, at p < 0.05, aer ascertaining the normal
distribution of variables. e changes in microbial community
structure were investigated by analysis of similarities (ANOSIM),

Diouf et al. 10.3389/fevo.2022.1055382
Frontiers in Ecology and Evolution 06 frontiersin.org
implemented in mothur (Schloss, 2008), using Bray-Curtis distances
and 10,000 iterations. Principal coordinate analyses (PCoA) were
constructed based on Bray-Curtis distances with the phyloseq R
package (McMurdie and Holmes, 2013). e eect of sampling time
points on the composition of the bacteriome and mycobiome was
tested by non-parametric permutational multivariate analysis of
variance (PERMANOVA), as implemented in the vegan function
adonis with 10,000 permutations (Oksanen etal., 2007). All the
other analyses were performed with R v3.4.4, and plots were
generated with ggplot2 (Wickham, 2016). To test the relationship
between matrices of the bacteriome and the mycobiome, a Mantel
test was performed using the ecodist package (Goslee and Urban,
2007), with Bray–Curtis dissimilarity matrices, Spearman’s rank
correlation coecient, and 10,000 random permutations.
Results
Comparison of the alates bacteriome
with the bacteriome parental colonies
e evaluation of the bacteriome of alates and workers directly
from wild colonies aimed to assess the extent to which the
composition of the bacteriomes of alates leaving the mother
colony could mirror the symbiotic microbiome of the colony
of origin.
e phylum-level distribution of the bacterial taxa highlighted
that a considerable proportion of the identied OTUs fell within
Bacteroidetes, Firmicutes, and Proteobacteria, which proportion
was not dierent between castes (Figure2A). e distribution of
OTUs was similar between minor and major workers within and
between colonies and between male and female alates. is similar
distribution pattern was much more visible at the genus level, with
the proportion of OTUs between the most representative genus-
level taxa varying hardly between castes and sexes (Figure2B).
Eect of sex on the size of the gut and
the structure of the microbiota
Guts of reproductives of both sexes dissected over the colony
foundation period starting from the swarming ight (alates) up
beyond the emergence of workers were used to assess the temporal
change in gut morphology and the structure of the microbiota.
Werst checked if the gut size and the structure of the microbiota
were aected by sex of the reproductive. e various parameters
of the gut morphology measured from image analyses of the
dissected guts of reproductives (Supplementary Figures S1A–C1)
did not signicantly dier between sexes during the colony
foundation period (p > 0.910 for the gut length, the hindgut area,
and the paunch width; and p = 0.761 for paunch area). Likewise,
there was no signicant dierence between sexes in the structure
and taxonomic composition of the mycobiome analyzed by
metabarcoding based on the fungal ITS and of the bacteriome
analysed by the metabarcoding based on the 16S rRNA gene
(ANOSIM, R = −0.026, p = 0.697 for the fungi; R = −0.037,
p = 0.759 for the bacteria). Consequently, the results presented
below were analyzed irrespective of the reproductive sexes.
Variations of the size of the gut of
reproductives over colony foundation
e area of the hindgut signicantly increased (p < 0.001)
between the alates (t = 0) and dealates collected on day 45
(Figure3). en this area remained stable until day 75 with no
signicant dierences between days 45, 60, and 75 (p > 0.05). e
whole gut length, the paunch area and the maximum paunch
width were also signicantly higher in dealates at days 60 and 75
compared to the alates (p < 0.005) (Supplementary Figure S2).
Beyond day 45, there were no variations of these parameters in
dealates. e paunch at this time appeared distended with soil-like
A
B
FIGURE2
Phylum-level (A) and genus-level (B) distribution of bacterial
OTUs in the gut of female alates (FA), male alates (MA), major
workers (MjW) and minor workers (MnW) of Macrotermes
subhyalinus from two distinct wild colonies. For both taxonomic
levels, only taxa covering a number of OTUs >0.5% of the total
number of OTUs of the dataset are displayed. The corresponding
taxa covered >97% of OTUs in each library for the phylum-level
and > 85% of OTUs for the genus-level. “Others” refers to the
cumulative percentage of the remaining OTUs. The dot sizes are
proportional to the relative number of OTUs for the
corresponding taxa in each library.

Diouf et al. 10.3389/fevo.2022.1055382
Frontiers in Ecology and Evolution 07 frontiersin.org
matter (Supplementary Figure S1D). Overall these gut morphology
measures were correlated, with the hindgut area being positively
correlated with the whole gut length (r = 0.734; p < 0.0001), the
area of the paunch sub-section (r = 0.988; p < 0.0001) and the
maximal width of the paunch (r = 0.835; p < 0.0001).
Diversity analyses from the Ion Torrent
metabarcoding data
For fungi, high-quality reads of ITS clustered into 3,439 OTUs
assigned to 3 phylum-level taxa from 25 classes and 165 genus-
level taxa (Spreadsheet “S1_fungi,” Supplementary Table S1).
Regarding the alpha diversity, both OTU richness (q = 0) and
diversity (exponential of Shannon index, q = 1) remained stable
between days 0 and 30 but signicantly increased at days 45 and
60 (Supplementary Table S2). Similarly, these two indices were
signicantly higher in the copularium than in gut samples.
Regarding the beta diversity the PCoA performed to assess the
eect of sampling time on the structure of the mycobiome showed
no clear clustering (Figure4A). An overall signicant eect of the
sampling time point emerged from the PERMANOVA test
(R2 = 0.275, p < 0.0001). e ANOSIM pairwise tests revealed
signicant dierences between days 0 and 45 and between days 30
and 45 (Supplementary Table S3).
For bacteria, high-quality reads of the 16S rRNA gene clustered
into 8,030 distinct OTUs, assigned to 27 phylum-level taxa, 74
classes, and 479 genus-level taxa (Spreadsheet “S1_Bacteria,”
Supplementary Table S1). e richness index did not vary between
gut samples over the 60-rst days (Supplementary Table S2).
Regarding the diversity, for the exponential of Shannon index,
(q = 1) and the multiplicative inverse of the Simpson index, (q = 2),
there was a signicant one-o decrease on days 30 to 45. e three
alpha-diversity indices were higher in the copularium environment
than in termite-gut samples. e structure of the bacteriomes
assessed by PCoA showed clustering of gut libraries per sampling
time point (Figure 4B). e bacterial communities in the
copularium samples clustered separately from those of the gut
samples. is clustering of the bacteriomes according to the
sampling time was underpinned by the PERMANOVA test
(R
2
= 0.425, p < 0.0001), showing that time explained more variance
for the bacteriome than for the mycobiome. Accordingly, the
ANOSIM pairwise test showed signicant dierences between
almost all time points except between days 10 and 30 and days 30
and 45 (Supplementary Table S3). Overall, there were distinct
patterns between the mycobiome and bacteriome, which was
further supported by the absence of signicant correlation between
both communities (Mantel test, rS = 0.06, p =0.26).
Taxonomic composition of the
microbiota
e gut mycobiome of alates was co-dominated by Ascomycota
and Basidiomycota (Figure 5A). From day 10, the relative
abundance of Basidiomycota decreased in favor of unclassied
fungi (i.e., fungal sequences that could not be taxonomically
assigned below the fungal kingdom level) except at 30 days.
Ascomycota accounted for about 90% of the copularium libraries.
At the genus level, of the 40 major taxa of the dataset, only four
clearly identied ones, from Ascomycota (Aspergillus, Retroconis,
Graphium, and Mortierella), were conserved over time (Figure5B).
e other conserved lineages were unclassied groups likely from
various distinct taxa. e ectosymbiotic fungus Termitomyces was
present in the gut of dealates until day 30, but not beyond.
Among the bacteria, Bacteroidetes, and to lesser extent Firmicutes
were the most represented phylum-level taxa in alates gut (Figure6A;
Supplementary Table S1). Bacteroidetes remained the dominant
phylum over time, but a gradual decrease was observed from day 0 to
day 45. Meanwhile, the relative abundance of Epsilonbacteraeota
increased. e proportion of Fibrobacteres in the gut was higher at day
120 than before. e relative abundance of Actinobacteria gradually
increased over time, exceeding 7% of reads at days 60 and 120. In the
copularium environment, Proteobacteria dominated, followed by
Actinobacteria and Firmicutes. At the genus-level, all the major
bacterial taxa of the dataset present in the gut of alates persisted in
dealates over the 60 rst days of the colony’s life, and only 4 were not
found by the end of the experiment (Figure6B). ese major genera
were dominant in the gut libraries, covering >94% of gut libraries
from days 0 to 60 and 88.2% at day 120. ey covered 56.2% of
libraries from the copularium. ese key genera primarily fell within
Bacteroidia, various lineages of Proteobacteria and Clostridiales, and
within well-known lineages in the gut of termites from the
Campylobacterales (Epsilonbacteraeota), Chitinivibrionia
(Fibrobacteres), Spirochaetaceae and Sphaerochaeta (Spirochaetes). A
noticeable increase of the relative abundance of Campylobacterales, of
which Arcobacter was observed from day 0 and to day 45.
FIGURE3
Boxplot representation of the change of the area of the whole
hindgut of M. subhyalinus reproductives during colony
foundation. For area measurements, the values in the figure
represent the half of the total surface (see
Supplementary Figure S1). Bars topped with p values <0.05
connect time points with statistically dierent measurements.

Diouf et al. 10.3389/fevo.2022.1055382
Frontiers in Ecology and Evolution 08 frontiersin.org
The core microbial OTUs and
Termitomyces
Core fungal OTUs were those found in the gut of reproductives
at all the sampling time points from the alates up to day 60. ey
were only 12 OTUs (out of 3,439; Sheet “Core OTUs of fungi,”
Supplementary Table S1). Together, they accounted for 5 to 10%
of gut libraries from days 0 to 30 and slightly more at days 45 and
60 (≈24%). is relative increase at 45 and 60 days was primarily
driven by a single OTU (unclassied Chaetomiaceae) that was the
most abundant in the copularium. Core fungal OTUs were even
more represented in the copularium than in gut samples.
Focusing on the ectosymbiotic fungus Termitomyces, the
phylogenetic inference showed that all OTUs from the
Lyophyllaceae in the dataset belonged to this genus Termitomyces
(Supplementary Figure S3). In alates, their average relative
AB
FIGURE4
PCoA based on Bray-Curtis distances displaying the change in the structure of the mycobiome (A) and of the bacteriome (B) in the gut of
M. subhyalinus reproductives over the colony foundation period, in comparison with the copularium at day 120. Day 0 to day 60 for fungi, and day
0 to day 120 for bacteria refer to gut samples at the corresponding time points.
AB
FIGURE5
Phylum-level (A) and genus-level (B) distribution of fungal taxa in the gut of M. subhyalinus reproductives over the colony foundation period, in
comparison with the copularium at day 120. For the genus-level, only the 40 most abundant taxa of the data set are displayed. The dot sizes are
proportional to the relative abundance of the corresponding taxa in each library and the absence of dot indicates the non-detection of the
corresponding taxon in a given library. « Others » refers to the sum of relative abundances of less abundant taxa.

Diouf et al. 10.3389/fevo.2022.1055382
Frontiers in Ecology and Evolution 09 frontiersin.org
abundance for both sexes was below 10%, with marked inter-
individual variability (Supplementary Table S4). eir proportion
sharply declined over time to zero beyond day 30. Members of
Lyophyllaceae were absent from the copularium
(Supplementary Table S1). With the taxon-specic PCR, 75% of
tests were positive in alates of both sexes from two distinct colonies
of M. subhyalinus, but no test was positive aer day 10. e
ectosymbionts of species from the genus Microtermes are vertically
transmitted through the gut of female reproductives. Alates of a
sympatric species Microtermes sp. were analysed for comparison
(n = 8 for each sex), resulting in 100% positive tests (data not shown).
Considering bacteria, 419 distinct OTUs fell within core OTUs
consisting of OTUs that persisted at all or at least at 5 of the 6
sampling time points from days 0 to 120 (Sheet “core OTUs of
bacteria,” Supplementary Table S1). ey accounted for more than
65% of OTUs in each library from days 0 to 60 and ≈ 60% at day
120 (Figure7A). e same was true in terms of abundance, since
they accounted for more than 73% of reads in all the guts libraries
up to 60 days (Figure7B) and remained highly representative at the
end of experiment. Compared to gut samples, core OTUs covered
only a small fraction of the number of OTUs and of the number of
reads in the copularium. eir taxonomic assignment mirrors the
taxonomic composition of the whole libraries. Besides unclassied
bacteria (i.e., bacterial sequences that could not betaxonomically
assigned below the Bacteria rank), Bacteroidetes, Firmicutes, and
Proteobateria were the most represented phyla. Only a few core
OTUs belonged to Epsilonibacteraeota, but these represented a
substantial fraction of gut libraries, especially at days 10 to 60.
Discussion
Variations of the size of reproductives’
gut during colony foundation
e survival of gut-resident microorganisms, and thereby
their persistence in reproductives during colony foundation,
was assumed to bedependent on their feeding behavior, which
was approximated by the gut morphology. e total gut length
and various dimensions of the hindgut did not dier between
male and female reproductives, as they do in distantly related
termites (Shimada etal., 2013). Similarly, the structure of the
microbiota did not dier between sexes. e absence of sexual
dimorphism on the microbiota was consistently observed on
protozoan communities of lower termites during the rst
months of colony foundation (Shimada etal., 2013; Michaud
etal., 2020), suggesting that both sexes are similarly involved in
feeding tasks at this stage. is involvement of both parents
(biparental care) was demonstrated in several studies on lower
termites (Shimada et al., 2013; Benjamino and Graf, 2016;
Inagaki etal., 2020; Velenovsky etal., 2021; Chouvenc, 2022)
and very recently on Macrotermes natalensis (Sinotte etal.,
2022) and was likely inherited from the common ancestor of
termites (Nalepa, 2015).
Notably, the gut size of M. subhyalinus dealates enlarged
during colony foundation, rather than shrinking as expected
under the hypothesis of a fasting at this stage. As illustrated in
Supplementary Figure S1, their paunch was lled with soil-like
AB
FIGURE6
Phylum-level (A) and genus-level (B) distribution of bacterial taxa in the gut of M. subhyalinus reproductives over the colony foundation period, in
comparison with the copularium at day 120. For the genus-level, only the 40 most abundant taxa of the data set are displayed. The dot sizes are
proportional to the relative abundance of the corresponding taxa in each library and the absence of dot indicates the non-detection of the
corresponding taxon in a given library. “Others” refer to the sum of relative abundances of less abundant taxa.

Diouf et al. 10.3389/fevo.2022.1055382
Frontiers in Ecology and Evolution 10 frontiersin.org
material. It is unclear whether the ingestion of soil material by
dealates is natural, or rather driven by the need of particulate
organic matter to compensate the unavailable wood material
needed to sustain the gut microbiome. e increase in the gut
size (Shimada etal., 2013) or relative weight (Nalepa etal.,
2001) seems common during the rst period of colony
foundations in lower termites. e enlargement of the gut seems
to bealigned with a gain in wet weight of reproductives during
this period (Nalepa etal., 2001; Chouvenc, 2022). However, the
dry body weight and the content in nutritional reserves decrease
at the same time (Johnston and Wheeler, 2007; Chouvenc,
2022), likely indicating that the increase of the gut size in
dealates reects the rehydration of the gut, rather than the
feeding behavior only. Albeit most of studies assessing the
feeding behavior concern lower termites, the engorgement of
dealates during the rst weeks of colony foundation may
be ubiquitous and can be a component of the observed
enlargement of the gut in the current study. Further
investigations taking into account weight indications and
analyzing the gut content should clarify the feeding behavior of
reproductives of this species during this period.
Contrast between bacteria and fungi
associated with reproductives
We characterized the whole gut mycobiome of
Macrotermitinae reproductives along colony foundation. e gut
of alates of M. subhylanus harbored dominantly Ascomycota and
Basidiomycota, among which all OTUs of Lyophyllaceae clustered
within Termitomyces. Despite the high frequency of detection in
alates (75% of alates, regardless sex), Termitomyces surprisingly
covered a low proportion of their gut mycobiome (< 10% of
reads), strikingly dierent from the worker caste, which is
dominated by Termitomyces in several Macrotermitinae species
(>98% of reads; Makonde, 2017).
In contrast, the bacteriome of alates of M. subhyalinus was similar
to that of workers. It predominately consists of Bacteroidetes,
Firmicutes, and Proteobacteria as in workers of other Macrotermes
species and some other Macrotermitinae (Hongoh et al., 2006;
Dietrich etal., 2014; Otani etal., 2014; Hu etal., 2019). e similarity
between alate and worker bacteriomes, which furthermore seems to
bewidespread (Hongoh etal., 2006; Diouf etal., 2018), indicates that
the bacteriome of alates is also representative of the host species.
Within colonies of Macrotermitinae, major and minor workers fulll
dierent tasks and have dierent feeding behavior (Sieber and
Leuthold, 1981). Yet, there was no marked dierence in the
composition of the bacteriome between major and minor workers
from wild colonies, which is congruent with previous reports on
other species of Macrotermintinae (Otani etal., 2019; Vidkjær etal.,
2021). Within-colony trophallaxis may therefore enable a suitable
transfer of the main bacterial taxa between nestmates. It is worth
mentioning that our study presented a fraction of reads assigned as
unclassied Bacteria (Figure6) and unclassied Fungi (Figure5).
is lack of taxonomic resolution can be explained by both the
sequencing technology used here (i.e., short-read sequencing of a few
hundred base pair amplicons, which limits the amount of
information) and by the origin of our samples (i.e., termite guts and
copularium) which are not investigated as much as human gut or soil
for instance. Regarding the former point, the use of shotgun
metagenomic combined with the reconstruction of metagenome-
assembled genomes has been successfully used to explore at the
AB
FIGURE7
Taxonomic distribution of core bacterial OTUs in libraries from the gut of M. susbhyalinus reproductives over the colony foundation period, in
comparison with the copularium at day 120. The significance of core OTUs in libraries is expressed either as a percentage of the number of OTUs
(A) or as a percentage of the number of reads (B).

Diouf et al. 10.3389/fevo.2022.1055382
Frontiers in Ecology and Evolution 11 frontiersin.org
genome level the prokaryotic diversity associated with termites
(Hervé etal., 2020; Arora etal., 2022). Further studies could thus use
this approach to monitor microbial transmission at the strain level (Su
et al., 2021). Regarding the latter point, there are indeed clades
associated with termites which has not been yet identied or
characterized, and this is particularly true for fungi (Větrovský etal.,
2020). is is further supported by the fact that for both fungal
(Figure5) and bacterial (Figure6) datasets, weconsistently found
much less unclassied bacterial and fungal sequences in the
copularium made of soil than in the termite guts.
Aside from alates, the microbiota of the reproductive caste of
Termitidae has only been studied in kings and queens of
Macrotermitinae from “mature” colonies several years aer
foundation. eir microbiota was markedly divergent and less
diverse than in workers’ gut, with only a single or very few
bacterial taxa undetectable in workers (Poulsen etal., 2014; Otani
etal., 2019), a compositional shi likely explained by trophallactic
food. is drastic divergence precluded any extrapolation on the
extent of their contribution in symbionts transmission. e
succession of the microbiota presented below partially lls the gap
between alates and these old reproductives. However, focusing on
the transitional phase where kings and queens pass from feeders
(parental care) to dependency to the nutritional resources
provided by ospring workers (alloparental care) should improve
our understanding the process leading to this erosion of the
bacteriome diversity in the gut of old kings and queens.
Succession of the microbiota at
foundation
e succession of bacteria and fungi in the gut microbiota
during colony foundation followed a distinct pattern, likely
reecting dierences in the tightness of the relationship. e
alpha diversity of the mycobiome, did not markedly vary up to
the day 30, which may reect the maintenance of fungal lineages
imported during the ight. Beyond day 30, the richness and the
exponential of Shannon indices increased together with gut
dimensions, possibly due to a transition in the feeding behavior
of dealates, which modied community structure. e widely
variable pattern of beta diversity of the mycobiome, even between
replicates, suggests that it mostly consists of transient fungi
ingested from the soil rather than resident, especially the rst
month. e dominance of Ascomycota, which are also the main
colonizers of the copularium supports an incidence of soil-borne
fungi. Similarly, the majority of Sordariomycetes core OTUs are
ubiquitous saprobic fungi (Zhang etal., 2006). Except for the
ectosymbiont, there is little information on strict associations
between Macrotermitinae and fungi. Taxon-specic PCR
detected Termitomyces in 75% of male and female alates of
M. subhyalinus. In comparison, 100% of positive tests were
recorded for both sexes of alates of the sympatric species
Microtermes sp. that inoculates its combs with conidia of
Termitomyces carried in the gut of female reproductives (Johnson,
1981; Johnson etal., 1981; Nobre etal., 2011a,b). While the
absence of sexual dimorphism for both species could indicate
that amplied DNA included material from the food bolus, the
lower score in M. subhyalinus could reect the dierence in the
transmission modes. e highlighting of vertical transmission in
Microtermes spp. and Macrotermes bellicosus in laboratory was
based on the presence of conidia bolus of Termitomyces in the
foregut of alates, and then the successful fertilization of fungus
combs without inoculation by the manipulator with comb
material from a mature colony (Johnson, 1981; Johnson etal.,
1981). is was not the case for M. subhyalinus (Johnson etal.,
1981) and though the molecular approach used here does not
make it possible to distinguish the nature of the fungal material,
the exhaustion of the molecular signal of Termitomyces before the
emergence of ospring, even with sensitive metabarcoding,
provides additional evidence of environmental acquisition of the
fungal ectosymbiont in M. subhyalinus i.e., the inoculation with
sexual spores harvested from the environment by the rst
foraging workers, as in the vast majority of Macrotermitinae. e
success of the colony foundations by reproductives in this case is
conditioned by the prior or simultaneous presence of
Termitomyces in the dispersal area (Nobre and Aanen, 2010). is
transmission mode, which is the most ancestral and the most
widespread in Macrotermitinae might have contributed to their
restricted geographic distribution only to Africa (the continent
of origin) and to Asia (Aanen and Eggleton, 2005). e large
prevalence of the horizontal transmission of Termitomyces also
led to the absence of host specicity. While a lineage-specic
pattern can bedrawn at higher taxonomic levels, host switches
are frequent at lower taxonomic levels, and no species of
Termitomyces is exclusively associated to a single termite host
(Aanen etal., 2002; Osiemo etal., 2010). Relevant dierences
were comparatively identied for bacteria. In contrast to fungi,
richness indexes for bacteria remained unchanged over the rst
months of the study, which is compatible with the hypothesis that
many bacterial taxa are not transiently ingested with the food but
are rather gut residents. One original aspect of our study was to
have analyzed each gut separately, preserving inter-individual
variability. Despite this, gut bacteriomes clustered by age and
diered from the copularium bacteriome, reinforcing the
endemism of bacterial lineages in the gut habitat. e temporary
decrease of Hill numbers q = 1 and q = 2 at days 30 and 45 under
constant richness coincided with the change in the gut
morphology. As the gut content was not analyzed, wequestion
whether it could beattributed to a dietary transition. is period
also coincides with the increase in the relative abundance of
Epsilonbacteraeota, and specially of a single OTU assigned to the
genus Arcobacter (21.27 to 24, 24% of reads at 30 and 45 days,
respectively); which may have decreased abundance-related
diversity indices. Arcobacter is a common genus of the core
bacterial communities of fungus-growing termites (Otani etal.,
2014, 2019). ough little is known about the metabolic functions
and factors driving its abundance in termites, the increase of the
relative abundance of Epsilonbacteraeota and specially of

Diouf et al. 10.3389/fevo.2022.1055382
Frontiers in Ecology and Evolution 12 frontiersin.org
Arcobacter could becompared with the protozoan pulse in the
gut lower termites during this rst period of the colony
foundation (Shimada etal., 2013; Benjamino and Graf, 2016;
Inagaki etal., 2020; Velenovsky etal., 2021), likely resulting from
the resumption of the wood-feeding activity of dealates to sustain
the protozoan fauna. Alongside with the compositional analyzes
performed here, quantifying bacterial density should clarify if the
microbial expansion found for protozoa applied to the gut
bacteriome of this higher termite.
Nevertheless, this transition did not markedly impact the
core bacteriome since the most abundant genus-level taxa and
core OTUs persisted for the 60 first days, indicating a high
conservation of key bacterial symbionts. These same key taxa
persisted until the end of the monitoring because only a few
were missing from the gut samples at day 120, where they still
dominated. To summarize, bacteria from the mother colony
are persistent in the gut and overcome the potential diet shift
associated with colony foundation, unlike fungi. As offspring
appeared as larvae by day 45 and as actively foraging workers
after 2 months, they co-exist with the core of bacteriome and
thus enable a vertical transmission within the colony of these
symbionts mostly adapted to the gut environment. Compared
to the transmission of the fungal exosymbiont, the bacteriome
the is likely transmitted through the same social exchanges
enabling the transmission of protozoa and that were acquired
in since the emergence of the ancestor of eusocial termites
(Nalepa, 2015; Chouvenc et al., 2021). This vertical
transmission likely explains the existence of termite-specific
clusters within numerous prokaryotic lineages, indicating a
coevolutionary process with termite hosts (Hongoh et al.,
2005; Abdul Rahman etal., 2015; Bourguignon etal., 2018).
In contrast to the gut bacteriome, the acquisition of the
externally cultured fungus symbiont is a derived trait, acquired
once by the ancestor of Macrotermitinae. Whether the loss of
protozoa that gave rise Termitidae evolved first a soil-feeding
behavior or rather an externalized digestion in combs is still
discussed (Bucek et al., 2019; Chouvenc et al., 2021).
Compared to the gut bacteriome, the delayed acquisition of
the fungal symbiont during the colony foundation seems to
support the hypothesis of postdated emergence of the ancestor
Macrotermitinae relative to the loss of protozoa. As suggested
by Bucek etal. (2019), fungus-growing termites may have
emerged from a soil-feeding ancestor by reversion toward a
wood feeding diet, fostered by the domestication of a saprobic
fungus predisposed to trive in insect fecal substrate (van de
Peppel etal., 2021).
Conclusion
e transmission of symbionts in termites has been
previously addressed for Protozoa that are unique to lower
termites and for fungal ectosymbionts that are unique to
Macrotermitinae. To our knowledge, this is the rst study
addressing the transmission of the gut symbionts in Termitidae,
which were assumed to have a distinct feeding behavior when
compared to lower termites, with a supposed fasting during the
colony foundation. To date, information on the microbiota of the
reproductive caste of Termitidae has been limited to alates, which
share a similar microbiota with workers, and to aged kings and
queens that have a distinct microbiota. Our study describes the
missing steps in the dynamics between these two extreme stages,
but further investigation from the end of our study up to old
kings and queens, and of the main changes taking place during
the transition from biparental to alloparantal care will provide a
more complete picture in the future. e persistence of gut
bacteria over the rst period of the colony’s life, the increase of
the gut size, and the presence of soil-borne fungi demonstrated
in this study did not support the existence of a fasting during the
colony foundation. Our ndings revealed no intimate relationship
between fungi and the gut habitat and the early exhaustion of
Termitomyces in the gut conrms that this symbiont is acquired
from the environment. Conversely, bacteria persisted in the gut
of reproductives throughout the colony foundation period, until
ospring workers emerged. e presence of these taxa in
ospring workers without any possible external source supports
the role of both reproductives in vertical transmission of gut
microbiota and, thereby, in the co-evolutionary history of
bacterial symbionts and their termite hosts.
Data availability statement
e original contributions presented in the study are included
in the article/Supplementary material, further inquiries can
bedirected to the corresponding author.
Author contributions
MD, CR-L, and M-AS conceived and designed the study. CR-F
and AN performed the eld sampling. SF, MD, and CR-L conducted
the lab rearing experiment, dissections, and morphological analyses.
SF, MD, JL, AB, and M-AS conducted molecular experiments. VH,
MJ, and EM conducted statistical and metabarcoding analyses. MD
and VH draed the manuscript. EM, M-AS, CR-F, AN, MJ, and AB
critically revised the intellectual content of the manuscript. e nal
content of the manuscript was edited and approved by all authors.
ey declare having agreed to beaccountable for all the aspects of the
study. All authors contributed to the article and approved the
submitted version.
Funding
is research was funded by the program Carrefour Écologie/
Sciences de l’Environnement of Sorbonne Universités
(CARESE-SU).

Diouf et al. 10.3389/fevo.2022.1055382
Frontiers in Ecology and Evolution 13 frontiersin.org
Acknowledgments
e authors would like to thank the Service de Systématique
Moléculaire (UMS 2700, MNHN, CNRS) for granting access to its
technical platform, the research teams of Laboratoire de Zoologie des
Invertébrés Terrestres (IFAN Ch. Anta Diop) and LEMSAT
(ECO&SOLS) for the allocation of human and logistic resources
during the eld mission. Wealso thank Stéphanie GIUSTI-MILLER
and Maxime FERREIRA for their technical and logistical support.
Conflict of interest
e authors declare that the research was conducted in the
absence of any commercial or nancial relationships that could
beconstrued as a potential conict of interest.
Publisher’s note
All claims expressed in this article are solely those of the
authors and do not necessarily represent those of their
affiliated organizations, or those of the publisher, the editors
and the reviewers. Any product that may be evaluated in this
article, or claim that may be made by its manufacturer, is not
guaranteed or endorsed by the publisher.
Supplementary material
e Supplementary material for this article can befound online
at: https://www.frontiersin.org/articles/10.3389/fevo.2022.1055382/
full#supplementary-material
References
Aanen, D. K., de Fine Licht, H. H., Debets, A. J. M., Kerstes, N. A. G.,
Hoekstra, R. F., and Boomsma, J. J. (2009). High symbiont relatedness stabilizes
mutualistic cooperation in fungus-growing termites. Science 326:1103. doi: 10.1126/
science.1173462
Aanen, D. K., and Eggleton, P. (2005). Fungus-growing termites originated in
African rain forest. Curr. Biol. 15, 851–855. doi: 10.1016/j.cub.2005.03.043
Aanen, D. K., Eggleton, P., Rouland-Lefèvre, C., Guldberg-Frøslev, T.,
Rosendahl, S., and Boomsma, J. J. (2002). e evolut ion of fungus-growing termites
and their mutualistic fungal symbionts. Proc. Natl. Acad. Sci. U. S.A. 99,
14887–14892. doi: 10.1073/pnas.22231309
Aanen, D. K., Ros, V. I. D., de Fine Licht, H. H., Mitchell, J., de Beer, Z. W.,
Slippers, B., et al. (2007). Patterns of interaction specicity of fungus-growing
termites and Ter mitomyces symbionts in SouthAfrica. BMC Evol. Biol. 7:115. doi:
10.1186/1471-2148-7-115
Abdul Rahman, N., Parks, D. H., Willner, D. L., Engelbrektson, A. L., Goredi, S. K.,
Warnecke, F., et al. (2015). A molecular survey of Australian and north American
termite genera indicates that vertical inheritance is the primary force shaping termite
gut microbiomes. Microbiome 3:5. doi: 10.1186/s40168-015-0067-8
Alberdi, A., and Gilbert, M. T. P. (2019). Hilldiv: an R package for the integral
analysis of diversity based on hill numbers. bio Rxiv. doi: 10.1101/545665
Anisimova, M., and Gascuel, O. (2006). Approximate likelihood-ratio test for
branches: a fast, accurate, and powerful alternative. Syst. Biol. 55, 539–552. doi:
10.1080/1063515 0600755453
Arora, J., Kinjo, Y., Šobotník, J., Buček, A., Clitheroe, C., Stiblik, P., et al. (2022).
e functional evolution of termite gut microbiota. Microbiome 10:78. doi: 10.1186/
s40168-022-01258-3
Badertscher, S., Gerber, C., and Leuthold, R. H. (1983). Polyethism in food supply
and processing in termite colonies of Macrotermes subhyalinus (Isoptera). Behav.
Ecol. Sociobiol 12, 115–119. doi: 10.1007/BF00343201
Benjamino, J., and Graf, J. (2016). Characterization of the Core and caste-specic
microbiota in the termite, Reticulitermes avipes. Front. Microbiol. 7:171. doi:
10.3389/fmicb.2016.00171
Bignell, D. E. (2016). “e role of symbionts in t he evolution of termites and their
rise to ecological dominance in the tropics” in e Mechanistic Benets of Microbial
Symbionts. ed. C. J. Hurst (Cham: Springer International Publishing)
Bourguignon, T., Lo, N., Cameron, S. L., Šobotník, J., Hayashi, Y., Shigenobu, S.,
et al. (2014). e evolutionary history of termites as inferred from 66 mitochondri al
genomes. Mol. Biol. Evol. 32, 406–421. doi: 10.1093/molbev/msu308
Bourguignon, T., Lo, N., Dietrich, C., Šobotník, J., Sidek, S., Roisin, Y., et al.
(2018). Rampant host switching shaped the termite gut microbiome. Curr. Biol. 28,
649–654.e2. doi: 10.1016/j.cub.2018.01.035
Brown, S. P., Callaham, M. A., Oliver, A. K., and Jumpponen, A. (2013). Deep ion
torrent sequencing identies soil fungal community shis aer frequent prescribed
res in a southeastern US forest ecosystem. FEMS Microbiol. Ecol. 86, 557–566. doi:
10.1111/1574-6941.12181
Brune, A. (2014). Symbiotic digestion of lignocellulose in termite guts. Nat. Rev.
Microbiol. 12, 168–180. doi: 10.1038/nrmicro3182
Bucek, A., Šobotník, J., He, S., Shi, M., McMahon, D. P., Holmes, E. C., et al.
(2019). Evolution of termite Symbiosis informed by transcriptome-based
phylogenies. Curr. Biol. 29, 3728–3734.e4. doi: 10.1016/j.cub.2019.08.076
Chao, A., Chiu, C. H., and Jost, L. (2016). “Phylogenetic diversity measures and
their decomposition: a framework based on hill numbers” in Biodiversity
conservation and phylogenetic systematics: Preser ving our evolutionary heritage in an
extinction crisis. eds. R. Pellens and P. Grandcolas, vol. 14 (Cham: Springer) doi:
10.1007/978-3-319-22461-9_8
Cheung, M. K., Wong, C. K., Chu, K. H., and Kwan, H. S. (2018). Community
structure, dynamics and interactions of bacteria, archaea and fungi in
subtropical coastal wetland sediments. Sci. Rep. 8:14397. doi: 10.1038/
s41598-018-32529-5
Chouvenc, T. (2022). Eusociality and the transition from biparental to alloparental
care in termites. Funct. Ecol. 36, 3049–3059. doi: 10.1111/1365-2435.14183
Chouvenc, T., Šobotník, J., Engel, M. S., and Bourguignon, T. (2021). Termite
evolution: mutualistic associations, key innovations, and the rise of Termitidae. Cell.
Mol. Life Sci. 78, 2749–2769. doi: 10.1007/s00018-020-03728-z
Collins, N.M. (1977). e population ecology and energetics of Macrotermes
bellicosus Smeathman (Isoptera). Doctoral thesis London: University of London.
da Costa, R. R., and Poulsen, M. (2018). Mixed-mode transmission shapes termite
gut community assemblies. Trends Microbiol. 26, 557–559. doi: 10.1016/j.
tim.2018.04.005
Dietrich, C., Köhler, T., and Brune, A. (2014). The cockroach origin of the
termite gut microbiota: patterns in bacterial community structure reflect major
evolutionary events. Appl. Environ. Microbiol. 80, 2261–2269. doi: 10.1128/
AEM.04206-13
Diouf, M., Hervé, V., Mora, P., Robert, A., Frechault, S., Rouland-Lefèvre, , et al. (2018).
Evidence from the gut microbiota of swarming alates of a vertical transmission of the
bacterial symbionts in Nasutitermes arborum (Termitidae, Nasutitermitinae). Antonie Van
Leeuwenhoek 111, 573–587. doi: 10.1007/s10482-017-0978-4
Diouf, M., and Rouland-Lefevre, C. (2018). “e fungus-growing termites:
biology, damage on tropical crops and specic management” in Termites and
Sustainable Management: Volume 2- Economic LOSSES and Management. eds. M. A.
Khan and W. Ahmad (Cham: Springer International Publishing)
Diouf, M., Roy, V., Mora, P., Frechault, S., Lefebvre, T., and Hervé, V., Cham
(2015). Proling the succession of bacterial communities throughout the life stages
of a higher termite Nasutitermes arborum (Termitidae, Nasutitermitinae) using 16S
rRNA gene pyrosequencing. PLoS One 10: 1–15. PMID: 26444989 doi:10.1371/
journal.pone.0140014
Edgar, R. C., Haas, B. J., Clemente, J. C., Quince, C., and Knight, R. (2011).
UCHIME improves sensitivity and speed of chimera detection. Bioinformatics 27,
2194–2200. doi: 10.1093/bioinformatics/btr381
Engel, M. S., Grimaldi, D. A., and Krishna, K. (2009). Termites (Isoptera): their
phylogeny, classication, and rise to ecological dominance. Am. Mus. Novit. 2009,
1–27. doi: 10.1206/651.1
Goslee, S., and Urban, D. (2007). e ECODIST package for dissimilarity-based
analysis of ecological data. J. Stat. Sow. 22, 1–19. doi: 10.18637/jss.v022.i07

Diouf et al. 10.3389/fevo.2022.1055382
Frontiers in Ecology and Evolution 14 frontiersin.org
Grassé, P.-P. (1984). “La fondation des nouvelles sociétés” in Termitologia Tome
II: Fondation des sociétés et construction. ed. P.-P. Grassé (Paris: Masson)
Grassé, P.-P., and Noirot, C. (1955). La fondation de nouvelles sociétés par
Bellicositermes natalensis H av. Insect. Soc. 2, 213–220. doi: 10.1007/BF02224382
Guindon, S., Dufayard, J.-F., Lefort, V., Anisimova, M., Hordijk, W., and
Gascuel, O. (2010). New algorithms and methods to estimate maximum-likelihood
phylogenies: assessing the performance of PhyML 3.0. Syst. Biol. 59, 307–321. doi:
10.1093/sysbio/syq010
Han, S. H., and Bordereau, C. (1982). Origin and formation of the royal fat body
of the higher termite queens. J. Morphol. 173, 17–28. doi: 10.1002/jmor.1051730103
Harit, A. K., Gajalakshmi, S., and Abbasi, S. A. (2016). Studies on the development
of captive termite colonies. Zoo. Ecol. 26, 301–312. doi:
10.1080/21658005.2016.1228731
Hervé, V., Liu, P., Dietrich, C., Sillam-Dussès, D., Stiblik, P., Šobotník, , et al.
(2020). Phylogenomic analysis of 589 metagenome-assembled genomes
encompassing all major prokaryotic lineages from the gut of higher termites. PeerJ
8:e8614. doi: 10.7717/peerj.8614
Hervé, V., and Lopez, P. J. (2020). Analysis of interdomain taxonomic patterns in
urban street mats. Environ. Microbiol. 22, 1280–1293. doi: 10.1111/1462-2920.14933
Hongoh, Y., Deevong, P., Inoue, T., Moriya, S., Trakulnaleamsai, S.,
Ohkuma, M., et al. (2005). Intra- and interspecic comparisons of bacterial
diversity and community structure support coevolution of gut microbiota and
termite host. Appl. Environ. Microbiol. 71, 6590–6599. doi: 10.1128/
AEM.71.11.6590-6599.2005
Hongoh, Y., Ekpornprasit, L., Inoue, T., Moriya, S., Trakulnaleamsai, S.,
Ohkuma, M., et al. (2006). Intracolony variation of bacteria l gut microbiota among
castes and ages in the fungus-growing termite Macrotermes gilvus. Mol. Ecol. 15,
505–516. doi: 10.1111/j.1365-294X.2005.02795.x
Hu, H., da Costa, R. R., Pilgaard, B., Schiøtt, M., Lange, L., and Poulsen, M.
(2019). Fungiculture in termites is associated with a Mycolytic gut bacterial
community. mSphere 4, e00165–e00119. doi: 10.1128/mSphere.00165-19
Inagaki, T., Yanagihara, S., Fuchikawa, T., and Matsuura, K. (2020). Gut microbial
pulse provides nutrition for parental provisioning in incipient termite colonies.
Behav. Ecol. Sociobiol. 74:64. doi: 10.1007/s00265-020-02843-y
Johnson, R. A. (1981). Colony development and establishment of the fungus comb
in Microtermes sp. nr. Usambaricus (Sjöstedt) (Isoptera: Macrotermitinae) from
Nigeria. Insect. Soc. 28, 3–12. doi: 10.1007/BF02223617
Johnson, R. A., omas, R. J., Wood, T. G., and Swi, M. J. (1981). e inoc ulat ion
of the fungus comb in newly founded colonies of some species of the Macrotermitinae
(Isoptera) from Nigeria. J. Nat. Hist. 15, 751–756. doi: 10.1080/00222938100770541
Johnston, M. L., and Wheeler, D. E. (2007). e role of storage proteins in
colony-founding in termites. Insect. Soc. 54, 383–387. doi: 10.1007/
s00040-007-0957-2
Katoh, K., and Standley, D. M. (2013). MAFFT multiple sequence alignment
soware version 7: improvements in performance and usability. Mol. Biol. Evol. 30,
772–780. doi: 10.1093/molbev/mst010
Kwong, W. K., Medina, L. A., Koch, H., Sing, K.-W., Soh, E. J. Y., Ascher, J. S., et al.
(2017). Dynamic microbiome evolution in social bees. Sci. Adv. 3, –e1600513. doi:
10.1126/sciadv.1600513
Lefort, V., Longueville, J.-E., and Gascuel, O. (2017). SMS: smart model selection
in PhyML. Mol. Biol. Evol. 34, 2422–2424. doi: 10.1093/molbev/msx149
Lepage, M., and Darlington, J. P. E. C. (2000). “Population dynamics of termites”
in Termites: Evolution, Sociality, Symbioses, Ecology. eds. T. Ab e, D. E. Bignell and M.
Higashi (Dordrecht: Springer Netherlands), 333–361.
Makonde, H. (2017). Fungal diversity and community structure in gut, mound
and surrounding soil of fungus-cultivating termites. Afr. J. Microbiol. Res. 11,
504–515. doi: 10.5897/AJMR2017.8484
McKnight, D. T., Huerlimann, R., Bower, D. S., S chwarzkopf, L., Alford, R. A., and
Zenger, K. R. (2019). Methods for normalizing microbiome data: an ecological
perspective. Methods Ecol. Evol. 10, 389–400. doi: 10.1111/2041-210X.13115
McMurdie, P. J., and Holmes, S. (2013). Phyloseq: an R package for reproducible
interactive analysis and graphics of microbiome census data. PLoS One 8:e61217.
doi: 10.1371/journal.pone.0061217
Michaud, C., Hervé, V., Dupont, S., Dubreuil, G., Bézier, A. M., Meunier, J.,
et al. (2020). Ecient but occasionally imperfect vertical transmission of gut
mutualistic protists in a wood-feeding termite. Mol. Ecol. 29, 308–324. doi:
10.1111/mec.15322
Mikaelyan, A., Dietrich, C., Köhler, T., Poulsen, M., Sillam-Dussès, D., and
Brune, A. (2015). Diet is the primary determinant of bacterial community structure
in the guts of higher termites. Mol. Ecol. 24, 5284–5295. doi: 10.1111/mec.13376
Mikaelyan, A., Meuser, K., and Brune, A. (2017). Microenvironmental
heterogeneity of gut compartments drives bacterial community structure in wood-
and humus-feeding higher termites. FEMS Microbiol. Ecol. 93:w 210. doi: 10.1093/
femsec/w210
Mitchell, J. D. (2020). Colony foundation and the development of incipient
laboratory colonies of Macrotermes natalensis (Haviland) (Termitidae:
Macrotermitinae). Afr. Entomol. 28, 215–224. doi: 10.4001/003.028.0215
Nalepa, C. A. (2015). Origin of termite eusociality: trophallaxis integrates the
social, nutritional, and microbial environments. Ecol. Entomol. 40, 323–335. doi:
10.1111/een.12197
Nalepa, C., Miller, L., and Lenz, M. (2001). Flight characteristics of Mastotermes
darwiniensis (Isoptera, Mastotermitidae). Insect. Soc. 48, 144–148. doi: 10.1007/
PL00001757
Nilsson, R. H., Larsson, K.-H., Taylor, A. F. S., Bengtsson-Palme, J., Jeppesen, T. S.,
Schigel, D., et al. (2018). e UNITE database for molecular ident ication of fungi:
handling dark taxa and parallel taxonomic classications. Nucleic Acids Res. 47,
D259–D264. doi: 10.1093/nar/gky1022
Nobre, T., and Aanen, D. K. (2010). Dispersion and colonisation by fungus-
growing termites: vertical transmission of the symbiont helps, but then..? Commun.
Integr. Biol. 3, 248–250. doi: 10.4161/cib.3.3.11415
Nobre, T., Koné, N. A., Konaté, S., Linsenmair, K. E., and Aanen, D. K. (2011a).
Dating the fungus-growing termites’ mutualism shows a mixture between ancient
codiversication and recent symbiont dispersal across divergent hosts. Mol. Ecol. 20,
2619–2627. doi: 10.1111/j.1365-294X.2011.05090.x
Nobre, T., Rouland-Lefèvre, C., and Aanen, D. K. (2011b). “Comparative
biology of fungus cultivation in termites and ants” in Biology of Termites: A
Modern Synthesis. eds. D. E. Bignell, Y. Roisin and N. Lo (Dordrecht:
Springer Netherlands)
Noda, S., Kitade, O., Inoue, T., Kawai, M., Kanuta, M., Hiroshima, K., et al. (2007).
Cospeciation in the triplex symbiosis of termite gut protists (Pseudotrichonympha
spp.), their hosts, and their bacterial endosymbionts. Mol. Ecol. 16, 1257–1266. doi:
10.1111/j.1365-294X.2006.03219.x
Nutting, W. L. (1969). “Flight and colony foundation” in Biology of Termites. eds.
K. Krishna and F. Weesner (New York: Academic Press)
Okot-Kotber, B. M. (1981). Polymorphism and the development of the rst
progeny in incipient colonies of Macrotermes michaelseni (Isoptera,
Macrotermitinae). Ins. Sci. Appl. 1, 147–150. doi: 10.1017/S174275840000031X
Oksanen, J., Kindt, R., Legendre, P., O’hara, B., Henry, M., and Maintainer, H.S.
(2007). e vegan package title community ecology package. Cran R-Project
Org/,Http://R-Forge R-Project Org/Projects/Vegan/
Otani, S., Mikaelyan, A., Nobre, T., Hansen, L. H., Koné, N. A., Sørensen, S. J.,
et al. (2014). Identifying the core microbial community in t he gut of fungus-growing
termites. Mol. Ecol. 23, 4631–4644. doi: 10.1111/mec.12874
Otani, S., Zhukova, M., Koné, N. A., da Costa, R. R., Mikaelyan, A., Sapountzis, P.,
et al. (2019). Gut microbial compositions mirror caste-specic diets in a major lineage
of social insects. Environ. Microbiol. Rep. 11, 196–205. doi: 10.1111/1758-2229.12728
Osiemo, Z.B., Marten, A., Kaib, M., Gitonga, L.M., Boga, H.I., and Brandl, R.
(2010). Open Relationships in the Castles of Clay: High Diversity and Low Host
Specicity of Termitomyces Fungi Associated with Fungus-Growing Termites in
Africa. Insect. Soc 57, 351–363. doi: 10.1007/s00040-010-0092-3
Poulsen, M., Hu, H., Li, C., Chen, Z., Xu, L., Otani, S., et al. (2014).
Complementary symbiont contributions to plant decomposition in a fungus-
farming termite. Proc. Natl. Acad. Sci. U. S. A. 111, 14500–14505. doi: 10.1073/
pnas.1319718111
Quast, C., Pruesse, E., Yilmaz, P., Gerken, J., Schweer, T., Yarza, P., et al.
(2012). The SILVA ribosomal RNA gene database project: improved data
processing and web-based tools. Nucleic Acids Res. 41, D590–D596. doi:
10.1093/nar/gks1219
Quinlan, A. R., and Hall, I. M. (2010). BEDTools: a exible suite of utilities for
comparing genomic features. Bioinformatics 26, 841–842. doi: 10.1093/
bioinformatics/btq033
Rouland-Lefèvre, C. (2000). “Symbiosis with fungi” in Termites: Evolution,
Sociality, Symbioses, Ecology. eds. T. Abe, D. E. Bignell and M. Higashi (Dordrecht
Netherlands: Springer)
Sands, W. A. (1960). e initiation of fungus comb construction in laboratory
colonies of Ancistrotermes guineensis (Silvestri). Insect. Soc. 7, 251–263. doi: 10.1007/
BF02224496
Schloss, P. D. (2008). Evaluating dierent approaches that test whether microbial
communities have the same structure. ISME J. 2, 265–275. doi: 10.1038/ismej.2008.5
Schloss, P. D., Westcott, S. L., Ryabin, T., Hall, J. R., Hartmann, M., Hollister, E. B.,
et al. (2009). Introducing mothur: open-source, platform-independent, community-
supported soware for describing and comparing microbial communities. Appl.
Environ. Microbiol. 75, 7537–7541. doi: 10.1128/AEM.01541-09
Schneider-Maunoury, L., Leclercq, S., Clément, C., Covès, H., Lambourdière, J.,
Sauve, M., et al. (2018). Is tuber melanosporum colonizing the roots of herbaceous,

Diouf et al. 10.3389/fevo.2022.1055382
Frontiers in Ecology and Evolution 15 frontiersin.org
non-ectomycorrhizal plants? Fungal Ecol. 31, 59–68. doi: 10.1016/j.
funeco.2017.10.004
Shimada, K., Lo, N., Kitade, O., Wakui, A., and Maekawa, K. (2013).
Cellulolytic Protist numbers rise and fall dramatically in termite Queens and
kings during Colony Foundation. Eukaryot. Cell 12, 545–550. doi: 10.1128/
EC.00286-12
Sieber, R. (1983). Establishment of fungus comb in laboratory colonies of
Macrotermes michaelseni and Odontotermes montanus (Isoptera, Macrotermitinae).
Insect. Soc. 30, 204–209. doi: 10.1007/BF02223870
Sieber, R., and Leuthold, R. H. (1981). Behavioural elements and their meaning
in incipient laboratory colonies of the fungus-growing termite Macrotermes
michaelseni (Isoptera: Macrotermitinae). Insect. Soc. 28, 371–382. doi: 10.1007/
BF02224194
Sinotte, V., Renelies-Hamilton, J., Andreu-Sánchez, S., Vasseur-Cognet, M., and
Poulsen, M. (2022). Extensive inheritance of gut microbial communities in a
superorganismal termite. doi: 10.21203/rs.3.rs-1978279/v1,
Su, Q., Wang, Q., Mu, X., Chen, H., Meng, Y., Zhang, X., et al. (2021). Strain-level
analysis reveals the vertical microbial transmission during the life cycle of
bumblebee. Microbiome 9:216. doi: 10.1186/s40168-021-01163-1
Turenne, C. Y., Sanche, S. E., Hoban, D. J., Karlowsky, J. A., and Kabani, A. M.
(1999). Rapid identication of fungi by using the ITS2 genetic region and an
automated uorescent capillary electrophoresis system. J. Clin. Microbiol. 37,
1846–1851. doi: 10.1128/JCM.37.6.1846-1851.1999
van de Peppel, L. J. J., Nieuwenhuis, M., Auxier, B., Grum-Grzhimaylo, A. A.,
Cárdenas, M. E. E., de Beer, Z. W., et al. (2021). Ancestral predisposition toward a
domesticated lifestyle in the termite-cultivated fungus Termitomyces. Curr. Biol. 31,
4413–4421.e5. doi: 10.1016/j.cub.2021.07.070
Velenovsky, J. F., Gile, G. H., Su, N. Y., and Chouvenc, T. (2021). Dynamic
protozoan abundance of Coptotermes kings and queens during the transition from
biparental to alloparental care. Insect. Soc. 68, 33–40. doi: 10.1007/
s00040-021-00808-6
Vesala, R., Niskanen, T., Liimatainen, K., Boga, H., Pellikka, P., and Rikkinen, J.
(2017). Diversity of fungus-growing termites (Macrotermes) and their fungal
symbionts (Termitomyces) in the semiarid Tsavo ecosystem, Kenya. Biotropica 49,
402–412. doi: 10.1111/btp.12422
Větrovský, T., Soukup, P., Stiblik, P., Votýpková, K., Chakraborty, A., Odriozola, I., et al.
(2020). Termites host specic f ungal communit ies that dier from t hose in their ambient
environments. Fungal Ecol. 48:100991. doi: 10.1016/j.funeco.2020.100991
Vidkjær, N. H., Schmidt, S., Hu, H., Bodawatta, K. H., Beemelmanns, C., and
Poulsen, M. (2021). Species- and caste-specic gut metabolomes in fungus-farming
termites. Meta 11:839. doi: 10.3390/metabo11120839
Wang, Q., Garrity, G. M., Tiedje, J. M., and Cole, J. R. (2007). Naïve Bayesian
classier for rapid assignment of rRNA sequences into the new bacterial taxonomy.
Appl. Environ. Microbiol. 73, 5261–5267. doi: 10.1128/AEM.00062-07
Waud, M., Busschaert, P., Ruyters, S., Jacquemyn, H., and Lievens, B. (2014). Impact
of primer choice on characterization of orchid mycorrhizal communities using 454
pyrosequencing. Mol. Ecol. Resour. 14, 679–699. doi: 10.1111/1755-0998.12229
Weesner, F. (1969). “External anatomy” in Biology of Termites. ed. K. Krishna
(New York, USA: Academic Press)
Westcott, S. L., and Schloss, P. D. (2017). Opti Clust, an improved method for
assigning amplicon-based sequence data to operational taxonomic units. mSphere
2, e00073–e00017. doi: 10.1128/mSphereDirect.00073-17
White, T., Bruns, T., Lee, S., and Taylor, J. (1990). “Amplication and direct
sequencing of fungal ribosomal RNA genes for phylogenetics” in PCR Protocols: A
Guide to Methods and Applications. eds. M. A. Innis, D. H. Gelfand, J. J. Sninsky and
T. J. White (New York: Academic Press)
Wickham, H. (2016). Package `ggplot 2`: Elegant graphics for data analysis.
Springer-Verlag New York
Zhang, N., Castlebury, L., Miller, A., Huhndorf, S. M., Schoch, C., Seifert, K., et al.
(2006). An overview of the systematics of the Sordariomycetes based on a four-gene
phylogeny. Mycologia 98, 1076–1087. doi: 10.1080/15572536.2006.11832635