of 80
1
Research Article - Discoveries
1
Molecular evolutionary trends and biosynthesis pathways in the Oribatida revealed by the
2
genome of
Archegozetes longisetosus
3
Adrian Brückner
1
*, Austen A. Barnett
2
, Igor A. Antoshechkin
1
and Sheila A. Kitchen
1
4
1
Division of Biology and Biological Engineering, California Institute of Technology, 1200 East
5
California Boulevard, Pasadena, CA 91125, United States of America
6
2
Department of Biology, DeSales University, 2755 Station Avenue, Center Valley, PA 18034,
7
United States of America
8
* corresponding author: Adrian Brückner
-
adrian.brueckner@gmail.com
9
10
Abstract
11
Oribatid mites are a specious order of microarthropods within the subphylum Chelicerata,
12
compromising about 11,000 described species. They are ubiquitously distributed across different
13
microhabitats in all terrestrial ecosystems around the world and were among the first animals
14
colonizing terrestrial habitats as decomposers and scavengers. Despite their species richness and
15
ecological importance genomic resources are lacking for oribatids. Here, we present a 190-Mb
16
genome assembly of the clonal, all-female oribatid mite species
Archegozetes longisetosus
Aoki
,
17
a model species used by numerous laboratories for the past 30 years. Comparative genomic and
18
transcriptional analyses revealed patterns of reduced body segmentation and loss of segmental
19
identity gene
abd-A
within Acariformes, and unexpected expression of key eye development
20
genes in these eyeless mites across developmental stages. Consistent with their soil dwelling
21
lifestyle, investigation of the sensory genes revealed a species-specific expansion of gustatory
22
receptors, the largest chemoreceptor family in the genome used in olfaction, and evidence of
23
horizontally transferred enzymes used in cell wall degradation of plant and fungal matter, both
24
components of the
A.
longisetosus
diet. Oribatid mites are also noted for their biosynthesis
25
capacities and biochemical diversity. Using biochemical and genomic data, we were able to
26
delineate the backbone biosynthesis of monoterpenes, an important class of compounds found in
27
the major exocrine gland system of Oribatida – the oil glands. Given the mite’s strength as an
28
.
CC-BY-NC-ND 4.0 International license
available under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprint
this version posted February 9, 2021.
;
https://doi.org/10.1101/2020.12.10.420141
doi:
bioRxiv preprint
2
experimental model, the new high-quality resources provided here will serve as the foundation
29
for molecular research in Oribatida and will enable a broader understanding of chelicerate
30
evolution.
31
32
Keywords
33
soil animal, terpene synthesis, horizontal gene transfer, parthenogenesis, chemoreceptors,
34
Hox genes, model organism, RNAseq, MinION long-read sequencing, Sarcoptiformes
35
36
Introduction
37
In the past couple of years, the number of sequenced animal genomes has increased
38
dramatically, especially for arthropods about 500 genomes sequences are now available
39
(Childers 2020; Thomas et al. 2020). The majority of these genomes, however, belong to the
40
insects (e.g. flies, beetles, wasp, butterflies and bugs (Thomas et al. 2020)) which compromise
41
the most diverse, yet evolutionarily young and more derived taxa of arthropods (Giribet and
42
Edgecombe 2019; Regier et al. 2010). In strong contrast, genome assemblies, many of which are
43
incomplete or not well annotated, exist for the Chelicerata (Childers 2020) – the other major
44
subphylum of arthropods (Giribet and Edgecombe 2019; Regier et al. 2010). Chelicerates include
45
sea spiders, spiders, mites and scorpions among other organisms, as well as several extinct taxa
46
(Ballesteros and Sharma 2019; Dunlop and Selden 1998). Chelicerates originated as marine
47
animals about 500 million years ago (Dunlop and Selden 1998; Dunlop 2010). Molecular
48
analyses suggest that one particular group, the omnivorous and detritivores acariform mites, may
49
have been among the first arthropods that colonized terrestrial habitats and gave rise to ancient,
50
simple terrestrial food webs (Dunlop and Alberti 2008; Schaefer et al. 2010; Walter and Proctor
51
1999).
52
So far, the well-annotated genomic data of chelicerates is limited to animal parasites
53
(including human pathogens and ticks), plant parasites, and predatory mites used in pest control
54
(Cornman et al. 2010; Dong et al. 2017; Dong et al. 2018; Grbić et al. 2011; Gulia-Nuss et al.
55
2016; Hoy et al. 2016; Rider et al. 2015). Other than some lower-quality genome assemblies
56
.
CC-BY-NC-ND 4.0 International license
available under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprint
this version posted February 9, 2021.
;
https://doi.org/10.1101/2020.12.10.420141
doi:
bioRxiv preprint
3
(Bast et al. 2016), there are no resources available for free-living soil and litter inhabiting
57
species. Such data are, however, pivotal to understanding the evolution of parasitic lifestyles
58
from a free-living condition and to bridge the gap between early aquatic chelicerates such as
59
horseshoe crabs, and highly derived terrestrial pest species and parasites (Klimov and OConnor
60
2013; Shingate et al. 2020; Weinstein and Kuris 2016). Because the phylogeny of Chelicerata
61
remains unresolved, additional chelicerate genomes are urgently needed for comparative
62
analyses (Ballesteros and Sharma 2019; Dunlop 2010; Lozano-Fernandez et al. 2019). To help
63
address this deficit, we report here the genome assembly of the soil dwelling oribatid mite
64
Archegozetes longisetosus
(Aoki, 1965;
Figure 1
) (Aoki 1965) and a comprehensive analysis in
65
the context of developmental genes, feeding biology, horizontal gene transfer and biochemical
66
pathway evolution of chelicerates.
67
Archegozetes longisetosus
(hereafter referred to as
Archegozetes
) is a member of the
68
Oribatida (Acariformes, Sarcoptiformes), an order of chelicerates well-known for their
69
exceptional biosynthesis capacities, biochemical diversity, unusual mode of reproduction,
70
unusually high pulling strength, mechanical resistances and pivotal ecological importance
71
(Brückner et al. 2020; Brückner et al. 2017b; Heethoff and Koerner 2007; Heethoff et al. 2009;
72
Maraun et al. 2007; Maraun and Scheu 2000; Norton and Palmer 1991; Raspotnig 2009;
73
Schmelzle and Blüthgen 2019).
Archegozetes
, like all members of its family Trhypochthoniidae
74
(
Figure 1a
), reproduce via thelytoky (Heethoff et al. 2013). That means the all-female lineages
75
procreate
via
automictic parthenogenesis with an inverted meiosis of the holokinetic
76
chromosomes, resulting in clonal offspring (Bergmann et al. 2018; Heethoff et al. 2006; Palmer
77
and Norton 1992; Wrensch et al. 1994). While studying a parthenogenetic species is useful for
78
the development of genetic tools as stable germ-line modifications can be obtained from the
79
clonal progeny without laboratory crosses, one is confronted with the technical and philosophical
80
problems of species delineation, cryptic diversity and uncertain species distribution (Heethoff et
81
al. 2013; Oxley et al. 2014). Reviewing all available data, Norton (Norton 1994; 2007) and
82
Heethoff et al. (Heethoff et al. 2013) concluded that
Archegozetes
is found widely on continents
83
and islands throughout the tropical and partly subtropical regions of the world and that it is a
84
middle-derived oribatid mite closely related to the suborder Astigmata.
85
.
CC-BY-NC-ND 4.0 International license
available under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprint
this version posted February 9, 2021.
;
https://doi.org/10.1101/2020.12.10.420141
doi:
bioRxiv preprint
4
One major feature of most oribatid mites is a pair of opisthonotal oil-glands and
86
Archegozetes
is no exception (Raspotnig 2009; Sakata and Norton 2001). These are a pair of
87
large exocrine glands, each composed of a single-cell layer invagination of the cuticle, which is
88
the simplest possible paradigm of an animal gland (Brückner and Parker 2020; Heethoff 2012).
89
The biological role of these glands was rather speculative for a long time; ideas ranged from a
90
lubricating and osmo- or thermoregulative function (Riha 1951; Smrž 1992; Zachvatkin 1941) to
91
roles in chemical communication (Heethoff et al. 2011a; Raspotnig 2006; Shimano et al. 2002).
92
So far about 150 different gland components have been identified from oribatid mites, including
93
mono- and sesquiterpenes, aldehydes, esters, aromatics, short-chained hydrocarbons, hydrogen
94
cyanide (HCN) and alkaloids (Brückner et al. 2017b; Brückner et al. 2015; Heethoff et al. 2018;
95
Raspotnig 2009; Saporito et al. 2007). While some chemicals appear to be alarm pheromones
96
(Raspotnig 2006; Shimano et al. 2002), most function as defensive allomones (Heethoff et al.
97
2011a). Interestingly, alkaloids produced by oribatids mites are the ultimate source of most
98
toxins sequestered by poison-frogs (Saporito et al. 2007; Saporito et al. 2009).
99
Terrestrial chelicerates predominately ingest fluid food. While phloem-feeding plant
100
pests like spider mites and ecotoparasites likes ticks adapted a sucking feeding mode, scorpions,
101
spiders and others use external, pre-oral digestion before ingestion by morphologically diverse
102
mouthparts (Bensoussan et al. 2016; Cohen 1995; Dunlop and Alberti 2008; Gulia-Nuss et al.
103
2016). Exceptions from this are Opiliones and sarcoptiform mites, i.e. oribatid and astigmatid
104
mites, all of which ingest solid food (Heethoff and Norton 2009; Norton 2007; Shultz 2007). In
105
general, oribatids feed on a wide range of different resources and show a low degree of dietary
106
specialization (Brückner et al. 2018b). The typical food spectrum of Oribatida, includes leaf-
107
litter, algae, fungi, lichens, nematodes, and small dead arthropods such as collembolans
108
(Heidemann et al. 2011; Riha 1951; Schneider and Maraun 2005; Schneider et al. 2004a;
109
Schneider et al. 2004b). In laboratory feeding trials, oribatid mites tend to prefer dark pigmented
110
fungi, but also fatty acid-rich plant-based food (Brückner et al. 2018b; Schneider and Maraun
111
2005). Additionally, stable-isotope analyses of
15
N and
13
C suggested that Oribatida are primary-
112
and secondary decomposers feeding on dead plant material and fungi, respectively (Maraun et al.
113
2011; Schneider et al. 2004a). The reasons for these preferences are still unknown, but they raise
114
the question of how oribatid mites are able to enzymatically digest the cell walls of plants and
115
.
CC-BY-NC-ND 4.0 International license
available under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprint
this version posted February 9, 2021.
;
https://doi.org/10.1101/2020.12.10.420141
doi:
bioRxiv preprint
5
fungi (Brückner et al. 2018a; Brückner et al. 2018b; Schneider et al. 2004b; Smrž and Čatská
116
2010).
117
Early studies on
Archegozetes
and other mites found evidence for cellulase, chitinase and
118
trehalase activity which was later attributed to symbiotic gut bacteria (Haq 1993; Luxton 1979;
119
Siepel and de Ruiter-Dijkman 1993; Smrž 2000; Smrž and Čatská 2010; Smrž and Norton 2004;
120
Zinkler 1971). While such bacterial symbionts are a possible explanation, genomic data of other
121
soil organisms and plant-feeding arthropods suggest a high frequency of horizontal transfer of
122
bacterial and fungal genes enabling the digestion of cell walls (Grbić et al. 2011; Mayer et al.
123
2011; McKenna et al. 2019; Wu et al. 2017; Wybouw et al. 2016; Wybouw et al. 2018). For
124
instance, an in-depth analysis of the spider mite
Tetranychus urticae
revealed a massive
125
incorporation of microbial genes into the mite’s genome (Grbić et al. 2011; Wybouw et al.
126
2018). Horizontal gene transfer appears to be a common mechanism for soil organisms,
127
including mites, to acquire novel metabolic enzymes (Dong et al. 2018; Faddeeva-Vakhrusheva
128
et al. 2016; Grbić et al. 2011; Hoffmann et al. 1998; Mayer et al. 2011; Wu et al. 2017), and
129
hence seems very likely for
Archegozetes
and other oribatid mite species that feed on plant or
130
fungal matter.
131
Archegozetes
has been established as a laboratory model organism for three decades,
132
having been used in studies, ranging from ecology, morphology, development and eco-
133
toxicology to physiology and biochemistry (Barnett and Thomas 2012; 2013a; 2013b; 2018;
134
Brückner et al. 2017a; Brückner et al. 2020; Heethoff et al. 2013). As such,
Archegozetes
is
135
among the few experimentally tractable soil organisms and by far the best-studied oribatid mite
136
species (Barnett and Thomas 2012; Heethoff et al. 2013; Thomas 2002). Since it meets the most
137
desirable requirements for model organisms (Thomas 2002), that is a rapid development under
138
laboratory conditions, a dedicated laboratory strain was named
Archegozetes longisetosus
ran
in
139
reference to its founder
R
oy
A
.
N
orton (Heethoff et al. 2013,
Figure 1b-c
). Their large number
140
of offspring enables mass cultures of hundreds of thousands of individuals, and their cuticular
141
transparency during juvenile stages, and weak sclerotization as adults are general assets of an
142
amenable model system (Brückner et al. 2018c; Brückner et al. 2016; Heethoff et al. 2013;
143
Heethoff and Raspotnig 2012). In the past 10 years,
Archegozetes
also received attention as a
144
model system for chemical ecology (Brückner and Heethoff 2018; Brückner et al. 2020;
145
.
CC-BY-NC-ND 4.0 International license
available under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprint
this version posted February 9, 2021.
;
https://doi.org/10.1101/2020.12.10.420141
doi:
bioRxiv preprint
6
Brückner et al. 2016; Heethoff and Rall 2015; Heethoff and Raspotnig 2012; Raspotnig et al.
146
2011; Thiel et al. 2018). Some of these studies focusing on the
Archegozetes
gland revealed
147
basic insights into the chemical ecology and biochemical capabilities of arthropods (Brückner et
148
al. 2020; Heethoff and Rall 2015; Thiel et al. 2018). Hence,
Archegozetes
is poised to become a
149
genetically tractable model to study the molecular basis of gland and metabolic biology.
150
The aim and focus of the current study were three-fold – to provide well-annotated, high-
151
quality genomic and transcriptomic resources for
Archegozetes longisetosus
(
Figure 1
), to reveal
152
possible horizontal gene transfers that could further explain the feeding biology of oribatids, and
153
to present
Archegozetes
as a research model for biochemical pathway evolution. Through a
154
combination of comparative genomic and detailed computational analyses, we were able to
155
generate a comprehensive genome of
Archegozetes
and provide it as an open resource for
156
genomic, developmental and evolutionary research. We further identified candidate horizontal
157
gene transfer events from bacteria and fungi that are mainly related to carbohydrate metabolism
158
and cellulose digestion, features correlated with the mite feeding biology. We also used the
159
genomic data together with stable-isotope labeling experiments and mass spectrometric
160
investigation to delineate the biosynthesis pathway of monoterpenes in oribatid mites.
161
162
Results and Discussion
163
Archegozetes longisetosus
genome assembly
164
Archegozetes longisetosus
(
Figure 1
) has a diploid chromosome number (2n) of 18
165
(Heethoff et al. 2006), most likely comprising 9 autosomal pairs, the typical number of nearly all
166
studied oribatid mite species (Norton et al. 1993). There are no distinct sex chromosomes in
167
Archegozetes
; this
appears to be ancestral in the Acariformes and persisted in the Oribatida
168
(Heethoff et al. 2006; Norton et al. 1993; Wrensch et al. 1994). Even though some XX:XO and
169
XX:XY genetic systems have been described in the closely related Astigmata, the sex
170
determination mechanism in oribatids, including
Archegozetes,
remains unknown (Heethoff et al.
171
2013; Heethoff et al. 2006; Norton et al. 1993; Oliver Jr 1983; Wrensch et al. 1994)
.
To provide
172
genetic resources, we sequenced and assembled the genome using both Illumina short-read and
173
Nanopore MinION long-read sequencing approaches (
Table 1
; see also “
Materials and
174
.
CC-BY-NC-ND 4.0 International license
available under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprint
this version posted February 9, 2021.
;
https://doi.org/10.1101/2020.12.10.420141
doi:
bioRxiv preprint
7
Methods
”). Analyses of the
k-mer
frequency distribution of short reads (
Table 1;
175
Supplementary Figure S1
) resulted in an estimated genome size range of 135-180 Mb, smaller
176
than the final assembled size of 190 Mb (
Table 1
; see also “
Materials and Methods
”). This
177
difference was suggestive of high repetitive content in the genome of
Archegozetes
and indeed,
178
repeat content was predicted to be 32 % of the genome (
see below
) (Alfsnes et al. 2017;
179
Simpson 2014). Compared to genome assemblies of other acariform mites, the assembled
180
genome size of
Archegozetes
is on the large end, but is smaller than that of mesostigmatid mites,
181
ticks and spiders (Bast et al. 2016; Dong et al. 2017; Dong et al. 2018; Grbić et al. 2011; Gulia-
182
Nuss et al. 2016; Hoy et al. 2016; Schwager et al. 2017). In the context of arthropods in general,
183
Archegozetes
’s genome (
Table 1
) is among the smaller ones and shares this feature with other
184
arthropod model species like the spider mite,
Drosophila
, clonal raider ant and red flour beetle
185
(Consortium 2008; dos Santos et al. 2015; Grbić et al. 2011; Oxley et al. 2014). Even though we
186
surface-washed the mites and only used specimens with empty alimentary tracts for sequencing,
187
we removed 438 contigs with high bacterial or fugal homology making up approximately 8.5 Mb
188
of contamination (
see supplementary Table S1
). The final filtered genome assembly was
189
composed of 1182 contigs with an N
50
contiguity of 994.5 kb (
Table 1
).
190
The official gene set and annotation of
Archegozetes
191
We generated the official gene set (OGS) for
Archegozetes
by an automated, multi-stage
192
process combining
ab inito
and evidenced-based (RNAseq reads, transcriptomic data and curated
193
protein sequences) gene prediction approaches (see “
Materials and Methods
”) yielding 23,825
194
gene models. In comparison to other mites and ticks as well as insects, this is well within the
195
range of the numbers discovered in other Chelicerata so far (
Figure 2a
). Chelicerates with a
196
large OSG, however, usually possess larger genomes (1-7 Gb), which suggests that
Archegozetes
197
may have a relatively dense distribution of protein-coding genes in its genome. On the other
198
hand, ticks can have giga-base sized genomes, but only a rather small number of gene models,
199
probably due to high repetitive content (Barrero et al. 2017; Gulia-Nuss et al. 2016; Palmer et al.
200
1994; Van Zee et al. 2007). Lacking more high-quality genomic resources of mites, it is thus not
201
clear whether the OGS of
Archegozetes
is the rule, or rather the exception within the Oribatida.
202
To compare if
Archegozetes’
OSG is similar to predicted genes of other oribatid mites as
203
well as Prostigmata and Astigmata, we first clustered genes by ortholog inference (OrthoFinder;
204
.
CC-BY-NC-ND 4.0 International license
available under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprint
this version posted February 9, 2021.
;
https://doi.org/10.1101/2020.12.10.420141
doi:
bioRxiv preprint
8
(Emms and Kelly 2015b)), removed species-specific genes and constructed a presence-absence
205
matrix of orthogroups to ordinate the data using non-metric multidimension scaling (NMDS,
206
Figure 2b
). Ordination revealed that the OGS of
Archegozetes
is well nested with other oribatid
207
mites and clearly separated from their closest relative the astigmatid mites as well as
208
prostigmatid mites (
Figure 2b
). As a first step in annotating the OSG, we ran KOALA (KEGG
209
Orthology And Links Annotation) to functionally characterize the genes (Kanehisa et al. 2016).
210
In total, 10,456 (43.9%) of all genes received annotation and about two thirds of all genes were
211
assigned either as metabolic genes (36%) or genes related to genetic information processing
212
(34%), while the remaining genes fell into different KEGG categories (
Figure 2c
). To further
213
annotate the genome, we followed the general workflow of funannotate with some modifications
214
(Palmer and Stajich 2017, see "Materials and Methods").
215
Overall, we found 15,236 genes (64%) of the OGS with homology to previously
216
published sequences (
Figure 2d
). For about half of all genes (51%), we were able to assign a full
217
annotation, 4% of all genes only showed homology to bioinformatically predicted proteins of
218
other species, while 9% of all genes only showed homology to hypothetical proteins (
Figure 2d
).
219
As only a few high-quality, annotated mite genomes are available and the two-spotted spider
220
mite is the sole species with any experimentally confirmed gene models, it is not surprising that
221
we were only able to confidently annotate about 55% of all genes of the OGS (
Figure 2d
).
222
Orthology and comparative genomics of chelicerates
223
To further access the protein-coding genes of the mite, we compared the OGS to other
224
chelicerates. Both concatenated maximum likelihood and coalescent species-tree phylogenomic
225
approaches based on 1,121 orthologs placed
Archegozetes,
as expected, within the Nothrina
226
(Heethoff et al. 2013; Pachl et al. 2012) with strong support and recovered previously found
227
oribatid clade topologies (
Figure 3a
). Our analysis placed the Astigmata as a sister group of
228
Oribatida and not nested within oribatids as suggested based on life-history, chemical defensive
229
secretions, morphology and several molecular studies (Alberti and Michalik 2004; Dabert et al.
230
2010; Domes et al. 2007; Klimov et al. 2018; Koller et al. 2012; Li and Xue 2019; Liana and
231
Witaliński 2005; Maraun et al. 2004; Norton 1994; 1998; Pepato and Klimov 2015; Sakata and
232
Norton 2001). The relationship of Oribatida and Astigmata has been challenging to resolve for
233
the past decades and several studies using different set of genes, ultra-conserved elements or
234
.
CC-BY-NC-ND 4.0 International license
available under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprint
this version posted February 9, 2021.
;
https://doi.org/10.1101/2020.12.10.420141
doi:
bioRxiv preprint
9
transcriptomic data reconstructed discordant phylogenies, some of which are similar to ours
235
(Dabert et al. 2010; Domes et al. 2007; Klimov et al. 2018; Li and Xue 2019; Lozano-Fernandez
236
et al. 2019; Maraun et al. 2004; Pepato and Klimov 2015; Van Dam et al. 2019). Overall, the
237
Oribatid-Astigmatid relationship remains unresolved and a broader taxon sampling, especially of
238
more basal Astigmata, will be necessary (Domes et al. 2007; Klimov et al. 2018; Lozano-
239
Fernandez et al. 2019; Norton 1994; 1998; Van Dam et al. 2019). We recovered Trombidiformes
240
(Prostigmata and Sphaerolichida) as sister group of the Sarcoptiformes (Oribatida and
241
Astigmata) constituting the Acariformes (
Figure 3a
). Neither the maximum likelihood
242
phylogeny (
Figure 3a
), nor the coalescence-based phylogeny (
Supplementary Figure S2
)
243
reconstructed the Acari (i.e. Acariformes and Parasitiformes) as a monophyletic taxon. Even
244
though there is morphological, ultrastructural and molecular evidence for a biphyletic Acari,
245
aswe recovered here, this relationship and larger-scale chelicerate relationships remain unclear
246
(Alberti 1984; 1991; Dabert 2006; Dunlop and Alberti 2008; Jeyaprakash and Hoy 2009; Li and
247
Xue 2019; Lozano-Fernandez et al. 2019; Van Dam et al. 2019).
248
To further assess the quality and homology of both the genome assembly (
Table 1
) and
249
the OGS (
Figure 2
), we used the1066 arthropod Benchmarking Universal Single-Copy Ortholog
250
(BUSCO) genes data set (Simão et al. 2015). Nearly all BUSCO genes were present in the
251
Archegozetes
assembly and OGS (96.2% and 97.3%, respectively;
Figure 3b
). Compared to
252
other genomes sequenced so far, the
Archegozetes
genome has the highest completeness among
253
oribatid mites and the OGS completeness is on par to the well curated genomes of other
254
chelicerate species and
Drosophila melanogaster
(
Figure 3b
). This result is not surprising
255
because the
Archegozetes
genome was assembled from long-read and short-read data, while all
256
other oribatid mite genomes were solely short reads sequenced on older Illumina platforms (Bast
257
et al. 2016). The fraction of duplicated BUSCO genes in
Archegozetes
(4%) was similar to that
258
of the spider mite and deer tick (Grbić et al. 2011; Gulia-Nuss et al. 2016), but very low
259
compared to the house spider (
Figure 3c
), whose genome underwent an ancient whole-genome
260
duplication (Schwager et al. 2017).
261
Overall, the high quality of both the genome assembly and OGS of
Archegozetes
262
compared to those of other oribatid mites, strongly indicates the importance of this genomic
263
resource. We next categorized all protein models from the OGS by conversation level based on a
264
.
CC-BY-NC-ND 4.0 International license
available under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprint
this version posted February 9, 2021.
;
https://doi.org/10.1101/2020.12.10.420141
doi:
bioRxiv preprint
10
global clustering orthology analysis (OrthoFinder; Emms and Kelly 2015b) of 23 species
265
(
Figure 3c; supplementary Figure S3
) representing Acariformes, Parasitiformes, several other
266
chelicerates and the fly
Drosophila
. As for most other species (Siepel et al. 2005; Thomas et al.
267
2020), about a third of all orthogroups was highly conserved (
Figure 3c
) across the arthropods,
268
being either in all species (10%;
Figure 3d
) or is most (22%;
Figure 3d
). Only 1% of all
269
Archegozetes
orthogroups did not show homology and were species specific (
Figure 3c and d
).
270
Only a low proportion (
Figure 3c
) of orthogroups was conserved across the higher taxonomic
271
levels (all <1% in
Archegozetes;
Figure 3d
), which is in line with previous studies that included
272
prostigmatid and mesostigmatid mites (Dong et al. 2017; Dong et al. 2018; Hoy et al. 2016).
273
Interestingly, there was a large proportion of orthogroups conserved across all Oribatida (43% in
274
Archegozetes;
Figure 3d
) and also about 19% of orthogroups in
Archegozetes
were shared only
275
with other Nothrina (
Figure 3d
). A fairly large percentage of these orthogroups may contain
276
potentially novel genes that await experimental verification and functional analyses (Emms and
277
Kelly 2015b; Nagy et al. 2020; Thomas et al. 2020). Especially the lack of homology within the
278
Sarcoptiformes (2-3%;
Figure 3c
) may explain the controversial placement of Astigmata as a
279
sistergroup of Oribatida that we recovered
(Figure 3a
). This grouping is likely caused by a long-
280
branch attraction artifact and the sister relationship was incorrectly inferred (Dabert 2006; Dabert
281
et al. 2010; Domes et al. 2007; Klimov et al. 2018; Pepato and Klimov 2015), because
282
orthogroup clustering could not detect enough homology between oribatids and the Astigmata so
283
far sequenced , which are highly derived. Hence, a broad taxon sampling of basal astigmatid mite
284
genomes seems necessary to resolve Oribatida-Astigmata relationship (Li and Xue 2019; Norton
285
1994; 1998; Pepato and Klimov 2015; Van Dam et al. 2019).
286
Repeat content analysis and transposable elements (TEs)
287
For clonal species like
Archegozetes,
reproducing in the absence of recombination, it has
288
been hypothesized that a reduced efficacy of selection could results in an accumulation of
289
deleterious mutations and repeats in the genome (Arkhipova and Meselson 2000; Barton 2010;
290
Charlesworth 2012; Muller 1964; Nuzhdin and Petrov 2003; Schön et al. 2009). There is,
291
however, no evidence for such an accumulation in oribatids or other arthropods (Bast et al.
292
2016). Generally, we found that most of the repetitive content in
Archegozetes
could not be
293
classified (57%;
Figure 4a
). The high proportion of unknown repeats likely corresponds to novel
294
.
CC-BY-NC-ND 4.0 International license
available under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprint
this version posted February 9, 2021.
;
https://doi.org/10.1101/2020.12.10.420141
doi:
bioRxiv preprint
11
predicted repetitive content, because of limited repeat annotation of mites in common repeat
295
databases such as RepBase (Bast et al. 2016). Regarding the two major classes of repeat content,
296
DNA transposons made up about 32% of total repeats, while only 5% represented
297
retrotransposons (
Figure 4a
). About 6% of total repetitive content comprised simple and low
298
complexity repeats (
Figure 4a
). Overall, the total repetitive content (32%,
Figure 4b
) seems to
299
be within a normal range for chelicerates and arthropods.
300
The repeat content found in other oribatid mites was lower (Bast et al. 2016), but recent
301
studies suggest that sequencing technology, read depth and assembly quality are paramount to
302
the capacity of identifying repeat content and TEs (Bourque et al. 2018; Panfilio et al. 2019).
303
Hence, it is very likely the current genomic data for other Oribatida underestimates the actual
304
total repetitive content. More low-coverage, long-read sequencing could reduce the assembly
305
fragmentation and likely reveal a higher proportion of repeats, closer to the actual repetitiveness
306
of oribatid genomes (Panfilio et al. 2019).
307
Different classes of transposable elements (TEs) are characterized by the mechanism they
308
use to spread within genomes and are known to influence population dynamics differently
309
(Bourque et al. 2018; Crescente et al. 2018; Finnegan 1989). We therefore analyzed the
310
evolutionary history of TE activity in
Archegozetes
in more detail
(
Figure 4c
). The main TE
311
superfamilies were
DNA transposons (
Figure 4a and c
), which seems to be a common pattern of
312
oribatid mite genomes. For
Archegozetes,
they appear to have accumulated in the genome for a
313
long time (i.e. they are more divergent from the consensus; (Waterston et al. 2002)) with
314
Tc1/mariner – a superfamily of interspersed repeats DNA transposons (Bourque et al. 2018) –
315
being the most abundant one (
Figure 4c
). Interestingly, we found an increase in TE activity with
316
0-3% sequence divergence range, indicating a recent burst (
Figure 4c
). This burst contained an
317
enrichment of DNA Mavericks, which are the largest and most complex DNA transposons with
318
homology to viral proteins (Bourque et al. 2018), but also several of retrotransposons. Among
319
these, is the Long Terminal Repeat (LTR) gypsy retroelement (
Figure 4c
), which is closely
320
related to retroviruses (Bourque et al. 2018). Like retroviruses, it encodes genes equivalent to
321
gag
,
pol
and
env
, but relatively little is known about how it inserts its DNA into the host genome
322
(Dej et al. 1998; Havecker et al. 2004). So far, it is unknown what these TEs do in
Archegozetes
,
323
.
CC-BY-NC-ND 4.0 International license
available under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprint
this version posted February 9, 2021.
;
https://doi.org/10.1101/2020.12.10.420141
doi:
bioRxiv preprint
12
but the recent burst in TE abundance might suggest that some changes in the genome might have
324
happened since the became a laboratory model nearly 30 years (Heethoff et al. 2013).
325
The
Archegozetes
Hox cluster
326
The Hox genes are a group of highly conserved transcription factor-encoding genes that
327
are used to pattern the antero-posterior axis in bilaterian metazoans (Holland and Hogan 1988;
328
Hrycaj and Wellik 2016). Ancestrally, arthropods likely had ten Hox genes arranged in a cluster
329
(Hughes and Kaufman 2002). During arthropod development, the Hox genes specify the
330
identities of the body segments, and mutations in Hox genes usually result in the transformation
331
of segmental identities (Hughes and Kaufman 2002). The importance of Hox genes in
332
development of metazoans makes knowledge of their duplication and disappearances important
333
for understanding their role in the evolution of body plans (Hughes and Kaufman 2002).
334
Mites largely lack overt, external signs of segmentation, other than the serially arranged
335
appendages of the prosoma (Dunlop and Lamsdell 2017). Signs of segmentation in the posterior
336
body tagma, the opisthosoma, do exist in adult members of Endeostigmata (van der Hammen
337
1970). However, these segmental boundaries are largely present only in the dorsal opisthosoma,
338
making it difficult to assess how these correspond to the ventral somites (Dunlop and Lamsdell
339
2017; van der Hammen 1970). Developmental genetic studies of the spider mite and
340
Archegozetes
suggest that acariform mites only pattern two segments in the posterior body
341
region, the opisthosoma, during embryogenesis (Barnett and Thomas 2012; 2013b; 2018; Grbić
342
et al. 2011). This stands in stark contrast to other studied chelicerate embryos. For example,
343
during embryogenesis the spider
Parasteatoda tepidariorum
patterns twelve opisthosomal
344
segments (Schwager et al. 2015) and the opilionid
Phalangium opilio
patterns seven (Sharma et
345
al. 2012). Furthermore, a member of Parasitiformes, the tick
Rhipicephalus microplus,
appears to
346
pattern eight opisthosomal segments during embryogenesis (Santos et al. 2013).
347
Parallel to the observation of segmental reduction in
T. urticae
, genomic evidence
348
suggests that this acariform mite has lost two of its Hox genes,
i.e.,
Hox3
and
abdominal-A
(abd-
349
A
) (Grbić et al. 2011). Interestingly, orthologs of
abd-A
in other studied arthropods pattern the
350
posterior segments as well. A genomic comparison of arthropod Hox clusters has also shown a
351
correlation between independent losses of
abd-A
and a reduction in posterior segmentation (Pace
352
et al. 2016). To investigate whether the loss of segmentation in
Archegozetes
is also due to an
353
.
CC-BY-NC-ND 4.0 International license
available under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprint
this version posted February 9, 2021.
;
https://doi.org/10.1101/2020.12.10.420141
doi:
bioRxiv preprint
13
absence in
abd-A
, we annotated its Hox cluster, paying close attention to the region between the
354
Hox genes
Ultrabithorax
(
Ubx
) and
Abdominal-B
(
Abd-B
), which is usually where this gene
355
resides in other arthropods (Hughes and Kaufman 2002). Our results suggest that the
356
Archegozetes
Hox genes are clustered in a contiguous sequence (tig00005200_pilon, total size
357
~7.5 Mbp) in the same order as suggested for the ancestral arthropod (Heethoff and Rall 2015).
358
Furthermore, we found no sequences suggestive of an
abd-A
ortholog in
Archegozetes
(
Figure
359
5a
). These data also support the findings of a previous PCR survey that retrieved no
abd-A
360
ortholog in
Archegozetes
(Cook et al. 2001). Genomic evidence from the Parasitiformes
Ixodes
361
scapularis
and
Metaseiulus occidentalis
reveal that these taxa maintain orthologs of all ten Hox
362
genes, however in
M. occidentalis
these genes are not clustered as they are in
I. scapularis
363
(Gulia-Nuss et al. 2016; Hoy et al. 2016).
364
Taken together, these observations suggest that the last common ancestor of acariform
365
mites likely lost its
abdominal-A
gene as well as experiencing a reduction in opisthosomal
366
segmentation (
Figure 5b
). Alternatively, these shared losses of
abd-A
may be due to
367
convergence due to similar selective pressures favoring a reduction in body size. The dorsal,
368
external segmentation of endeostigmatid mites does not necessarily contradict the hypothesis of a
369
loss of
abd-A
at the base of the acariform mites. As Hox genes are usually deployed after the
370
genetic establishment of segments in arthropods (Hughes and Kaufman 2002), the opisthosomal
371
segments in endeostigmatid mites may still develop in the absence of
abd-A
. However, this
372
hypothesis needs further testing with observations of segmental gene expression in
373
endeostigmatids as well as additional acariform species.
374
Life-stage specific RNA expression patterns
375
Developmental and gene expression data from
Archegozetes
embryos (
Figure 5 d
and
e
)
376
have elucidated many of the potential mechanisms driving the morphogenesis of
377
manydevelopmental peculiarities. These peculiarities include the suppression of the fourth pair
378
of walking legs during embryogenesis as well as the reduction of opisthosomal segmentation
379
(Barnett and Thomas 2012; 2013a; 2013b; 2018; Telford and Thomas 1998; Thomas 2002). In
380
typical acariform mites, embryogenesis ends with the first instar, the prelarva, which usually
381
remains within the egg chorion, as in
Archegozeetes
. Hatching releases the second instar, the
382
larva, which is followed by three nymphal instars (proto-, deutero- and tritonymph) and the
383
.
CC-BY-NC-ND 4.0 International license
available under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprint
this version posted February 9, 2021.
;
https://doi.org/10.1101/2020.12.10.420141
doi:
bioRxiv preprint
14
adult, for a total of six instars. (Heethoff et al. 2007). Thus far, methodological limitations have
384
made it difficult to examine how mite segmentation and limb development progress throughout
385
these instars.
386
To this end, we used RNAseq to calculate the transcripts per million (tpm) values of
387
genes known to be, or suspected to be, involved in limb development and segmentation
388
throughout the six different instars of
Archegozetes
. Prior to comparing these tpm values, gene
389
orthology was confirmed
via
phylogenetic analyses (
supplementary Figures S4-S11
; see
Table
390
S2
for phylogenetic statistics
and
Table S3
for tpm values). Regarding the total number of genes
391
expressed across the different life stages, we found that earlier instars generally expressed a
392
higher number of genes (
Figure 5c
). While most expressed genes were shared across all instars,
393
more transcripts were shared between the eggs and the larvae and among all five juvenile instars.
394
Additionally, we found that earlier instars expressed a larger number of stage-specific genes as
395
compared to later instars and adults (
Figure 5c
).
396
Gene expression, SEM and time-lapse data have revealed that the development of the
397
fourth pair of walking legs in
Archegozetes
is suppressed until after the larval instar (Barnett and
398
Thomas 2012; 2018; Telford and Thomas 1998). The resulting larva is thus hexapodal (see also
399
embryo in
Figure 5e
), which constitutes a putative synapomorphy of Acari, if they are
400
monophyletic (Dunlop and Alberti 2008). In arthropods, the development of the limbs is
401
generally accomplished via the activity of highly conserved regulatory genes, termed the “limb
402
gap genes.” These genes are expressed along their proximo-distal axes to establish the specific
403
identities of the limb podomeres. The limb gap genes include
extradenticle
(
exd
) and
404
homothorax
(
hth
), which act together to specify the proximal limb podomeres,
dachshund
(
dac
),
405
which specifies the medial podomeres, and
Distal-less
(
Dll
) which specifies the distal-most
406
podomeres. It was previously shown that the deployment of these genes in the anterior
407
appendages of
Archegozetes
,
i.e
., the chelicerae, pedipalps and first three pairs of walking legs
408
(
Figure 5d
and
e
), is similar to that of other chelicerate taxa (Barnett and Thomas 2013a;
409
Schwager et al. 2015; Sharma et al. 2015). However, in the anlagen of the fourth pair of walking
410
legs, only the proximal-specifying genes,
exd
and
hth
, are expressed (Barnett and Thomas
411
2013a).
412
.
CC-BY-NC-ND 4.0 International license
available under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprint
this version posted February 9, 2021.
;
https://doi.org/10.1101/2020.12.10.420141
doi:
bioRxiv preprint
15
Whether the limb gap genes are re-deployed during the transition from the prelarval to
413
larval instars in order to activate the development of the fourth pair of walking legs remains an
414
open question. We therefore compared the average tpm values of verified limb gap genes (i.e.,
415
Al-
Dll
, Al-
Hth
, Al-
exd
, and Al-
dac
(Barnett and Thomas 2013a)) in embryos and at each instar
416
stage (
Figure 5f
). We also compared the tpm values of the
Archegozetes
orthologs of
Sp6-9
and
417
optomotor blind
, genes shown to be involved in limb formation in spiders (Heingård et al. 2019;
418
Königsmann et al. 2017). We hypothesized that limb development genes would show high
419
expression in the larval stage leading to the development of the octopodal protonymph. We did
420
observe an increase in the tpm averages of Al
-hth
as well as Al-
optomotor-blind
, however the
421
aforementioned limb gap gene expression levels were similar between these instars (
Figure 5f
).
422
Taken together, these genes may not be up-regulated for the formation of the fourth pair of
423
walking legs between these two instars.
424
Chelicerate embryos segment their bodies through a “short/intermediate germ”
425
mechanism, whereby the anterior (prosomal) segments are specified asynchronously (Schwager
426
et al. 2015). This usually occurs well before the sequential addition of posterior segments from a
427
posterior growth zone. Based on neontological and paleontological data, chelicerate arthropods
428
may have ancestrally had an opisthosoma comprised of 12 or more segments (Dunlop and Selden
429
1998; Dunlop 2010; Dunlop and Lamsdell 2017). Embryonic expression data for the segment
430
polarity genes, those genes that delineate the boundaries of the final body segments, have shown
431
that in most studied chelicerate embryos opisthosomal segments are delineated during
432
embryogenesis (Dunlop and Lamsdell 2017; Schwager et al. 2015). However, as discussed
433
above, expression data in
Archegozetes
embryos suggest that only two opisthosomal segments
434
are patterned during embryogenesis (Barnett and Thomas 2012; 2018); this indicates that mites
435
have significantly reduced their number of opisthosomal segments either by loss or by fusion.
436
Further complicating this is the observation that many mites add segments as they progress
437
through the larval instars, a phenomenon known as anamorphic growth (Dunlop and Lamsdell
438
2017).
439
To determine by what genetic process
Archegozetes
may add segments during post-
440
embryonic ontogeny, we assessed the expression of known chelicerate and arthropod
441
segmentation genes in each instar transcriptome (
Figure 5f
) (Schwager et al. 2015). We
442
.
CC-BY-NC-ND 4.0 International license
available under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprint
this version posted February 9, 2021.
;
https://doi.org/10.1101/2020.12.10.420141
doi:
bioRxiv preprint
16
observed an up-regulation of the segmentation genes
hedgehog
and
engrailed
in the larvae, as
443
well as the slight up-regulation of
patched
and
pax3/7
. Furthermore, the segmentation gene
444
wingless
was slightly up-regulated in the protonymph, as well as a slight up-regulation of
445
hedgehog
in the tritonymph. Lastly, we found that transcripts of the genes
pax3/7
and
runt
were
446
up-regulated in adults. These results suggest that
Archegozetes
does pattern body segments
447
during the progression through the it’s instars similar to other Chelicerata.
448
Another peculiarity of
Archegozetes
is that these mites lack eyes (see more details
449
below). Eye loss has been documented in other arachnid clades, including independently in other
450
members of Acari (Evans 1992; Walter and Proctor 1999), and it has been recently demonstrated
451
that a species of whip spider has reduced its eyes by reducing the expression of retinal
452
determination genes that are shared throughout arthropods (Gainett et al. 2020). We sought to
453
determine if eye loss in
Archegozetes
also is associated with the reduced expression of these
454
genes (see also analysis of photoreceptor genes below). The genes, which have been shown to be
455
expressed in the developing eyes of spiders and whip scorpions, include
Pax-6
,
six1/sine oculis
456
(
so
),
eyes absent
(
eya
),
Eyegone
,
Six3
/
Optix
, and
atonal
(Gainett et al. 2020; Samadi et al. 2015;
457
Schomburg et al. 2015). We also followed the expression of Al-
orthodenticle
, a gene previously
458
shown to be expressed in the ocular segment of
Archegozetes
(Telford and Thomas 1998).
459
Surprisingly, all of these genes, excluding the
Pax-6
isoform A and
eyegone
, are indeed
460
expressed during embryogenesis (
Figure 5f
). Aside from the larval expression of the
Pax-6
461
isoform A during the larval stage, these eye-development genes remain quiescent until the adult
462
stage, where all but
Pax-6
isoform A,
six3
and
atonal
are up-regulated (
Figure 5f
). These results
463
are exceedingly surprising, given the conserved role of genes in retinal patterning. They suggest
464
a novel role for these genes, or alternatively, these expression patterns could be the result of early
465
expression of a retinal determination pathway followed by negative regulation by other genes to
466
suppress eye development.
467
Photoreceptor and chemosensory system of
Archegozetes longisetosus
468
Unlike insects and crustaceans, chelicerates do not have compounds eyes – with
469
horseshoe crab being an exception. Generally, mites are eyeless or possess one or two pairs of
470
simple ocelli (Alberti and Coons 1999; Alberti and Moreno-Twose 2012; Exner 1989; Harzsch et
471
al. 2006; Patten 1887). Ocelli are common in Prostigmata and Endeostigmata , among
472
.
CC-BY-NC-ND 4.0 International license
available under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprint
this version posted February 9, 2021.
;
https://doi.org/10.1101/2020.12.10.420141
doi:
bioRxiv preprint
17
Acariformes, as well Opilioacarida – the most likely sister group to the Parasitiformes – but are
473
absent in most Oribatida, Astigmata, Mesostigmata and ticks (Norton and Franklin 2018; Norton
474
and Fuangarworn 2015; Walter and Proctor 1998; Walter and Proctor 1999). This suggests that
475
the presence of eyes might be an ancestral condition for both Acariformes and Parasitiformes,
476
while more derived mites rely largely on chemical communication systems (Alberti and Coons
477
1999).
478
In oribatid mites, detailed morphological and ultrastructural investigations have
479
suggested that setiform sensilla are the most obvious sensory structures (
Figure 6a
) (Alberti
480
1998; Alberti and Coons 1999; Walter and Proctor 1999). The trichobothria are very complex,
481
highly modified (e.g. filiform, ciliate, pectinate, variously thickened or clubbed) no-pore setae
482
which are anchored in a cup-like base and likely serve as mechanosensory structures. In contrast,
483
the setal shafts of solenidia and eupathidia (
Figure 6a
) both possess pores (Alberti 1998; Alberti
484
and Coons 1999; Walter and Proctor 1999). Solenidia have transverse rows of small pores visible
485
under a light microscope and likely function in olfaction, while the eupathidia have one or
486
several terminal pores and likely are used as contact/gustatory sensilla (
Figure 6a
) (Alberti
487
1998; Alberti and Coons 1999). Previous work demonstrated that oribatid mites indeed use
488
olfactory signals in the context of chemical communication and food selection (Brückner et al.
489
2018a; Brückner et al. 2018b; Heethoff et al. 2011a; Heethoff and Raspotnig 2012; Raspotnig
490
2006; Shimano et al. 2002).
491
Interestingly, detailed morphological and ultrastructural studies showed that light-
492
sensitive organs exist in some Palaeosomata and Enarthronota (probably true eyes) as well as in
493
Brachypylina (the secondary lenticulus), representing lower and highly derived oribatid mites,
494
respectively (Alberti and Coons 1999; Alberti and Moreno-Twose 2012; Norton and Franklin
495
2018; Norton and Fuangarworn 2015).
Archegozetes
and most other oribatids, however, are
496
eyeless, yet there is scattered experimental and some anecdotal evidence that even these mites
497
show some response to light and seem to avoid it (‘negative phototropism’ or ‘negative
498
phototaxis’) (Madge 1965; Trägårdh 1933; Walter and Proctor 1999; Woodring 1966). Hence,
499
we mined the genome of
Archegozetes
for potential photoreceptor genes and found two genes of
500
the
all-trans retinal
peropsin
class and one gene related to spider mite
rhodopsin-7-like
gene
501
(
Figure 6b
).
Peropsin-like
genes are also present in other eyeless ticks. In jumping spiders it
502
.
CC-BY-NC-ND 4.0 International license
available under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprint
this version posted February 9, 2021.
;
https://doi.org/10.1101/2020.12.10.420141
doi:
bioRxiv preprint
18
encodes for nonvisual, photosensitive pigments, while
rhodopsin-7
may be involved in basic
503
insect circadian photoreception (Eriksson et al. 2013; Koyanagi et al. 2008; Nagata et al. 2010;
504
Senthilan et al. 2019; Senthilan and Helfrich-Förster 2016; Shen et al. 2011). Taken together, this
505
might suggest that eyeless species like
Archegozetes
use
peropsin
- and
rhodopsin-7-like
genes
506
for reproductive and diapause behaviors, or to maintain their circadian rhythm, as well as
507
negative phototaxis.
508
However, the main sensory modality soil mites use is chemical communication via
509
olfaction (Alberti 1998; Alberti and Coons 1999; Brückner et al. 2018a; Brückner et al. 2018b;
510
Raspotnig 2006; Shen et al. 2011; Walter and Proctor 1999). In contrast to insects, but similar to
511
crustaceans and Myriapoda, mites do not have the full repertoire of chemosensory classes, they
512
are missing odorant receptors and odorant-binding proteins (
Table 2
) (Dong et al. 2017; Dong et
513
al. 2018; Hoy et al. 2016; Maraun et al. 2007; Raspotnig 2009; Sánchez-Gracia et al. 2009;
514
Sánchez‐Gracia et al. 2011; Vieira and Rozas 2011). Although chemosensory protein (CSP)
515
encoding genes are absent in most mite genomes, we identified one gene encoding for such a
516
protein in
Archegozetes
and one CSP has been previously found in the deer tick (Table 2). Hence,
517
Archegozetes
should primarily rely on gustatory receptors (GRs) and ionotropic receptors (IRs).
518
Both the number of GRs (68 genes;
Figure 6d
) and IRs (3 genes;
Figure 6c
) was very well within
519
the range of most mites and ticks and there was no evidence for any massive chemoreceptor
520
expansion like in the spider mite (
Table 2
) (Ngoc et al. 2016). This was surprising because
521
Archegozetes
, like other acariform mites have many multiporous solenidia, present on all legs and
522
the palp, but appear to only have a limited number of chemoreceptors.
523
Canonical ionotropic glutamate receptors (iGluRs) are glutamate-gated ion channels with
524
no direct role in chemosensation, which come in two major subtypes: either NMDA iGluRs
525
which are sensitive to N-methyl-D-aspartic acid (NMDA) or non-NMDA iGluRs. The latter
526
group – at least in
Drosophila
– seems to have essential functions in synaptic transmission in the
527
nervous system and have been associated with sleep and vision (Benton et al. 2009; Croset et al.
528
2010; Ngoc et al. 2016; Sánchez-Gracia et al. 2009; Sánchez‐Gracia et al. 2011). None of the IRs
529
we found in the
Archegozetes
genome belonged to the NMDA iGluRs and most were classified
530
as non-NMDA iGluRs (
Figure 6c
). Nothing is known about their functions in mites. It is,
531
however, likely that they perform similar tasks in synaptic transmission in the brain and
532
.
CC-BY-NC-ND 4.0 International license
available under a
(which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made
The copyright holder for this preprint
this version posted February 9, 2021.
;
https://doi.org/10.1101/2020.12.10.420141
doi:
bioRxiv preprint