Spindle-shaped viruses infect marine ammonia-oxidizing thaumarchaea

Abstract

Ammonia-oxidizing archaea (AOA) from the phylum Thaumarchaeota are ubiquitous in marine ecosystems and play a prominent role in carbon and nitrogen cycling. Previous studies have suggested that, like all microbes, thaumarchaea are infected by viruses and that viral predation has a profound impact on thaumarchaeal functioning and mortality, thereby regulating global biogeochemical cycles. However, not a single virus capable of infecting thaumarchaea has been reported thus far. Here we describe the isolation and characterization of three Nitrosopumilus spindle-shaped viruses (NSVs) that infect AOA and are distinct from other known marine viruses. Although NSVs have a narrow host range, they efficiently infect autochthonous Nitrosopumilus strains and display high rates of adsorption to their host cells. The NSVs have linear double-stranded DNA genomes of ∼28 kb that do not display appreciable sequence similarity to genomes of other known archaeal or bacterial viruses and could be considered as representatives of a new virus family, the “Thaspiviridae.” Upon infection, NSV replication leads to inhibition of AOA growth, accompanied by severe reduction in the rate of ammonia oxidation and nitrite reduction. Nevertheless, unlike in the case of lytic bacteriophages, NSV propagation is not associated with detectable degradation of the host chromosome or a decrease in cell counts. The broad distribution of NSVs in AOA-dominated marine environments suggests that NSV predation might regulate the diversity and dynamics of AOA communities. Collectively, our results shed light on the diversity, evolution, and potential impact of the virosphere associated with ecologically important mesophilic archaea.

Full text

Spindle-shaped viruses infect marine ammonia-
oxidizing thaumarchaea
Jong-Geol Kim
a
, So-Jeong Kim
b
, Virginija Cvirkaite-Krupovic
c
, Woon-Jong Yu
a
, Joo-Han Gwak
a
, Mario López-Pérez
d
,
Francisco Rodriguez-Valera
d
, Mart Krupovic
c
, Jang-Cheon Cho
e
, and Sung-Keun Rhee
a,1
a
Department of Microbiology, Chungbuk National University, Heungduk-gu, 361-763 Cheongju, South Korea;
b
Geologic Environment Research Division,
Korea Institute of Geoscience and Mineral Resources, 34132 Daejeon, Republic of Korea;
c
Department of Microbiology, Institut Pasteur, 75015 Paris, France;
d
Evolutionary Genomics Group, Universidad Miguel Hernandez, San Juan, 03540 Alicante, Spain; and
e
Department of Biological Sciences, Inha University,
22212 Incheon, Republic of Korea
Edited by Edward F. DeLong, University of Hawaii at Manoa, Honolulu, HI, and approved June 21, 2019 (received for review April 3, 2019)
Ammonia-oxidizing archaea (AOA) from the phylum Thaumarchaeota
are ubiquitous in marine ecosystems and play a prominent role in
carbon and nitrogen cycling. Previous studies have suggested that,
like all microbes, thaumarchaea are infected by viruses and that viral
predation has a profound impact on thaumarchaeal functioning and
mortality, thereby regulating global biogeochemical cycles. However,
not a single virus capable of infecting thaumarchaea has been
reported thus far. Here we describe the isolation and characterization
of three Nitrosopumilus spindle-shaped viruses (NSVs) that infect
AOA and are distinct from other known marine viruses. Although
NSVs have a narrow host range, they efficiently infect autochtho-
nous Nitrosopumilus strains and display high rates of adsorption
to their host cells. The NSVs have linear double-stranded DNA ge-
nomes of ∼28 kb that do not display appreciable sequence simi-
larity to genomes of other known archaeal or bacterial viruses and
could be considered as representatives of a new virus family, the
“Thaspiviridae.”Upon infection, NSV replication leads to inhibition
of AOA growth, accompanied by severe reduction in the rate of am-
monia oxidation and nitrite reduction. Nevertheless, unlike in the case
of lytic bacteriophages, NSV propagation is not associated with de-
tectable degradation of the host chromosome or a decrease in cell
counts. The broad distribution of NSVs in AOA-dominated marine
environments suggests that NSV predation might regulate the diver-
sity and dynamics of AOA communities. Collectively, our results
shed light on the diversity, evolution, and potential impact of the
virosphere associated with ecologically important mesophilic archaea.
spindle-shaped virus
|
ammonia-oxidizing archaea
|
viral predation
|
chronic infection
Members of the phylum Thaumarchaeota are widespread
and abundant in marine ecosystems and play key roles in
nitrogen cycles by mediating ammonia oxidation (1, 2). Ammonia
oxidation is implicated in controlling the availability of nitrogen
species, production of N
2
O (3, 4), and is associated with carbon
fixation in the deep ocean. Thus, information on key factors af-
fecting abundance and composition of the communities of
ammonia-oxidizing archaea (AOA) is crucial for understanding
the biogeochemical processes of nitrogen cycling in the oceans.
The relative contribution of resource competition (bottom-up)
and predation (top-down control) are the key drivers of bio-
geochemical cycles, affecting microbial activity and community
structures. To understand the abundance and composition of the
AOA communities, the effects of physicochemical factors and
metabolic traits of AOA ecotypes on the efficiency of resource
utilization have been thoroughly assessed (2, 5).
Predation pressure can also influence AOA communities but
has been rarely studied. Flagellate grazing was proposed to affect
the distribution and abundance of AOA in planktonic microbial
assemblages (6, 7). Danovaro et al. (8) suggested that viral in-
fection represents a key mechanism controlling the turnover of
archaea, especially AOA, in surface deep-sea sediments. Putative
thaumarchaeal proviruses related to tailed bacterial and archaeal
viruses of the order Caudovirales have been previously identified in
the genomes of the soil thaumarchaeon Nitrososphaera viennensis
(9) and the extremely thermophilic thaumarchaeon Candidatus
Nitrosocaldus cavascurensis (10). Furthermore, several meta-
genomic and single-cell genomic studies have resulted in the as-
sembly of putative AOA virus genomes, all related to members of
the order Caudovirales (11–13). Notably, some of these assembled
virus genomes were found to carry putative genes encoding the
ammonia monooxygenase subunit C (amoC), a key component of
ammonia monooxygenase (AMO) (13, 14), suggesting an active role
of viruses in nitrogen cycling in the oceans. Nevertheless, not a single
thaumarchaeal virus–host system has been isolated or cultivated thus
far, precluding functional studies on the virus–host interactions and
the effect of viruses on the metabolic activity of thaumarchaea.
Archaea are associated with a remarkably diverse virosphere,
which is characterized by unique morphotypes not observed
among viruses infecting Bacteria and Eukarya. These include
virions with spindle-shaped, bottle-shaped, droplet-shaped, coil-
shaped, and other morphologies (15–18). Among these archaea-
specific morphotypes, spindle-shaped viruses are among the
most widely distributed (19) and were found not only in extreme
geothermal and hypersaline habitats, but also in marine environments,
Significance
Ammonia-oxidizing archaea (AOA) are major players in global
nitrogen cycling. The physicochemical and metabolic factors af-
fecting the composition of AOA communities and their efficiency
of resource utilization have been studied extensively. However,
viral predation on AOA remains unexplored due to lack of iso-
lated virus–host systems. Here we report on the isolation and
characterization of three Nitrosopumilus spindle-shaped viruses
(NSVs) that infect AOA hosts. NSVs represent a potentially im-
portant group of marine viruses with a chronic infection cycle,
providing important insights into the diversity and evolution of
the archaeal virosphere. The wide spread of NSVs in AOA-
containing marine environments suggests that NSV predation
might regulate the diversity and dynamics of AOA communities,
thereby affecting the carbon and nitrogen cycling.
Author contributions: J.-G.K. and S.-K.R. designed research; J.-G.K., V.C.-K., W.-J.Y., and
J.-H.G. performed research; S.-J.K. and M.L.-P. contributed new reagents/analytic tools;
S.-J.K., V.C.-K., M.L.-P., F.R.-V., M.K., and J.-C.C. analyzed data; and J.-G.K., F.R.-V., M.K.,
and S.-K.R. wrote the paper.
The authors declare no conflict of interest.
This article is a PNAS Direct Submission.
Published under the PNAS license.
Data deposition: DNA sequencing data have been deposited in the GenBank database
with the identifiers MK570053 to MK570059. Accession number of the Nitrosopumilus
strain SW genome is CP035425.
1
To whom correspondence may be addressed. Email: rhees@chungbuk.ac.kr.
This article contains supporting information online at www.pnas.org/lookup/suppl/doi:10.
1073/pnas.1905682116/-/DCSupplemental.
Published online July 16, 2019.
www.pnas.org/cgi/doi/10.1073/pnas.1905682116 PNAS
|
July 30, 2019
|
vol. 116
|
no. 31
|
15645–15650
MICROBIOLOGY

including coral surfaces (20), particulate matter-rich bays (21), and the
oceanic basement (22), although their hosts and genome sequences
were not determined.
In this study, we isolated and characterized spindle-shaped
viruses infecting AOA from coastal seawater and revealed
properties of their life cycles. We show that virus infection has a
dramatic effect on ammonia oxidation and is likely to affect the
population structure and functioning of the AOA community. Our
results shed light on the diversity, evolution, and potential impact of
the virosphere associated with ecologically important archaea.
Results and Discussion
Isolation, Morphology, and Stability. Three virus strains, designated
Nitrosopumilus spindle-shaped viruses 1, 2, and 3 (NSV1, NSV2,
and NSV3, respectively), were isolated from suspended partic-
ulate matter (SPM)-rich seawater samples taken from the west-
ern coast of the Korean Peninsula using as a host the axenic
AOA strain SW (SI Appendix, Fig. S1), which was isolated from
surface water (20 m deep) at the Yellow Sea, Korea (23) (see
below for further information). Since AOA could not form lawns
on agar plates, the dilution-to-extinction method was used to
isolate the viruses from enrichment cultures. General features of
NSVs are summarized in Table 1. The sizes of the spindle-
shaped NSV virions were similar (Fig. 1 and SI Appendix, Fig.
S2), measuring 64 ±3 nm in diameter and 112 ±6 nm in length,
with a short tail at one pole (Fig. 1A). The morphological fea-
tures of NSVs were very similar to those of viruses in the family
Fuselloviridae and in the genus Salterprovirus, which infect
hyperthermophilic and hyperhalophilic archaea, respectively (19,
24, 25). A large fraction of produced virions remained at-
tached to the cell surface (Fig. 1B). Elongation of virions into
arrowhead-shaped particles with long tails was observed 6 h post
infection (Fig. 1C), suggesting that flexibility of virions might be
important to the infection process—as has been observed for
spindle-shaped virions of the fusellovirus SSV1 (26) and the
bicaudavirus ATV (27). The adsorption of NSVs to AOA cells
was rapid, with ∼50% of virions bound to cell surfaces within 10
min (SI Appendix, Fig. S3). Certain hyperthermophilic archaeal
viruses exhibit comparably rapid adsorption kinetics (28); con-
versely, halophilic archaeal viruses—including those with spindle-
shaped virions—generally exhibit slow adsorption kinetics (29).
Spindle-shaped viruses are frequently observed in extreme
environments, and it is suggested that this morphotype has been
selected for its robustness under a wide range of extreme envi-
ronmental conditions (24, 30, 31). To study how the isolated
NSV virions respond to physicochemical fluctuations in the en-
vironment, their stabilities were tested under varying regimes of
pH, salinity, and temperature. NSV virions remained stable and
infectious between pH 3 and 9, salinities between 0.1 and 20%,
and temperatures up to 55 °C (SI Appendix, Fig. S4). These re-
sults showed that NSV virions were well-adapted to both survive
environmental fluctuations and interact with the unique archaeal
cell surface, which consists of a cytoplasmic membrane and
proteinaceous S-layer (32–35). Notably, mesophilic AOA are
believed to have evolved from a (hyper)thermophilic ancestor
(36, 37). Thus, the observed resilience of the NSV particles could
have been inherited from an ancestral extremophilic virus.
Host Specificity. Based on the comparison of average nucleotide
identities (ANIs), strain SW belongs to the genus Nitrosopumilus,
but represents a species that is most closely related to Nitro-
sopumilus maritimus SCM1 (SI Appendix, Fig. S5). Thus, host
specificities of the three NSVs were tested using strains of
Nitrosopumilus species closely related to the strain SW (>98%
sequence similarity of the 16S rRNA gene; SI Appendix, Fig. S1)
SCM1 (33), DDS1 (23), HCA1 (33), and BC. In the presence of
NSVs, neither virus production nor inhibition of ammonia oxi-
dation by these AOA strains was observed, indicating that the
host range of NSVs might be rather narrow. The sensitivity of
closely related strains to viral infection could be potentially af-
fected by the presence of host defense mechanisms or a lack of
specific receptors on the cell surface. To date, there is no report
of a CRISPR-Cas–dependent viral defense system in the thau-
marchaeal group I.1a (38). Similarly, homologs of the Dnd de-
fense system could not be identified in the available genomes of
AOA strains (i.e., strains SCM1, DDS1, and SW) (39) used in this
study. These findings suggest that resistance might instead be due
to the absence or modification of cell surface receptors. Notably,
comparison of the closely related Nitrosopumilus maritimus SCM1
and Nitrosopumilus sp. SW genomes revealed that genomic island 2
predictedtobeinvolvedincellsurface modification had different
gene content in the two strains, which might explain the different
susceptibility to NSVs (SI Appendix,Fig.S6and Dataset S1).
Table 1. General features of NSVs and scaffolds related to NSV
Isolation site
Virus/putative
viral
scaffolds*
Adsorption
rate, 50% (min)
Latent
period (h)
Attached
fraction (%)
Genome
size (kb)
Number
of ORFs G+C mol%
AOA-like
genes
Accession
number
Bulcheon
(36°57′N, 126°20′E)
NSV1 5 6 69 27.5 48 29.8 5 MK570053
NSV2 10 6–8 80 28.9 51 29.8 4 MK570055
Scaffold83 ———14.6 30 27.1 1 MK570056
Scaffold98 ———13.4 20 31.3 1 MK570057
Scaffold261 ———7.1 18 29.7 0 MK570058
Scaffold342 ———6.0 13 27.2 2 MK570059
Daecheon
(36°58′N, 126°20′E)
NSV3 <5 6 70 27.5 48 29.8 5 MK570054
—, not applicable.
*Scaffold is obtained from early phase enrichment culture for NSV2.
AB C
Fig. 1. Transmission electron microscopy images of negatively stained NSV1
virions. (A) NSV1 virions. (Scale bar, 50 nm.) (B) NSV1 particles attached to
the surface of an SW cell. (Scale bar, 200 nm.) (C) Elongated NSV1 particles
attached to the surface of an SW cell after 12 h of infection. The arrows
indicate elongated NSV1 particles. (Scale bar, 200 nm.)
15646
|
www.pnas.org/cgi/doi/10.1073/pnas.1905682116 Kim et al.

Genome Analysis. The genomes of NSV1, NSV2, and NSV3
consist of linear dsDNA molecules of ∼27, ∼29, and ∼27 kb,
respectively, containing 176-bp terminal inverted repeats (Table 1).
The numbers of ORFs in the genomes of NSV1, NSV2, and
NSV3 were predicted to be 48, 51, and 48, respectively. The
NSV1 and NSV3 genomes were highly similar (ANI =99.8%;
Fig. 2), whereas the NSV2 genome was slightly more divergent
(ANI =95%; Fig. 2). Divergent partial NSV genomes were also
obtained as scaffolds 83, −98, −261, and −342 from meta-
genomes of the initial NSV2 enrichment culture, indicating a
greater diversity of NSVs (Fig. 2). Despite being closely related
to each other, NSVs did not display appreciable sequence simi-
larity to other known archaeal or bacterial viruses in BLASTP
searches (E value cutoff: 0.001; Dataset S2). Indeed, none of the
NSV ORFs gave BLAST best-hits to known viral proteins, and
only 9 out of 48 (18.7%) NSV1 ORFs yielded significant matches
in the nonredundant sequence databases to known cellular
proteins—a common trend in archaeal viruses (15). Five of the
nine hits were to proteins encoded by various marine thau-
marchaea, and four were to bacterial proteins (Dataset S2). The
thaumarchaea-like ORFs encode putative genome replication
proteins (ORF3 and ORF43), glycosyltransferases (ORF20 and
ORF39), and a DNA-binding protein (ORF44) (see below).
Accordingly, NSVs might be considered as representatives of a
new archaeal virus family with the proposed name “Thaspiviridae”
(for Thaumarchaeal spindle-shaped viruses).
More sensitive sequence analysis based on profile hidden
Markov model (HMM) comparisons allowed for functional an-
notation of 15 putative NSV1 genes (Dataset S2). ORF3 and
ORF43, respectively, encode predicted protein-primed family B
DNA polymerase (pPolB) and a DNA sliding clamp known as
proliferating cell nuclear antigen (PCNA) that are likely to be
involved in NSV genome replication. ORF15 encodes a pre-
dicted Cdc6-like AAA+ATPase that may also be involved in
genome replication. However, given the broad functional di-
versity of AAA+ATPases, involvement of ORF15 in other steps
of the infection cycle cannot be ruled out. Similarly to many
other archaeal viruses (40), NSV1 encodes two predicted gly-
cosyltransferases (ORF20 and ORF39), with the closest homo-
logs present in thaumarchaeal genomes, and one predicted DNA
methyltransferase (ORF33)—apparently recruited from bacteria
by horizontal gene transfer. ORFs 6, 7, 29, 31, 32, and 37 encode
short proteins with predicted zinc-binding domains; ORFs 18, 26,
and 44 encode putative DNA-binding proteins with winged
helix-turn-helix, looped-hinge helix, and ribbon-helix-helix do-
mains, respectively (Dataset S2). To gain further understanding on
the proteins encoded by NSVs, purified NSV1 virions were
subjected to proteomic characterization by liquid chromatography
coupled to tandem mass spectrometry (LC-MS/MS). Ten NSV1
proteins were detected by proteomic analysis (Dataset S3).
Evolutionary Relationships to Other Archaeal Viruses. Despite the
lack of direct sequence similarity, general features of the genome
organization and proteome of NSV1 are reminiscent of those of
other archaeal viruses. In particular, the pPolB of NSVs is shared
with several groups of archaeal viruses and nonviral mobile ge-
netic elements, which, like NSV, have linear genomes with terminal
inverted repeats. These include the haloarchaeal spindle-shaped
(genus Salterprovirus) and pleomorphic (family Pleolipoviridae,ge-
nus Gammapleolipovirus) viruses His1 (41) and His2 (30), re-
spectively; hyperthermophilic bottle-shaped (family Ampullaviridae)
(25) and ellipsoid (family Ovaliviridae) (42) viruses; and casposons,
which integrate into the genomes of diverse thaumarchaea (43, 44).
To understand the relationship between NSVs and other pPolB-
encoding archaeal and bacterial mobile genetic elements, a maxi-
mum likelihood phylogenetic analysis of their respective pPolB
sequences was performed. When midpoint rooted, a well-supported
phylogeny splits between bacterial and archaeal sequences, with the
only exception being the ovalivirus SEV1, which groups with bac-
terial rather than archaeal homologs (Fig. 3A). The pPolB se-
quences from NSVs form a sister group to the clade that includes
halophilic viruses His1 and His2, as well as sequences from marine
Fig. 2. Comparative genomics of NSVs. Genomic maps of NSVs and related scaffolds. Shared ORFs are connected by color-coded shaded areas based on
sequence identity. Genomes of NSVs are flanked by terminal inverted repeats. %GC represents mol% G+C content of DNA.
Kim et al. PNAS
|
July 30, 2019
|
vol. 116
|
no. 31
|
15647
MICROBIOLOGY

sediment metagenomes. At the base of this clade are casposons
from the marine Thaumarchaeota.Interestingly,G+Cmol%values
of His1 (39%) and His2 (40%) are different from that of their host,
Haloarcula hispanica (63%), but close to those of AOA (∼34%).
Collectively, these results suggest horizontal exchange of the pPolB
genes between casposons, NSVs, and haloarchaeal viruses. Fur-
thermore, phylogenetic analysis suggests that pPolB genes of
thaumarchaeal mobile elements are ancestral to those of His1-like
viruses of halophilic archaea. However, many more viral genomes
of both His1-like and NSV-like viruses are needed to substantiate
this hypothesis.
NSVs and the salterprovirus His1 are the only known spindle-
shaped viruses with linear dsDNA genomes carrying pPolB genes,
suggesting a specific evolutionary connection between the two virus
groups. Notably, despite the lack of detectable sequence similarity,
the two viruses possess a similar gene repertoire (Fig. 3B), including
many genes encoding zinc-binding proteins, AAA+ATPase, gly-
cosyltransferases, and a putative terminal protein which is found
by LC-MS/MS in the virions of both NSV1 (Dataset S3) and His1
(45). Furthermore, all known spindle-shaped viruses—including
His1—encode relatively short (70–140 aa) major capsid proteins
containing two highly hydrophobic α-helical regions predicted to
form transmembrane domains (19), but with little sequence sim-
ilarity to each other. Among NSV1 proteins, only one, encoded by
ORF12, fits these characteristics (SI Appendix,Fig.S7) and was
detected in the virions by LC-MS/MS (Dataset S3). Based on these
shared properties, NSVs appear to be distantly related to His1 and,
more generally, to other spindle-shaped archaeal viruses.
All previously characterized archaeal viruses with unique vi-
rion morphologies not observed among viruses of bacteria or
eukaryotes infect extremophilic hosts (15, 25), suggesting that
these archaea-specific morphotypes have evolved as an adap-
tation to extreme environments. The fact that genomes of
thaumarchaeal viruses previously discovered by metagenomics
are all related to those of tailed bacteriophages (11–13) is con-
sistent with this possibility. However, the identification of spindle-
shaped viruses infecting mesophilic marine thaumarchaea strongly
suggests that the spread of unique archaeal morphotypes extends to
mesophilic archaea and possibly encompasses other archaeal line-
ages. Furthermore, these results suggest that spindle-shaped viruses
have a deep evolutionary history within the domain Archaea, which
likely dates back to the last archaeal common ancestor.
Viral Impact on Host Metabolism. The effects of NSVs on the
physiology and metabolism of their AOA host were examined
next. During the first 2 d post infection (dpi), viral DNA repli-
cation occurred concurrently with AOA growth and was ac-
companied by a normal rate of ammonia oxidation (Fig. 4 and SI
Appendix, Fig. S8). However, host cell growth ceased 2 dpi, while
ammonia oxidation continued until 4 dpi at a rate similar to that
in the uninfected AOA (Fig. 4 and SI Appendix, Fig. S8). Pre-
sumably, during this period, the energy obtained from ammonia
oxidation was directed to virus replication, consistent with the
observed increase in the viral titer until 6 dpi. However, after 5
dpi, the rate of ammonia oxidation and nitrite production de-
creased dramatically, indicating that NSVs had a severe effect on
metabolic activity in their AOA hosts. Notably, virus production
was not associated with detectable degradation of the host
chromosome determined by quantification of archaeal 16S
rRNA gene (Fig. 4B) or a decrease in cell counts measured using
epifluorescence microscopy (SI Appendix, Fig. S9). Consistent
with this observation, cells with damaged cell envelopes, as ob-
served for some lytic archaeal viruses (46), were not detectable by
transmission electron microscopy (TEM). Instead, virions were
observed in abundance on the cell surface without obvious asso-
ciated perturbations (Fig. 1B), suggesting that viral replication and
release did not lead to cell lysis. It has been previously shown that
spindle-shaped viruses of hyperthermophilic and halophilic ar-
chaea are also released from their hosts without causing cell lysis
(26, 47). In the case of Sulfolobus spindle-shaped virus SSV1, vi-
rions are assembled during the budding of the viral nucleoprotein
through the cell membrane, which remains intact, in a process
highly similar to the egress of enveloped eukaryotic viruses (26).
We hypothesize that NSV virions are assembled and released
from the cell by a similar mechanism without lysing the host cells.
As suggested by the TEM analysis (Fig. 1B) and the adsorp-
tion assays (SI Appendix, Fig. S3), high proportions (>60%) of
the produced NSVs remained cell-associated, presumably both
as adsorbed virions on the host cell surface and as intracellular
replicated genome copies (Fig. 4Cand SI Appendix, Fig. S8 C
and D). In resource-poor oceans, a nonlytic mode of replication
(phi29-like phages)
(monoderm hosts)
(diderm hosts)
"Autolykiviridae"
ATY46514.1_Sulfolobus ellipsoi d virus 1
MGYP000338936204
MGYP000401327420
candidate archaeal division MSBL1
metagenomic sequences
Ampullaviridae
Thaumarchaeal casposons (marine)
NSV1_gp3
NSV2_gp4
NSV3_gp3
scaffold98_gp4
scaffold83_gp11
Ovaliviridae
YP_529524.1_His1
YP_529644.1_His2
MGYP000367844654
MGYP000577203607
MGYP000041105224
MGYP000220981334
Marine
sediment
metagenome
A
B
Nitrosopumilus
spindle-
-shaped viruses
Gamma-
pleolipovirus
Salterprovirus
Picovirinae
Tectiviridae
Tectiviridae
94
63
96
97
88
100
99
100 100
100
100
100
100
100 100
100
100
100
100
100
99
99
99
99
99
99
92
Fig. 3. Phylogeny of pPolB and comparative genome maps of NSV1 and
His1. (A) Maximum likelihood phylogenetic analysis of pPolB sequences from
bacterial and archaeal viruses. In this tree, archaeal viruses comprise the
following taxa: Ampullaviridae,Pleolipoviridae,Ovaliviridae,andSalterprovirus.
Sequences originating from marine environments and hyperhalophilic
archaeal viruses are highlighted with light blue and green backgrounds,
respectively. Sequences from metagenomic datasets are indicated with
gray font. (B) Comparison of the NSV1 and His1 genome maps. Function-
ally equivalent genes are indicated with matching colors. Genes encoding
proteins detected in the purified virus particles are shown in cyan. Genes
encoding small proteins containing Zn-binding domains are shown in
green. Abbreviations: pPolB, protein-primed family B DNA polymerase; MCP,
major capsid protein (putative); wHTH, winged helix-turn-helix; GTase, gly-
cosyltransferase; MTase, DNA methyltransferase; PCNA, proliferating cell
nuclear antigen; RHH, ribbon-helix-helix. The question mark next to the
putative MCP denotes the uncertainty of this prediction.
15648
|
www.pnas.org/cgi/doi/10.1073/pnas.1905682116 Kim et al.

and high adsorption rate might represent an optimal strategy for
survival of viruses infecting chemolithoautotrophic AOA hosts.
The total number of NSV particles produced under laboratory
conditions was ∼298 ±18 virions per AOA cell at 6 dpi and
∼10
3
virus particles per micromole of NH
3
oxidized, with slight
strain-specific variations (Fig. 4 and SI Appendix,Fig.S8). The
nonlytic replication strategy of NSVs and other spindle-shaped ar-
chaeal viruses (26, 47), which allows for continuous virion production
and release, is radically different from the lytic life cycle of tailed
viruses of the order Caudovirales (48) but, in certain aspects, re-
sembles the nonlytic production of filamentous bacteriophages of the
Inoviridae family (49). Consequently, conventional ecological
models which are tailored to the mode of bacterial predation by
lytic head–tail phages (50, 51) might benefit from revision ac-
counting for alternative virus life cycles—exemplified here by
NSVs and the Nitrosopumilus sp. SW.
Environmental Distribution. Distribution of NSVs in marine envi-
ronments was analyzed using quantitative real-time PCR of the
NSV-specific pPolB gene. NSVs were detected in abundance in
various marine sediments (10
4
–10
6
NSV genome copies per gram of
sediment), coastal seawaters (10
4
–10
6
NSV genome copies per liter
of seawater), and coral-rich seawater (∼10
4
NSV genome copies per
liter of seawater) (SI Appendix,TableS1). By contrast, NSVs were
below the detection limit in a liter of typical oligotrophic seawater,
consistent with a previous study showing that spindle-shaped virus
particles corresponded to <1% of virions in tested seawater samples
(52). A comparison of the ratio of AOA counts to NSV counts
(Pearson’s correlation coefficient 0.557; Pvalue 0.015) indicated
that NSV levels might be positively correlated with the abundance of
AOA present. However, the correlation of counts of strain SW-
specific gene slp2 encoding an S-layer protein to those of NSV
counts was weak, suggesting that viral host specificity might be as-
sociated to the different versions of this protein (53).
Metagenomic recruitment analysis showed that NSVs were not
detected in any of the available metaviromes from marine envi-
ronments including those from sediments. A high adsorption rate of
NSV particles to host cells and particulate matter (Figs. 1Band 4C)
might have caused low recovery of NSVs from the viral fraction of
the sediments during virome preparations, which could explain the
lack of reads matching NSVs in the public metaviromes. Indeed, this
possibility was confirmed experimentally: only ∼1% of NSV particles
present in marine sediments could be extracted into the viral frac-
tion using the approach commonly used for virome preparation (SI
Appendix,Fig.S10). The adsorption of NSV to particulate matter is
also evidenced by frequent observation of NSV-like morphotypes in
SPM-rich bays (21) and surface microlayers of corals (20). Thus,
NSV-like genomes could be underrepresented in the viral fractions
extracted by conventional means from marine environments.
In addition to resource competition (bottom-up control), predation
(top-down control) can act as a key driver of biogeochemical cycles by
affecting microbial activity and community structures. Thus far, a
virus capable of infecting thaumarchaea had not been isolated, which
had limited the fundamental knowledge of the impact of viral in-
fection on the functioning and mortality of AOA. In this study,
spindle-shaped viruses were found to infect a marine ammonia-
oxidizing thaumarchaeon. The genome architecture and life cycle of
the examined NSVs indicate that they are distantly related to spindle-
shaped viruses that infect hyperthermophilic and hyperhalophilic ar-
chaea but represent a distinct new viral family. Characterization of
the infection strategy employed by NSVs suggests they might be
adapted for efficient infection of chemolithoautotrophic AOA hosts
and survival in resource-poor oceans. This study provides evidence
that viral predation severely affects the metabolism of infected AOA
cells, with a potential impact on global carbon and nitrogen cycling.
0.8
0.6
0.4
0.2
0.0
100
80
60
40
20
0
106
107
108
10
9
10
10
0
24
68
10 12 14
Time (days)
)%(
surivdedn
e
psus
ylee
rF
dnaenegANRrS61
lm(rebmunypocenegBloP
p
lariV
-1
))Mm(etirtindnaainommA
AO A
Tot al N S V1
1.0
Ammonia (NSV1-infected)
Ammonia (Non-infected control)
Nitrite (NSV1-infected)
Nitrite (Non-infected control)
A
B
C
Fig. 4. Properties of the NSV1 infection cycle. Strain SW cells were in-
fected with NSV1. Error bars represent SDs for three biological repli-
cates. (A) Comparison of ammonia oxidation by NSV1-infected and
noninfected control cells. (B) Virus production by strain SW cells infected
with NSV1. AOA growth and virus production were measured by qPCR
quantification of 16S rRNA and pPolB genes, respectively. (C)Fractionof
nonadsorbed NSV1 virions, estimated by qPCR of viral pPolB gene. Viral
genomic DNA of nonadsorbed virions was prepared from the culture
supernatant.
Kim et al. PNAS
|
July 30, 2019
|
vol. 116
|
no. 31
|
15649
MICROBIOLOGY

Materials and Methods
Isolation of AOA viruses, analysis of infection cycle of NSVs, and sequencing
and annotation of viral genomes are described in SI Appendix,SI Materials
and Methods. Details of DNA extraction from seawater and marine sedi-
ment, quantification of AOA and NSV from marine environments, and
metagenomic read recruitments are provided in SI Appendix,SI Materials
and Methods. Adsorption, host range, and stability of NSVs were tested as
described in SI Appendix,SI Materials and Methods. Phylogenetic analysis
of pPolB, comparative genomic analysis of AOA strains and NSVs, and
proteome analysis of NSV1 are described in SI Appendix,SI Materials and
Methods.
ACKNOWLEDGMENTS. This research was supported by National Research
Foundation of Korea (NRF) grants (Mid-Career Researcher Program [NRF-
2018R1A2B6008861], Basic Research Laboratory Program [NRF-2015R1A4A1041869],
and C1 Gas Refinery Program [NRF-2015M3D3A1A01064881]) funded by the
Ministry of Science, Information and Communication Technology, and Future
Planning. M.K. was supported by a grant from l’Agence Nationale de la Recherche
(#ANR-17-CE15-0005-01). The authors would like to thank Thibault Chaze and
Mariette Matondo (Pasteur Proteomics Platform) for help with the proteomics
analyses. M.L.-P. was supported by a postdoctoral fellowship (Juan de la Cierva)
from the Spanish Ministerio de Economía y Competitividad (IJCI-2017-34002).
F.R.-V. was supported by grant CGL2016-76273-P (Agencia Estatal de Investiga-
ción/European Development Regional Fund [FEDER], EU), (cofounded with FEDER
funds) from the Spanish Ministerio de Economía, Industria y Competitividad.
1. M. B. Karner, E. F. DeLong, D. M. Karl, Archaeal dominance in the mesopelagic zone
of the Pacific Ocean. Nature 409, 507–510 (2001).
2. A. E. Santoro, K. L. Casciotti, C. A. Francis, Activity, abundance and diversity of nitri-
fying archaea and bacteria in the central California Current. Environ. Microbiol. 12,
1989–2006 (2010).
3. A. E. Santoro, C. Buchwald, M. R. McIlvin, K. L. Casciotti, Isotopic signature of N(
2)
O
produced by marine ammonia-oxidizing archaea. Science 333, 1282–1285 (2011).
4. T. Nunoura et al., Distribution and niche separation of planktonic microbial com-
munities in the water columns from the surface to the hadal waters of the Japan
Trench under the Eutrophic Ocean. Front. Microbiol. 7, 1261 (2016).
5. A. M. Durbin, A. Teske, Sediment-associated microdiversity within the Marine Group I
Crenarchaeota. Environ. Microbiol. Rep. 2, 693–703 (2010).
6. M. Ballen-Segura, M. Felip, J. Catalan, Some mixotrophic flagellate species selectively
graze on archaea. Appl. Environ. Microbiol. 83, e02317-16 (2016).
7. D. De Corte, G. Paredes, T. Yokokawa, E. Sintes, G. J. Herndl, Differential response of
Cafeteria roenbergensis to different bacterial and archaeal prey characteristics. Mi-
crob. Ecol. 78,1–5 (2019).
8. R. Danovaro et al., Virus-mediated archaeal hecatomb in the deep seafloor. Sci. Adv.
2, e1600492 (2016).
9. M. Krupovic, A. Spang, S. Gribaldo, P. Forterre, C. Schleper, A thaumarchaeal provirus
testifies for an ancient association of tailed viruses with archaea. Biochem. Soc. Trans.
39,82–88 (2011).
10. S. S. Abby et al., Candidatus Nitrosocaldus cavascurensis, an ammonia oxidizing, ex-
tremely thermophilic archaeon with a highly mobile genome. Front. Microbiol. 9,28
(2018).
11. J. M. Labonté et al., Single-cell genomics-based analysis of virus-host interactions in
marine surface bacterioplankton. ISME J. 9, 2386–2399 (2015).
12. C.-E. T. Chow, D. M. Winget, R. A. White, 3rd, S. J. Hallam, C. A. Suttle, Combining
genomic sequencing methods to explore viral diversity and reveal potential virus-host
interactions. Front. Microbiol. 6, 265 (2015).
13. N. A. Ahlgren, C. A. Fuchsman, G. Rocap, J. A. Fuhrman, Discovery of several novel,
widespread, and ecologically distinct marine Thaumarchaeota viruses that encode
amoC nitrification genes. ISME J. 13, 618–631 (2019).
14. S. Roux et al.; Tara Oceans Coordinators, Ecogenomics and potential biogeochemical
impacts of globally abundant ocean viruses. Nature 537, 689–693 (2016).
15. M. Krupovic, V. Cvirkaite-Krupovic, J. Iranzo, D. Prangishvili, E. V. Koonin, Viruses of
archaea: Structural, functional, environmental and evolutionary genomics. Virus Res.
244, 181–193 (2018).
16. D. Prangishvili et al., The enigmatic archaeal virosphere. Nat. Rev. Microbiol. 15, 724–
739 (2017).
17. N. Dellas, J. C. Snyder, B. Bolduc, M. J. Young, Archaeal viruses: Diversity, replication,
and structure. Annu. Rev. Virol. 1, 399–426 (2014).
18. M. K. Pietilä, T. A. Demina, N. S. Atanasova, H. M. Oksanen, D. H. Bamford, Archaeal
viruses and bacteriophages: Comparisons and contrasts. Trends Microbiol. 22, 334–
344 (2014).
19. M. Krupovic, E. R. J. Quemin, D. H. Bamford, P. Forterre, D. Prangishvili, Unification of
the globally distributed spindle-shaped viruses of the Archaea. J. Virol. 88, 2354–2358
(2014).
20. J. E. Davy, N. L. Patten, Morphological diversity of virus-like particles within the sur-
face microlayer of scleractinian corals. Aquat. Microb. Ecol. 47,37–44 (2007).
21. W. N. S. Arlidge, “Coral reef viruses in Kane’ohe Bay, Hawai’i,”Masters thesis, Victoria
University of Wellington, Wellington, New Zealand, (2012).
22. O. D. Nigro et al., Viruses in the oceanic basement. MBio 8, e02129-16 (2017).
23. J.-G. Kim et al., Hydrogen peroxide detoxification is a key mechanism for growth of
ammonia-oxidizing archaea. Proc. Natl. Acad. Sci. U.S.A. 113, 7888–7893 (2016).
24. C. Schleper, K. Kubo, W. Zillig, The particle SSV1 from the extremely thermophilic
archaeon Sulfolobus is a virus: Demonstration of infectivity and of transfection with
viral DNA. Proc. Natl. Acad. Sci. U.S.A. 89, 7645–7649 (1992).
25. D. Prangishvili, M. Krupovic; ICTV Report Consortium, ICTV virus taxonomy profile:
Ampullaviridae. J. Gen. Virol. 99, 288–289 (2018).
26. E. R. J. Quemin et al., Eukaryotic-like virus budding in archaea. MBio 7, e01439-16
(2016).
27. M. Häring et al., Virology: Independent virus development outside a host. Nature 436,
1101–1102 (2005).
28. E. R. J. Quemin et al., First insights into the entry process of hyperthermophilic ar-
chaeal viruses. J. Virol. 87, 13379–13385 (2013).
29. P. Kukkaro, D. H. Bamford, Virus-host interactions in environments with a wide range
of ionic strengths. Environ. Microbiol. Rep. 1,71–77 (2009).
30. C. Bath, T. Cukalac, K. Porter, M. L. Dyall-Smith, His1 and His2 are distantly related,
spindle-shaped haloviruses belonging to the novel virus group, Salterprovirus.Virol-
ogy 350, 228–239 (2006).
31. E. R. J. Quemin et al., Sulfolobus spindle-shaped virus 1 contains glycosylated capsid
proteins, a cellular chromatin protein, and host-derived lipids. J. Virol. 89, 11681–
11691 (2015).
32. S.-V. Albers, B. H. Meyer, The archaeal cell envelope. Nat. Rev. Microbiol. 9, 414–426
(2011).
33. W. Qinet al ., Nitrosopumilus maritimus gen. nov., sp. nov., Nitrosopumilus cobalaminigenes
sp. nov., Nitrosopumilus oxyclinae sp. nov., and Nitrosopumilus ureiphilus sp. nov.,
four marine ammonia-oxidizing archaea of the phylum Thaumarchaeota. Int. J. Syst.
Evol. Microbiol. 67, 5067–5079 (2017).
34. M. Kerou et al., Proteomics and comparative genomics of Nitrososphaera viennensis
reveal the core genome and adaptations of archaeal ammonia oxidizers. Proc. Natl.
Acad. Sci. U.S.A. 113, E7937–E7946 (2016).
35. M. Stieglmeier et al., Nitrososphaera viennensis gen. nov., sp. nov., an aerobic and
mesophilic, ammonia-oxidizing archaeon from soil and a member of the archaeal
phylum Thaumarchaeota. Int. J. Syst. Evol. Microbiol. 64, 2738–2752 (2014).
36. P. López-García, Y. Zivanovic, P. Deschamps, D. Moreira, Bacterial gene import and
mesophilic adaptation in archaea. Nat. Rev. Microbiol. 13, 447–456 (2015).
37. J. P. Beam, Z. J. Jay, M. A. Kozubal, W. P. Inskeep, Niche specialization of novel
Thaumarchaeota to oxic and hypoxic acidic geothermal springs of Yellowstone Na-
tional Park. ISME J. 8, 938–951 (2014).
38. M. Stieglmeier, R. J. E. Alves, C. Schleper, “The phylum Thaumarchaeota”in The
Prokaryotes: Other Major Lineages of Bacteria and the Archaea, E. Rosenberg,
E. F. DeLong, S. Lory, E. Stackebrandt, F. Thompson, Eds., (Springer, Berlin, Heidelberg),
pp 347–362 (2014).
39. L. Xiong et al., A new type of DNA phosphorothioation-based antiviral system in
archaea. Nat. Commun. 10, 1688 (2019).
40. J. Iranzo, E. V. Koonin, D. Prangishvili, M. Krupovic, Bipartite network analysis of the
archaeal virosphere: Evolutionary connections between viruses and capsidless mobile
elements. J. Virol. 90, 11043–11055 (2016).
41. C. Bath, M. L. Dyall-Smith, His1, an archaeal virus of the Fuselloviridae family that
infects Haloarcula hispanica.J. Virol. 72, 9392–9395 (1998).
42. H. Wang et al., Novel Sulfolobus Virus with an exceptional capsid architecture.
J. Virol. 92, e01727-17 (2018).
43. M. Krupovic, P. Béguin, E. V. Koonin, Casposons: Mobile genetic elements that gave
rise to the CRISPR-Cas adaptation machinery. Curr. Opin. Microbiol. 38,36–43 (2017).
44. M. Krupovic et al., Integrated mobile genetic elements in Thaumarchaeota. Environ.
Microbiol. 21, 2056–2078 (2019).
45. M. K. Pietilä, N. S. Atanasova, H. M. Oksanen, D. H. Bamford, Modified coat protein
forms the flexible spindle-shaped virion of haloarchaeal virus His1. Environ. Microbiol.
15, 1674–1686 (2013).
46. E. I. Rensen et al., A virus of hyperthermophilic archaea with a unique architecture
among DNA viruses. Proc. Natl. Acad. Sci. U.S.A. 113, 2478–2483 (2016).
47. J. Svirskait _
e, H. M. Oksanen, R. Daugelaviˇ
cius, D. H. Bamford, Monitoring physiolog-
ical changes in haloarchaeal cell during virus release. Viruses 8, 59 (2016).
48. R. Young, Phage lysis: Three steps, three choices, one outcome. J. Microbiol. 52, 243–
258 (2014).
49. B. Loh, A. Kuhn, S. Leptihn, The fascinating biology behind phage display: Filamen-
tous phage assembly. Mol. Microbiol. 111, 1132–1138 (2019).
50. Y. Zhao et al., Abundant SAR11 viruses in the ocean. Nature 494, 357–360 (2013).
51. I. Kang, H.-M. Oh, D. Kang, J.-C. Cho, Genome of a SAR116 bacteriophage shows the
prevalence of this phage type in the oceans. Proc. Natl. Acad. Sci. U.S.A. 110, 12343–
12348 (2013).
52. J. R. Brum, R. O. Schenck, M. B. Sullivan, Global morphological analysis of marine
viruses shows minimal regional variation and dominance of non-tailed viruses. ISME
J. 7, 1738–1751 (2013).
53. S. Avrani, O. Wurtzel, I. Sharon, R. Sorek, D. Lindell, Genomic island variability facil-
itates Prochlorococcus-virus coexistence. Nature 474, 604–608 (2011).
15650
|
www.pnas.org/cgi/doi/10.1073/pnas.1905682116 Kim et al.