Downloaded from Genome Website

Home , Acholeplasma laidlawii, Anaeroplasmatales, Entomoplasmatales, Haloplasma, Tenericutes

bioRxiv preprint doi: https://doi.org/10.1101/2020.11.18.388454; this version posted November 19, 2020. The copyright holder for this preprint (which was not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission.

1 Characterization of the first cultured free-living representative of

2 Candidatus Izimaplasma uncovers its unique biology

3 Rikuan Zheng1,2,3,4, Rui Liu1,2,4, Yeqi Shan1,2,3,4, Ruining Cai1,2,3,4, Ge Liu1,2,4, Chaomin Sun1,2,4*

1 4 CAS Key Laboratory of Experimental Marine Biology & Center of Deep Sea

5 Research, Institute of Oceanology, Chinese Academy of Sciences, Qingdao, China

2 6 Laboratory for Marine Biology and Biotechnology, Qingdao National Laboratory

7 for Marine Science and Technology, Qingdao, China

3 8 College of Earth Science, University of Chinese Academy of Sciences, Beijing,

9 China

10 4Center of Ocean Mega-Science, Chinese Academy of Sciences, Qingdao, China

12 * Corresponding author

13 Chaomin Sun Tel.: +86 532 82898857; fax: +86 532 82898857.

14 E-mail address: [email protected]

17 Key words: Candidatus Izimaplasma, uncultivation, biogeochemical cycling,

18 extracellular DNA, in situ, deep sea

19 Running title: Characterization of the first cultured Izimaplasma

22 Abstract

23 Candidatus Izimaplasma, an intermediate in the reductive evolution from Firmicutes

24 to Mollicutes, was proposed to represent a novel class of free-living wall-less bacteria

25 within the phylum Tenericutes found in deep-sea methane seeps. Unfortunately, the

26 paucity of marine isolates currently available has limited further insights into their

27 physiological and metabolic features as well as ecological roles. Here, we present a

28 detailed description of the phenotypic traits, genomic data and central metabolisms

29 tested in both laboratorial and deep-sea environments of the novel strain zrk13, which

30 allows for the first time the reconstruction of the metabolic potential and lifestyle of a

31 member of the tentatively defined Candidatus Izimaplasma. On the basis of the

32 description of strain zrk13, the novel species and genus Xianfuyuplasma coldseepsis is

33 proposed. Notably, DNA degradation driven by X. coldseepsis zrk13 was detected in

34 both laboratorial and in situ conditions, strongly indicating it is indeed a key DNA

35 degrader. Moreover, the putative genes determining degradation broadly distribute in

36 the genomes of other Izimaplasma members. Given extracellular DNA is a

37 particularly crucial phosphorus as well as nitrogen and carbon source for

38 microorganisms in the seafloor, Izimaplasma bacteria are thought to be important

39 contributors to the biogeochemical cycling in the deep ocean.

43 Introduction

44 The phylum Tenericutes is composed of bacteria lacking a peptidoglycan cell wall

45 and consists of bacteria that evolved from the phylum Firmicutes1-3. Two distinctive

46 features set Tenericutes apart from the Firmicutes: the inability to synthetize

47 precursors of peptidoglycan and thereby forming a cell wall4-7, and extreme reduction

48 of the size of genome (between 530 to 2220 kbp)8,9. Tenericutes includes the class

49 Mollicutes2, three taxa in provisional class Candidatus Izimaplasma4, and several taxa

50 of unclassified status. Tenericutes bacteria have evolved a broad range of lifestyles,

51 including free-living, commensalism and parasitism2. Up to date, almost all reported

52 Mollicutes (including five orders Mycoplasmatales, Entomoplasmatales,

53 Haloplasmatales, Acholeplasmatales, and Anaeroplasmatales)10 are commensals or

54 obligate parasites of humans, domestic animals, plants and insects2. In comparison,

55 some free-living species are found to be associated with inert substrates (e.g.

56 Candidatus Izimaplasma)4 or animal/plant surfaces (e.g. Acholeplasma laidlawii)11,12.

57 Tenericutes bacteria ubiquitously exist in numerous environments13.

58 Environmental 16S rRNA surveys have identified a large number of unknown

59 Tenericutes clades in diverse environments including the deep oceans, introducing the

60 possibility that these Tenericutes may represent free-living microorganisms which

61 conducting a non-host-associated life style2. Indeed, free-living Candidatus

62 Izimaplasma4 and Haloplasma14 were reported in a deep-sea cold seep and brine pool,

63 respectively. These deep-sea free-living Tenericutes exhibit metabolic versatility and

64 adaptive flexibility, showing the possibility to isolate more Tenericutes bacteria from

65 the oceans and even other extreme environments.

66 Among the free-living Tenericutes identified in the marine environments,

67 Candidatus Izimaplasma was mainly studied based on the enrichment and

68 corresponding assembled genomes given the absence of pure cultures4,15. Candidatus

69 Izimaplasma was proposed to represent a novel class of free-living wall-less bacteria

70 within the phylum Tenericutes4, and an intermediate in the reductive evolution from

71 Firmicutes to Mollicutes4. Members of the Izimaplasma are thought to be heterotrophs

72 engaging primarily in fermentative metabolism4, and thereby producing lactate and

73 possibly other small molecules such as acetate or ethanol, which in turn may be

74 utilized by other members of the cold seep15. Notably, based on their genomic

75 analysis, Candidatus Izimaplasma bacteria are believed to be important DNA

76 degraders4,15. Recent studies have shown high concentrations of extracellular DNA in

77 marine sediments from shallow depths down to the abyssal floor16-18. Given

78 extracellular DNA is a particularly crucial phosphorus as well as nitrogen and carbon

79 source for microorganisms in the seafloor, Candidatus Izimaplasma might be a key

80 contributor to the biogeochemical cycling in the deep ocean. However, the paucity of

81 isolates currently available has greatly limited further insights into the physiological,

82 ecological and evolutional studies of Candidatus Izimaplasma.

83 In this study, we successfully cultivated the first free-living representative of the

84 class Candidatus Izimaplasma. Using growth assay and transcriptomic methods, we

85 further found organic nutrient and thiosulfate could significantly promote the growth

86 of the new isolate. Of note, the novel isolate was found to be an effective degrader of

87 extracellular DNA in both laboratorial and real deep-sea conditions. Lastly, its

88 metabolic pathways were reconstructed based on multi-omics results presented in this

89 study and its potential ecological roles were also discussed.

90 Results

91 Quantification of the abundance of Tenericutes and Candidatus Izimaplasma in

92 the deep-sea cold seep sediments. To gain some preliminary understanding of

93 Tenericutes in the deep-sea, operational taxonomic units (OTUs) sequencing was

94 performed to detect the abundance of Tenericutes present in the cold seep sediments

95 at depth intervals of 0-10 cm, 30-50 cm, 90-110 cm, 150-170 cm, 210-230 cm,

96 230-250 cm from the surface to the deep layer. As previously reported2, the phylum

97 Tenericutes had a very low abundance in the cold seep sediments (Supplementary Fig.

98 1a). For example, the percentages of Tenericutes only respectively accounted for

99 0.011%, 0.222% and 0.003% of the whole bacterial domain in the samples ZC1, ZC3

100 and ZC4 (Supplementary Fig. 1a). We even could not detect any Tenericutes in other

101 three samples. Thus we propose that extreme low abundance might be a major reason

102 limiting the isolation of bacteria from this phylum. To obtain further insights into the

103 Tenericutes in deep-sea sediments, we performed metagenomics sequencing of the

104 sample ZC3 and analyzed all the annotated genes. The results showed that sequences

105 associated with phylum Tenericutes represented 0.250% of all annotated bacterial 5

106 sequences (Supplementary Fig. 1b), which was consistent with the OTUs result of the

107 ZC3 sample (Supplementary Fig. 1a). Of note, the proportion of genes associated with

108 Candidatus Izimaplasma to those related to Tenericutes was up to 93.289%

109 (Supplementary Fig. 1c), indicating the dominant status of Candidatus Izimaplasma

110 within the deep-sea Tenericutes.

111 DNA degradation-driven isolation of the first cultured free-living representative

112 of Candidatus Izimaplasma from the deep-sea cold seep. Despite what has been

113 learned from cultivation independent methods4, the lack of cultured free-living

114 representatives of deep-sea Candidatus Izimaplasma has hampered a more detailed

115 exploration of the group. With this, we need cultivated representatives to provide

116 further understanding of Candidatus Izimaplasma from the deep-sea environment. In

117 service of this goal, we improved the enrichment strategy by using a specific medium

118 containing only some inorganic salts supplemented with Escherichia coli genomic

119 DNA as the nutrient sources, given the previously predicted superior DNA

120 degradation capability of Candidatus Izimaplasma4,15. Using this medium, we

121 anaerobically enriched the deep-sea sediment samples at 28 °C for six months.

122 Enriched samples were then plated on solid medium in Hungate tubes and individual

123 colonies with distinct morphology were picked and cultured (Fig. 1a). Excitedly, most

124 cultured colonies were identified as members of Candidatus Izimaplasma. Among

125 them, strain zrk13 possessed a fast growth rate and was chosen for further study.

126 Under transmission electron microscopy (TEM) observation, the cells of strain zrk13

127 were coccoid (Figs. 1b and 1e), approximately 300-800 nm in size, and had no

128 flagellum. As expected, cells of strain zrk13 had no distinct cell wall compared with

129 the Gram-positive bacteria cell wall found within members of the Firmicutes (Figs. 1c

130 and 1d), which was confirmed by the TEM observation of ultrathin-sections of strain

131 zrk13 and a typical Firmicutes bacterium (Figs. 1f and 1g). Overall, we successfully

132 obtained the first cultured free-living representative of Candidatus Izimaplasma from

133 the cold seep through a DNA degradation-driven strategy, which might be useful to

134 enrich and isolate other uncultured candidates of Izimaplasma in the future.

135 Genomic characteristics and phylogenetic analysis of strain zrk13. To understand

136 more characteristics of the strain zrk13, its whole genome was sequenced and

137 analyzed. The genome size of strain zrk13 is 1,958,905 bp with a DNA G+C content

138 of 38.2% (Supplementary Fig. 2 and Supplementary Table 1). Annotation of the

139 genome of strain zrk13 revealed it consisted of 1,872 predicted genes that included 64

140 RNA genes (three rRNA genes, 42 tRNA genes and 28 other ncRNAs). The genome

141 relatedness values were calculated by the amino acid identity (AAI), the average

142 nucleotide identity (ANI), in silico DNA-DNA similarity (isDDH) and the

143 tetranucleotide signatures (Tetra), against the four genomes (strains zrk13, HR1, HR2

144 and ZiA1) (Supplementary Table 1). The AAI values of zrk13 with HR1, HR2 and

145 ZiA1 were 68.8%, 66.5% and 64.3%, respectively. The average nucleotide identities

146 (ANIb) of zrk13 with HR1, HR2 and ZiA1 were 68.72%, 67.42% and 66.57%,

147 respectively. The average nucleotide identities (ANIm) of zrk13 with HR1, HR2 and

148 ZiA1 were 85.94%, 86.53% and 81.98%, respectively. The tetra values of zrk13 with

149 HR1, HR2 and ZiA1 were 0.85192, 0.77694 and 0.837. Based on digital DNA-DNA

150 hybridization employing the Genome-to-Genome Distance Calculator GGDC, the in

151 silico DDH estimates for zrk13 with HR1, HR2 and ZiA1 were 17.50%, 18.50% and

152 15.80%, respectively. These results together demonstrated the genome of strain zrk13

153 to be clearly below established ‘cut-off’ values (ANIb: 95%, ANIm: 95%, AAI: 95%,

154 isDDH: 70%, Tetra: 0.99) for defining bacterial species, strongly suggesting it

155 represents a novel taxon within Candidatus Izimaplasma as currently defined.

156 To clarify the taxonomic status of strain zrk13, we further performed the

157 phylogenetic analyses with 16S rRNA genes from cultured Tenericutes and

158 Firmicutes representatives, unclassified Haloplasma contractile SSD-17B and some

159 uncultured Candidatus Izimaplasma. The maximum likelihood tree of 16S rRNA

160 placed the clade ‘Izimaplasma’ as a sister group of the class Mollicutes, which

161 together form a distinct cluster as Tenericutes separating from the phylum Firmicutes

162 and unclassified Haloplasma contractile SSD-17B (Fig. 1h). Of note, the position of

163 Candidatus Izimaplasma was just located between the phylum Firmicutes and

164 Mollicutes, indicating Candidatus Izimaplasma could represent an intermediate in the

165 reductive evolution from Firmicutes to Mollicutes. Given that strain zrk13 is the first

166 cultured free-living representative of a proposed class Candidatus Izimaplasma4,

167 which is qualified to be named as Izimaplasma from now on. A sequence similarity

168 calculation using the NCBI server indicated that the closest relative of strain zrk13

169 was uncultivated Izimaplasma HR1 (95.26%). Therefore, we propose that strain zrk13

170 is classified as the type strain of a novel genus in the class Izimaplasma, for which the

171 name Xianfuyuplasma coldseepsis gen. nov., sp. nov. is proposed.

172 Description of Xianfuyuplasma gen. nov. and Xianfuyuplasma coldseepsis sp. nov..

173 For the genus name Xianfuyuplasma Xian.fu.yu'plas.ma. L. fem. n. Xianfuyu comes

174 from a strange animal’s name described in the Classic of Mountains and Rivers-a very

175 famous Chinese mythology. Xianfuyu is a kind of strange animal possessing a fish’s

176 head and pig’s body (Supplementary Fig. 3), which is similar to Izimaplasma that

177 possessing both characteristics of Tenericutes and Firmicutes. -plasma, formed or

178 molded, refers to the lack of cell wall. The genus Xianfuyuplasma is a kind of strictly

179 anaerobic microorganisms, whose cells are non-motile and coccoid. Its phylogenetic

180 position is classified in the family Izimaplasmaceae, order Izimaplasmales, class

181 Izimaplasma within the bacterial phylum Tenericutes. The type species is

182 Xianfuyuplasma coldseepsis.

183 Xianfuyuplasma coldseepsis (cold.seep'sis. L. gen. pl. n. coldseepsis of or

184 belonging to the deep-sea). For this species, cells are coccoid, approximately 300-800

185 nm in size, and had no flagellum; strictly anaerobic; the temperature range for growth

186 is 24-32 °C with an optimum at 28 °C. Growing at pH values of 6.0-8.0 (optimum, pH

187 7.0); growth occurs at NaCl concentrations between 0.0-8.0% with an optimum

188 growth at 4.0% NaCl; growth is stimulated by using glucose, maltose, butyrate,

189 acetate, formate, fructose, sucrose, mannose, starch, isomaltose, trehalose, lactate,

190 ethanol, glycerin and rhamnose as a sole carbon source (Supplementary Table 2),

191 showing quite differences toward the carbon sources predicted by other members of

192 the class Izimaplasma. The type strain, zrk13T, was isolated from the sediment of

193 deep-sea cold seep, P. R. China.

194 Organic nutrient and thiosulfate significantly promote the growth of

195 X.coldseepsis zrk13. In the previous study, uncultivated Izimaplasma HR1 and HR2

196 were found to be better enriched in the medium amended with small amount of

197 glucose (5 g/L) and yeast extract (0.2 g/L) than that in the inorganic medium4.

198 Similarly, the growth rate of X. coldseepsis zrk13 was also significantly promoted in

199 the organic medium containing yeast extract (1 g/L) and peptone (1 g/L) compared

200 with that in the oligotrophic medium containing less yeast extract (0.1 g/L) and

201 peptone (0.1 g/L) (Fig. 2a), strongly indicating strain zrk13 is a heterotrophic

202 bacterium. Given the tricarboxylic cycle was incomplete in the previous4 and the

203 present Izimaplasma genomes, we sought to ask how does Izimaplasma utilize

204 organic nutrient. Therefore, we performed the transcriptomic analysis of X.

205 coldseepsis zrk13 cultured in oligotrophic and rich media. The results revealed that

206 the expressions of genes encoding different glycoside hydrolases (Fig. 2b), FeFe

207 hydrogenases (Fig. 2c), NADH-ubiquinone oxidoreductase (Fig. 2d) and ATP

208 synthase (Fig. 2e) were significantly up-regulated. Glycoside hydrolases are enzymes

209 that catalyze the hydrolysis of the glycosidic linkage of glycosides, leading to the

210 formation of small molecules sugars for easy utilization by organisms

211 (http://www.cazypedia.org/index.php/Glycoside_hydrolases). Correspondingly,

212 expressions of different glycoside hydrolases including families 1, 16, 17, 30, 35, 65

213 were evidently up-regulated, suggesting many categories of glycans could be

214 efficiently degraded by X. coldseepsis zrk13. Indeed, all the present and two

215 uncultured Izimaplasma HR1 and HR2 genomes contained a complete set of genes

216 associated with Embden-Meyerhof-Parnas (EMP) glycolysis and pentose phosphate

217 pathways4, indicating Izimaplasa bacteria possess a significant capability of glycan

218 metabolism. Of note, the expressions of all three iron hydrogenase maturation genes

219 (hydEFG) were up-regulated about 30-fold (Fig. 2b), and this kind of hydrogenase is

220 usually linked with NADP and may generate NADPH for the production of

221 biomolecules4. Correspondingly, the expressions of many genes related to

222 NADH-ubiquinone oxidoreductase were up-regulated 20-fold (Supplementary Fig. 7),

223 which catalyzed the transfer of two electrons from NADH to reduce ubiquinone to

224 ubiquinol and was also the entry point for a large fraction of the electrons that traverse

225 the respiratory chain19,20. With this, it is reasonable to see that the expressions of

226 different genes encoding ATP synthase were also greatly up-regulated (Fig. 2e),

227 which converted the energy of protons (H+) moving down their concentration gradient

228 into the synthesis of ATP and thereby promoting the bacterial growth. Overall, we

229 conclude that X. coldseepsis zrk13 could effectively utilize organic nutrients for

230 growth through anaerobic fermentation and has the potential to produce hydrogen4.

231 X. coldseepsis zrk13 was isolated from a deep-sea cold seep, and different sulfur

232 sources were detected to ubiquitously exist in the environment in our previous study21.

233 Therefore, we tested the effects of different sulfur-containing inorganic substances (e.

234 g. Na2SO4, Na2SO3, Na2S2O3, Na2S) on the growth of X. coldseepsis zrk13. The

235 results showed that only the supplement of Na2S2O3 could significantly promote the

236 growth of strain zrk13 (Fig. 3a). We thus performed the transcriptomic analysis of

237 strain zrk13 cultured in the organic medium amended with or without Na2S2O3 to

238 explore the underlying mechanism of growth promotion. Surprisingly, we didn’t find

239 obvious up-regulation of typical genes associated with sulfur metabolism.

240 Alternatively, the expressions of many genes encoding [2Fe-2S] binding proteins or

241 [4Fe-4S] di-cluster domain-containing proteins were markedly up-regulated (Fig.

242 3b), and these proteins are thought to play a variety of functions including substrate

243 binding and activation, electron transport, gene expression regulation and enzyme

244 activity in response to external stimuli22. Meanwhile, the expressions of large amount

245 of genes encoding enzymes responsible for saccharides degradation (Fig. 3c) and

246 sugar transport (Fig. 3d) were evidently up-regulated. In combination of the results

247 showed in Fig. 2, we speculate that thiosulfate may accelerate the hydrolysis and

248 uptake of saccharides under the control of [Fe-S] associated proteins and thereby

249 synthesizing energy and promoting the growth of strain zrk13. Together, we believe X.

250 coldseepsis zrk13 possesses a capability to utilize both organic nutrients (saccharides)

251 and inorganic sulfur-containing compounds (thiosulfate) ubiquitously existing in the

252 deep-sea environments.

253 X.coldseepsis zrk13 possesses a significant capability of degrading DNA in both

254 laboratorial and deep-sea environments. Based on several previous reports2,4,15 and

255 our present isolation strategy, Izimaplasma is firmly believed to degrade extracellular

256 DNA and use as a nutrient source for growth. However, this hypothesis is still lack of

257 solid proof due to the uncultivation status of Izimaplasma. With that, detection of

258 DNA degradation and utilization was further performed with strain zrk13. First, an

259 in-depth analysis of the genome of strain zrk13 revealed the existence of a putative

260 DNA-degradation related gene cluster, which contained genes encoding DNA

261 degradation protein EddB, DNase/RNase endonuclease, thermonuclease,

262 phosphohydrolases, ABC-transporters, phosphate transporters and other proteins

263 associated with DNA degradation (Fig. 4a). In order to test the ability of strain zrk13

264 to degrade DNA, genomic DNA without removing rRNA of E. coli was extracted for

265 further assays. The results showed that the supernatant of zrk13 cultures began to

266 digest chromosomal DNA within 5 min (Fig. 4b, lane III) and degraded all DNA

267 within 15 min (Fig. 4b, lane IX). However, the supernatant of zrk13 didn’t degrade

268 RNA at all, indicating its specific DNA degradation capability (Fig. 4b). Moreover,

269 the supplement of genomic DNA in the inorganic medium could significantly

270 promote the growth rate and biomass amount of zrk13 (Fig. 4c), strongly suggesting

271 this bacterium could efficiently utilize the digested DNA as a nutrient source to

272 support further growth. Consistently, the expressions of genes related to DNA

273 degradation were significantly up-regulated, especially the gene encoding DNA

274 degradation protein EddB, which was up-regulated about 20-fold (Fig. 4d), indicating

275 the key roles of the DNA-degradation gene cluster in the course of DNA degradation

276 and utilization. Notably, compared with 62 finished and draft genomes of Tenericutes

277 and Firmicutes bacteria, the members of Izimaplasma harbored more genes associated

278 with DNA degradation than those in Mollicutes and Firmicutes (Supplementary Table

279 3 and Supplementary Fig. 4), confirming the strong capability of DNA degradation of

280 Izimaplasma.

281 Taken all the above results, we are clear about the metabolic landscape of X.

282 coldseepsis zrk13 based on the laboratorial conditions. However, its real metabolisms

283 performed in the deep-sea environment are still obscure. With that, we performed the

284 in situ cultivation in the June of 2020 to study the metabolisms of strain zrk13 in the

285 deep-sea cold seep, where we isolated this bacterium and lives a lot of typical cold

286 seep animals such as mussels and shrimps (Figs. 5a, 5b). The transcriptomics results

287 based on the in situ cultivated cells showed that the expressions of many genes

288 encoding exonuclease, endonuclease and ribonuclease were obviously up-regulated

289 compared with control group (Fig. 5c), indicating DNA-degradation indeed happened

290 in the deep sea. Correspondingly, the expressions of all genes belonging to a gene

291 cluster closely associated with purine nucleobase catabolism (such as adenine

292 deaminase, guanine deaminase and xanthine dehydrogenase) were significantly

293 up-regulated up to 16-fold (Fig. 5d), strongly suggesting zrk13 had a remarkable

294 ability to degrade purine and obtain energy to maintain growth in situ. Surprisingly,

295 the expressions of almost all the genes involved in EMP glycolysis were significantly

296 down-regulated (Figs. 5e and 5f), which might be due to the deficiency of organic

297 nutrients in the deep-sea environment. Instead, strain zrk13 could be able to maintain

298 a good growth status mainly by digesting DNA to obtain energy for supporting

299 growth given its strong DNA degradation capability and the ubiquitous existence of

300 extracellular DNA in the deep-sea sediments15,23.

301 Discussion

302 Microbes in deep ocean sediments represent a large portion of the biosphere, and

303 resolving their ecology is crucial for understanding global ocean processes24-26.

304 Despite the global importance of these microorganisms, majority of deep-sea

305 microbial diversity remains uncultured and poorly characterized24. Description of the

306 metabolisms of these novel taxa is advancing our understanding of their

307 biogeochemical roles, including the coupling of elements and nutrient cycling, in the

308 deep oceans. Given the importance of these communities to the oceans, there is an

309 urgent need to obtain more uncultivated isolates for better resolving the diversity and

310 ecological roles of these uncultured taxa. Candidatus Izimaplasma bacteria were

311 proposed to represent a novel class of free-living representatives from a Tenericutes

312 clade found in deep-sea methane seeps4, and they were believed to actively participate

313 in the primary degradation of extracellular DNA in anoxic marine sediments and 15

314 contribute to the biogeochemical cycling of deep biosphere4,15. Unfortunately, up to

315 date, there is no any available pure culture of Izimaplasma bacteria, which seriously

316 hinders the accurate determination of their features such as growth, metabolism,

317 physiology and ecology.

318 In the present study, we developed an effective enrichment method driven by

319 DNA-degradation (Fig. 1a), and successfully obtained a novel Izimaplasma isolate, X.

320 coldseepsis zrk13, which is the first pure culture in the Izimaplasma class. Based on

321 the isolate, we detailedly explored its physiology, genomic traits, phylogenetics and

322 metabolisms through bioinformatics, biochemical and transcriptomic methods, and

323 proposed a model describing its central metabolic pathways (Fig. 6). Clearly, DNA

324 and organic matter degradation and utilization play key roles for energy production

325 and thereby driving growth of X. coldseepsis zrk13 (Fig. 6).

326 In the laboratorial culturing condition, X. coldseepsis zrk13 prefers to perform a

327 heterotrophic life given the easily available organic matter in the medium (Fig. 2a).

328 Therefore, the expressions of large amount of genes associated with sugar degradation

329 were significantly up-regulated compared with the bacterium cultured in the

330 oligotrophic medium (Fig, 2b). In contrast, the expressions of genes responsible for

331 sugar metabolisms (EMP glycolysis) were down-regulated in the in situ cultivated X.

332 coldseepsis zrk13 (Figs. 5e and 5f) compared with that incubated in the rich medium.

333 Considering the exchanges between the medium inside the incubation bag and the

334 outside seawater, we speculate that the down-regulation of saccharide metabolism in

335 situ might be the lower content of organic matters in the seawater than that in the

336 organic medium. Notably, thiosulfate also promotes the saccharides metabolism and

337 thereby benefiting the bacterial growth (Fig. 3), which endows X. coldseepsis zrk13 a

338 better adaptability to the deep-sea environment given ubiquitous existence of

339 thiosulfate in the cold seep21. Therefore, we believe saccharide metabolism is crucial

340 for maintaining the growth of X. coldseepsis zrk13 both in the laboratorial and

341 deep-sea environments. In combination of previous report4, we predict part of the

342 ecological role of Izimaplasma in the cold seep is likely in the fermentation of

343 products from the degradation of organic matter to produce lactate and possibly other

344 small molecules such as acetate or ethanol, which in turn are utilized by other

345 members of the cold seep microbes (Fig. 6).

346 In comparison, the degradation of DNA mediated by X. coldseepsis zrk13

347 happened in both laboratorial (Fig. 4) and deep-sea conditions (Fig. 5), strongly

348 indicating this bacterium is indeed a key DNA degrader. Consistently, an intact locus

349 containing genes encoding enzymes for further digestion of DNA into

350 oligonucleotides and nucleosides, as well as removal of phosphates from nucleotides

351 is identified in the genome of X. coldseepsis zrk13 (Fig. 4a). And nucleases within

352 and outside of this gene cluster were demonstrated to function in the course of DNA

353 degradation (Fig. 4 and Fig. 5). Comparative genomics indicated that DNA can be

354 digested by diverse members of Izimaplasma15 as well as X. coldseepsis zrk13

355 (Supplementary Fig. 4), strongly suggesting Izimaplasma bacteria are specialized

356 DNA-degraders that encode multiple extracellular nucleases. DNA degradation

357 process is closely associated with amino acids and saccharides metabolisms (Fig. 6)

358 as well as the production of acetyl-CoA4. The products of nucleic acids (such as urea

359 and ammonium, CO2 and acetate) are also important nutrients for other members of

360 microbial communities4,15.

361 It is noting that extracellular DNA is a major macromolecule in global element

362 cycles, and is a particularly crucial source of phosphorus, nitrogen and carbon for

363 microorganisms in the seafloor15. It was estimated that extracellular DNA available

364 for degradation by extracellular nucleases, supplies microbial communities of both

365 coastal and deep-sea sediments with 2-4% of their carbon requirements, 4-7% of their

366 nitrogen needs, and a remarkable 20-47% of their phosphorus demands23,27.

367 Considering the tremendous body of marine water, the oceans harbor a massive

368 Izimaplasma population composed of diverse novel linages. Therefore, extracellular

369 available DNAs and their key degraders like Izimaplasma contribute substantially to

370 oceanic and sedimentary biogeochemical cycles, and provide an enormous energy

371 source for microbial communities. Overall, based on the comprehensive

372 investigations of the first cultured isolate of Izimaplasma in both laboratorial and

373 deep-sea conditions, our present study expands the ecophysiological understanding of

374 Izimaplasma bacteria by showing that they actively participate in the primary

375 degradation of extracellular DNA and other organic matter in anoxic deep-sea

376 sediments.

377 Methods

378 Samples sampling and statistical analysis of 16S rRNA genes of Tenericutes in

379 the deep-sea cold seep. The deep-sea samples were collected by RV KEXUE from a

380 typical cold seep in the South China Sea (E119º17'07.322'', N22º06'58.598'') as

381 described previously21. To understand the abundance of Tenericutes in deep-sea cold

382 seep sediments, we selected and collected six sedimentary samples (RPC, ZC1, ZC2,

383 ZC3, ZC4 and ZC5 at depth intervals of 0-10, 30-50, 90-110, 150-170, 210-230 and

384 230-250 cm, respectively) for operational taxonomic units (OTUs) sequencing

385 performed by Novogene (Tianjin, China). Total DNAs from these samples were

386 extracted by the CTAB/SDS method28 and were diluted to 1 ng/µL with sterile water

387 and used for PCR template. 16S rRNA genes of distinct regions (16S V3/V4) were

388 amplified using specific primers (341F: 5’- CCTAYGGGRBGCASCAG and 806R:

389 5’- GGACTACNNGGGTATCTAAT) with the barcode. These PCR products were

390 purified with a Qiagen Gel Extraction Kit (Qiagen, Germany) and prepared to

391 construct libraries. Sequencing libraries were generated using TruSeq® DNA

392 PCR-Free Sample Preparation Kit (Illumina, USA) following the manufacturer’s

393 instructions. The library quality was assessed on the Qubit@ 2.0 Fluorometer

394 (Thermo Scientific) and Agilent Bioanalyzer 2100 system. And then the library was

395 sequenced on an Illumina NovaSeq platform and 250 bp paired-end reads were

396 generated. Paired-end reads were merged using FLASH (V1.2.7,

397 http://ccb.jhu.edu/software/FLASH/)29, which was designed to merge paired-end

398 reads when at least some of the reads overlap those generated from the opposite end

399 of the same DNA fragments, and the splicing sequences were called raw tags. Quality

400 filtering on the raw tags were performed under specific filtering conditions to obtain

401 the high-quality clean tags30 according to the QIIME (V1.9.1,

402 http://qiime.org/scripts/split_libraries_fastq.html) quality controlled process. The tags

403 were compared with the reference database (Silva database, https://www.arb-silva.de/)

404 using UCHIME algorithm (UCHIME Algorithm,

405 http://www.drive5.com/usearch/manual/uchime_algo.html)31 to detect chimera

406 sequences, and then the chimera sequences were removed32. And sequences analyses

407 were performed by Uparse software (Uparse v7.0.1001, http://drive5.com/uparse/)33.

408 Sequences with ≥97% similarity were assigned to the same OTUs. The representative

409 sequence for each OTU was screened for further annotation. For each representative

410 sequence, the Silva Database (http://www.arb-silva.de/)34 was used based on Mothur

411 algorithm to annotate taxonomic information.

412 Metagenomic sequencing, assembly, binning and annotation. We selected three

413 cold seep sediment samples (C1, C2 and C4, 20 g each) for metagenomic analysis in

414 BGI (BGI, China). Briefly, total DNAs from these samples were extracted using the a

415 Qiagen DNeasy® PowerSoil® Pro Kit (Qiagen, Germany) and the integrity of DNA

416 was evaluated by gel electrophoresis. 0.5 μg DNA of each sample was used for

417 libraries preparation with an amplification step. Thereafter, DNA was cleaved into 50

418 ~ 800 bp fragments using Covaris E220 ultrasonicator (Covaris, UK) and some

419 fragments between 150 ~ 250 bp were selected using AMPure XP beads (Agencourt,

420 USA) and repaired using T4 DNA polymerase (ENZYMATICS, USA). All NGS

421 sequencing was performed on the BGISEQ-500 platform (BGI, China), generating

422 100 bp paired-end raw reads. Quality control was performed by SOAPnuke (v1.5.6)

423 (setting: -l 20 -q 0.2 -n 0.05 -Q 2 -d -c 0 -5 0 -7 1)35 and the clean data were

424 assembled using MEGAHIT (v1.1.3) (setting:--min-count 2 --k-min 33 --k-max 83

425 --k-step 10)36.

426 Isolation and cultivation of deep-sea Izimaplasma. To enrich the Izimaplasma

427 bacteria, the sediment samples were cultured at 28 °C for six months in an anaerobic

428 enrichment medium containing (per liter seawater): 1.0 g NH4Cl, 1.0 g NaHCO3, 1.0

. 429 g CH3COONa, 0.5 g KH2PO4, 0.2 g MgSO4 7H2O, 1.0 mg E. coli genomic DNA, 0.7

430 g cysteine hydrochloride, 500 µL 0.1 % (w/v) resazurin and pH 7.0. This medium was

431 prepared anaerobically as previously described37. During the course of enrichment,

432 1.0 mg genomic DNA was added once a month. After a six-month enrichment, 50 µL

433 dilution portion was spread on Hungate tube covered by modified organic medium

434 named ORG in this study, which contains 1 g yeast extract, 1 g peptone, 1 g NH4Cl, 1

. 435 g NaHCO3, 1 g CH3COONa, 0.5 g KH2PO4, 0.2 g MgSO4 7H2O, 0.7 g cysteine

436 hydrochloride, 500 µL 0.1 % (w/v) resazurin, 1 L seawater, 15 g agar, pH7.0. The

437 Hungate tubes were anaerobically incubated at 28 °C for 7 days. Individual colonies

438 with distinct morphology were picked using sterilized bamboo sticks and then

439 cultured in the ORG broth. Strain zrk13 was isolated and purified with ORG medium

440 by repeated use of the Hungate roll-tube methods for several rounds until it was

441 considered to be axenic. The purity of strain zrk13 was confirmed routinely by TEM

442 and repeated partial sequencing of the 16S rRNA gene. Strain zrk13 is preserved at

443 -80 °C in ORG broth supplemented with 20% (v/v) glycerol.

444 The temperature, pH and NaCl concentration ranges for the growth of strain

445 zrk13 were determined in the ORG broth with incubation at 28 °C for 14 days.

446 Growth assays were performed at different temperatures (4, 16, 28, 30, 37, 45, 60, 70,

447 80 °C). The pH range for growth was tested from pH 2.0 to pH 10.0 with increments

448 of 0.5 pH units. Salt tolerance was tested in the modified ORG broth (with distilled

449 water) supplemented with 0-10 % (w/v) NaCl (0.5 % intervals). Substrates utilization

450 of strain zrk13 was tested in the medium (pH 7.0) consisting of (L-1): 5 g NaCl, 1 g

451 NH4Cl, 0.5 g KH2PO4, 0.2 g MgSO4, 0.02 g yeast extract. Single substrate (including

452 glucose, maltose, butyrate, fructose, sucrose, acetate, formate, starch, isomaltose,

453 trehalose, galactose, cellulose, xylose, lactate, ethanol, D-mannose, glycerin,

454 rhamnose and sorbitolurea) was added from sterile filtered stock solutions to the final

455 concentration at 20 mM, respectively. Cell culture containing only 0.02 g yeast

456 extract (L-1) without adding any other substrates was used as a control. All the

457 cultures were incubated at 28 °C for 14 days. For each substrate, three biological

458 replicates were performed.

459 TEM observation. To observe the morphological characteristics of zrk13, its cell

460 suspension was washed with Milli-Q water and centrifuged at 5,000 × g for 5 min,

461 and taken by immersing copper grids coated with a carbon ﬁlm for 20 min, washed

462 for 10 min in distilled water and dried for 20 min at room temperature38.

463 Ultrathin-section electron microscopic observation was performed as described

464 previously39,40. Briefly, the samples were firstly preserved in 2.5% (v/v)

465 glutaraldehyde overnight at 4 °C, washed three times with PBS and dehydrated in

466 ethanol solutions of 30%, 50%, 70%, 90% and 100% for 10 min each time, and then

467 the samples were embedded in a plastic resin. Ultrathin sections (50~70 nm) of cells

468 were prepared with an ultramicrotome (Leica EM UC7), stained with uranyl acetate

469 and lead citrate. All samples were examined using TEM (HT7700, Hitachi, Japan)

470 with a JEOL JEM 12000 EX (equipped with a ﬁeld emission gun) at 100 kV.

471 Genome sequencing and genomic characteristics of strain zrk13. Genomic DNA

472 was extracted from strain zrk13 cultured for 6 days at 28 °C . The DNA library was

473 prepared using the Ligation Sequencing Kit (SQK-LSK109), and sequenced using a

474 FLO-MIN106 vR9.4 flow-cell for 48 h on MinKNOWN software v1.4.2 (Oxford

475 Nanopore Technologies (ONT), United Kingdom). Whole-genome sequence

476 determinations of strain zrk13 were carried out with the Oxford Nanopore MinION

477 (Oxford, United Kingdom) and Illumina MiSeq sequencing platform (San Diego, CA).

478 A hybrid approach was utilized for genome assembly using reads from both platforms.

479 Base-calling was performed using Albacore software v2.1.10 (Oxford Nanopore

480 Technologies). Nanopore reads were processed using protocols toolkit for quality

481 control and downstream analysis41. Filtered reads were assembled using Canu version

482 1.842 using the default parameters for Nanopore data. And then the genome was

483 assembled into a single contig and was manually circularized by deleting an

484 overlapping end.

485 The genome relatedness values were calculated by multiple approaches: Average

486 Nucleotide Identity (ANI) based on the MUMMER ultra-rapid aligning tool (ANIm),

487 ANI based on the BLASTN algorithm (ANIb), the tetranucleotide signatures (Tetra),

488 and in silico DNA-DNA similarity. ANIm, ANIb, and Tetra frequencies were

489 calculated using JSpecies WS (http://jspecies.ribohost.com/jspeciesws/). The

490 recommended species criterion cut-offs were used: 95% for the ANIb and ANIm and

491 0.99 for the Tetra signature. The amino acid identity (AAI) values were calculated by

492 AAI-profiler (http://ekhidna2.biocenter.helsinki.fi/AAI/). The in silico DNA-DNA

493 similarity values were calculated by the Genome-to-Genome Distance Calculator

494 (GGDC) (http://ggdc.dsmz.de/)43. The isDDH results were based on the recommended

495 formula 2, which is independent of genome size and, thus, is robust when using

496 whole-genome sequences. The prediction of genes involved in DNA degradation for

497 individual genomes was performed using Galaxy (Galaxy Version 2.6.0,

498 https://galaxy.pasteur.fr/)44 with the NCBI BLAST+ blastp method.

499 Phylogenetic analysis. The 16S rRNA gene tree was constructed with the full-length

500 16S rRNA sequences by the neighbor-joining algorithm, maximum likelihood and

501 minimum-evolution methods. The full-length 16S rRNA gene sequence (1,537 bp) of

502 strin zrk13 was obtained from the genome (accession number MW132883), and other

503 related taxa used for phylogenetic analysis were obtained from NCBI

504 (www.ncbi.nlm.nih.gov/). Phylogenetic trees were constructed using W-IQ-TREE

505 web server (http://iqtree.cibiv.univie.ac.at)45 with LG+F+I+G4 model. The online tool

506 Interactive Tree of Life (iTOL v4)46 was used for editing the tree.

507 Growth assay of X. coldseepsis zrk13. Growth assays were performed at

508 atmospheric pressure. Briefly, 16 mL strain zrk13 culture was inoculated in 2 L

509 Hungate bottles containing 1.6 L of inorganic medium supplemented with or without

510 1.0 mg/L E. coli genomic DNA, ORG medium supplemented with 100 mM Na2S2O3

511 and oligotrophic medium (0.1 g/L yeast extract, 0.1 g/L peptone, 1.0 g/L NH4Cl, 1.0

. 512 g/L NaHCO3, 1.0 g/L CH3COONa, 0.5 g/L KH2PO4, 0.2 g/L MgSO4 7H2O, 0.7 g/L

513 cysteine hydrochloride, 500 µL/L of 0.1 % (w/v) resazurin and pH 7.0), respectively.

514 These Hungate bottles were anaerobically incubated at 28 °C. Bacterial growth status

515 was monitored by measuring the OD600 value every day until cell growth reached the

516 stationary phase. Since strain zrk13 grew very slow in the inorganic medium, we also

517 adopted the quantitative PCR (qPCR) to measure its growth status.

518 Detection of DNA degradation by strain zrk13. Electrophoresis was performed to

519 detect the ability of DNA degradation of strain zrk13. Briefly, the E. coli genomic

520 DNA and RNA were extracted from overnight cultured E. coli DH5α cells with a

521 Genomic DNA and RNA Kit (Tsingke, China) by omiting the RNA removal step. The

522 reaction system (20 µL) contained 2 µL bacterial supernatant, 2 µL buffer and 16 µL

523 E. coli genomic DNA and rRNA extracts (2 µg). In parallel, the supernatant of a

524 Clostridia bacterium was used as a control. All bacterial supernatants were obtained

525 by centrifugation with 12,000 g for 10 min from 10-day cultures. Assays were

526 performed at 37 °C for 5 min, 10 min and 15 min, respectively. Finally, the reaction

527 solutions were detected by 1% agarose gel electrophoresis at 180 V for 20 min. The

528 imaging of the gel was taken by the Gel Image System (Tanon 2500, China).

529 Quantitative real-time PCR assay. For qRT-PCR, cells of strain zrk13 were

530 cultured in inorganic medium supplemented with or without 1.0 mg/L E. coli genomic

531 DNA or organic medium at 28 °C for 15 days with harvesting cells from 100 mL

532 medium every day. Total RNAs from each sample were extracted using the Trizol

533 reagent (Solarbio, China) and the RNA concentration was measured using Qubit®

534 RNA Assay Kit in Qubit® 2.0 Flurometer (Life Technologies, CA, USA). Then RNA

535 from corresponding sample was reverse transcribed into cDNA and the transcriptional

536 levels of different genes were determined by qRT-PCR using SybrGreen Premix Low

537 rox (MDbio, China) and the QuantStudioTM 6 Flex (Thermo Fisher Scientific, USA).

538 The PCR condition was set as following: initial denaturation at 95 °C for 3 min,

539 followed by 40 cycles of denaturation at 95 °C for 10 s, annealing at 60 °C for 30 s,

540 and extension at 72 °C for 30 s. 16S rRNA was used as an internal reference and the

541 gene expression was calculated using the 2-ΔΔCt method, with each transcript signal

542 normalized to that of 16S rRNA. Transcript signals for each treatment were compared

543 to those of control group. Speciﬁc primers for genes related to DNA degradation and

544 16S rRNA were designed using Primer 5.0 as shown in Supplementary Table 4. The

545 standard curve was generated using five independent dilution series of plasmid DNA

546 carrying a fragment (143 bp) of the strain zrk13T 16S rRNA gene. This qPCR assay

547 standard curve had a slope of -2.5162, a y-intercept of 30.53 and an R2 value of

548 0.9924. All qRT-PCR runs were performed in three biological and three technical

549 replicates.

550 Transcriptomic analysis. Transcriptomic analysis was performed by Novogene

551 (Tianjin, China). Briefly, cells suspension of X. coldseepsis zrk13 cultured in

552 oligotrophic medium, ORG medium supplemented with or without 100 mM Na2S2O3

553 for 7 days was respectively collected for further transcriptomic analysis. All cultures

554 were performed in 2 L anaerobic bottles. For in situ experiments, X. coldseepsis zrk13

555 was firstly cultured in ORG medium for 5 days, and then was divided into two parts:

556 one part was divided into three gas samples bags (which not allowing any exchanges

557 between inside and outside; Aluminum-plastic composite film, Hede, China) and used

558 as the control group; the other part was divided into three dialysis bags (8,000-14,000

559 Da cutoff, which allowing the exchanges of substances smaller than 8,000 Da but

560 preventing bacterial cells from entering or leaving the bag; Solarbio, China) and used

561 as the experimental group. All the samples were placed simultaneously in the

562 deep-sea cold seep (E119°17′04.429″, N22°07′01.523″) for 10 days in the June 2020

563 during the cruise of Kexue vessel. After 10 days incubation in the deep sea, the bags

564 were taken out and the cells were immediately collected and saved in the -80 freezer

565 until further use. Thereafter, the cells were checked by 16S rRNA sequencing to

566 confirm the purity of the culture and performed further transcriptomic analysis. For

567 transcriptomic analyses, total X. coldseepsis zrk13 RNAs were extracted using TRIzol

568 reagent (Invitrogen, USA) and DNA contamination was removed using the MEGA

569 clear™ Kit (Life technologies, USA). Detailed protocols of the following procedures

570 including library preparation, clustering and sequencing and data analyses were

571 described in the Supplementary Information.

572 Data availability. The whole 16S rRNA sequence of X. coldseepsis zrk13 has

573 been deposited in the GenBank database (accession number MW132883). The

574 complete genome sequence of X. coldseepsis zrk13 has been deposited at GenBank

575 under the accession number CP048914. The raw sequencing reads for transcriptomic

576 analysis have been deposited to NCBI Short Read Archive (accession numbers:

577 PRJNA664657, PRJNA669477 and PRJNA669478). The raw amplicon sequencing

578 data have also been deposited to NCBI Short Read Archive (accession number:

579 PRJNA675395).

580 Acknowledgements

581 This work was funded by the Strategic Priority Research Program of the Chinese

582 Academy of Sciences (Grant No. XDA22050301), China Ocean Mineral Resources

583 R&D Association Grant (Grant No. DY135-B2-14), National Key R and D Program

584 of China (Grant No. 2018YFC0310800), the Taishan Young Scholar Program of

585 Shandong Province (tsqn20161051), and Qingdao Innovation Leadership Program

586 (Grant No. 18-1-2-7-zhc) for Chaomin Sun. This study is also funded by the Open

587 Research Project of National Major Science & Technology Infrastructure (RV KEXUE)

588 (Grant No. NMSTI-KEXUE2017K01).

589 Author contributions

590 RZ and CS conceived and designed the study; RZ conducted most of the experiments;

591 RL, YS and GL collected the samples from the deep-sea cold seep; RC helped to

592 analyze the metagenomes; RZ and CS lead the writing of the manuscript; all authors

593 contributed to and reviewed the manuscript.

594 Conflict of interest

595 The authors declare that there are no any competing financial interests in relation to

596 the work described.

597 References

598 1. Hug, L. A. et al. A new view of the tree of life. Nat Microbiol 1 (2016). 599 2. Wang, Y. et al. Phylogenomics of expanding uncultured environmental Tenericutes 600 provides insights into their pathogenicity and evolutionary relationship with Bacilli. BMC 601 Genomics 21 (2020). 602 3. Castelle, C. J. & Banfield, J. F. Major new microbial groups expand diversity and alter our 603 understanding of the tree of life. Cell 172, 1181-1197 (2018). 604 4. Skennerton, C. T. et al. Phylogenomic analysis of Candidatus 'Izimaplasma' species: 605 free-living representatives from a Tenericutes clade found in methane seeps. ISME J. 10, 606 2679-2692 (2016). 607 5. Claessen, D. & Errington, J. Cell wall deficiency as a coping strategy for stress. Trends 608 Microbiol 27, 1025-1033 (2019). 609 6. Miyata, M. & Ogaki, H. Cytoskeleton of Mollicutes. J. Mol. Microb. Biotech. 11, 256-264 610 (2006). 611 7. Ku, C., Lo, W. S. & Kuo, C. H. Molecular evolution of the actin-like MreB protein gene 612 family in wall-less bacteria. Biochem Bioph Res Co 446, 927-932 (2014).

613 8. Cao, Y. H., Trivellone, V. & Dietrich, C. H. A timetree for phytoplasmas (Mollicutes) 614 with new insights on patterns of evolution and diversi fication. Mol. Phylogenet. Evol. 149 615 (2020). 616 9. Clark, G. W. & Tillier, E. R. M. Loss and gain of GroEL in the Mollicutes. Biochem Cell 617 Biol 88, 185-194 (2010). 618 10.Parte, A. C. LPSN - list of prokaryotic names with standing in nomenclature (bacterio.net), 619 20 years on. Int. J. Syst. Evol. Micr. 68, 1825-1829 (2018). 620 11. Trachtenberg, S. Mollicutes. Curr. Biol. 15, R483-R484 (2005). 621 12. Ermolaeva, S. A. et al. Nonthermal plasma affects viability and morphology of 622 Mycoplasma hominis and Acholeplasma laidlawii. J Appl Microbiol 116, 1129-1136 623 (2014). 624 13. Razin, S., Yogev, D. & Naot, Y. Molecular biology and pathogenicity of mycoplasmas. 625 Microbiol. Mol. Biol. R. 62, 1094-1156 (1998). 626 14. Antunes, A. et al. A new lineage of halophilic, wall-less, contractile bacteria from a 627 brine-filled deep of the Red Sea. J. Bacteriol. 190, 3580-3587 (2008). 628 15. Wasmund, K. et al. DNA-foraging bacteria in the seafloor. bioRxiv, 528695 (2019). 629 16. Novitsky, J. A. & Karl, D. M. Influence of deep ocean sewage outfalls on the microbial 630 activity of the surrounding sediment. Appl. Environ. Microb. 50, 1464-1473 (1985). 631 17. Danovaro, R., Dell'anno, A., Pusceddu, A. & Fabiano, M. Nucleic acid concentrations 632 (DNA, RNA) in the continental and deep-sea sediments of the eastern Mediterranean: 633 relationships with seasonally varying organic inputs and bacterial dynamics. Deep-Sea 634 Res. Pt. I 46, 1077-1094 (1999). 635 18. Dell'Anno, A., Fabiano, M., Duineveld, G. C. A., Kok, A. & Danovaro, R. Nucleic acid 636 (DNA, RNA) quantification and RNA/DNA ratio determination in marine sediments: 637 Comparison of spectrophotometric, fluorometric, and high-performance liquid 638 chromatography methods and estimation of detrital DNA. Appl. Environ. Microb. 64, 639 3238-3245 (1998). 640 19. Efremov, R. G. & Sazanov, L. A. Structure of the membrane domain of respiratory 641 complex I. Nature 476, 414-U462 (2011). 642 20. Walker, J. E. The Nadh - ubiquinone oxidoreductase (complex-I) of respiratory chains. Q. 643 Rev. Biophys. 25, 253-324 (1992). 644 21. Zhang, J. et al. A novel bacterial thiosulfate oxidation pathway provides a new clue about 645 the formation of zero-valent sulfur in deep sea. ISME J. 14, 2261-2274 (2020). 646 22. Gao, H. Y. et al. Arabidopsis thaliana Nfu2 accommodates [2Fe-2S] or [4Fe-4S] clusters 647 and is competent for in vitro maturation of chloroplast [2Fe-2S] and [4Fe-4S] 648 cluster-containing proteins. Biochemistry-Us 52, 6633-6645 (2013). 649 23. Dell'Anno, A. & Danovaro, R. Extracellular DNA plays a key role in deep-sea ecosystem 650 functioning. Science 309, 2179-2179 (2005). 651 24. Baker, B. J., Appler, K. E. & Gong, X. New microbial biodiversity in marine sediments. 652 Ann. Rev. Mar. Sci. (2020). 653 25. Lewis, W. H., Tahon, G., Geesink, P., Sousa, D. Z. & Ettema, T. J. G. Innovations to 654 culturing the uncultured microbial majority. Nat. Rev. Microbiol. (2020).

655 26. Zheng, R. K. & Sun, C. M. Sphingosinithalassobacter tenebrarum sp. nov., isolated from 656 a deep-sea cold seep. Int. J. Syst. Evol. Micr. 70, 5561-5566 (2020). 657 27. Corinaldesi, C., Dell'Anno, A. & Danovaro, R. Early diagenesis and trophic role of 658 extracellular DNA in different benthic ecosystems. Limnol Oceanogr 52, 1710-1717 659 (2007). 660 28. Murray, M. G. & Thompson, W. F. Rapid isolation of high molecular-weight plant DNA. 661 Nucleic Acids Res. 8, 4321-4325 (1980). 662 29. Magoc, T. & Salzberg, S. L. FLASH: fast length adjustment of short reads to improve 663 genome assemblies. Bioinformatics 27, 2957-2963 (2011). 664 30. Bokulich, N. A. et al. Quality-filtering vastly improves diversity estimates from Illumina 665 amplicon sequencing. Nat. Methods 10, 57-U11 (2013). 666 31. Edgar, R. C., Haas, B. J., Clemente, J. C., Quince, C. & Knight, R. UCHIME improves 667 sensitivity and speed of chimera detection. Bioinformatics 27, 2194-2200 (2011). 668 32. Haas, B. J. et al. Chimeric 16S rRNA sequence formation and detection in Sanger and 669 454-pyrosequenced PCR amplicons. Genome Res. 21, 494-504 (2011). 670 33. Edgar, R. C. UPARSE: highly accurate OTU sequences from microbial amplicon reads. 671 Nat. Methods 10, 996-+ (2013). 672 34. Quast, C. et al. The SILVA ribosomal RNA gene database project: improved data 673 processing and web-based tools. Nucleic Acids Res. 41, D590-D596 (2013). 674 35. Chen, Y. X. et al. SOAPnuke: a MapReduce acceleration-supported software for 675 integrated quality control and preprocessing of high-throughput sequencing data. 676 Gigascience 7 (2017). 677 36. Li, D. H., Liu, C. M., Luo, R. B., Sadakane, K. & Lam, T. W. MEGAHIT: an ultra-fast 678 single-node solution for large and complex metagenomics assembly via succinct de Bruijn 679 graph. Bioinformatics 31, 1674-1676 (2015). 680 37. Fardeau, M. L. et al. Thermotoga hypogea sp. nov., a xylanolytic, thermophilic bacterium 681 from an oil-producing well. Int. J. Syst. Bacteriol. 47, 1013-1019 (1997). 682 38. Buchan, A., LeCleir, G. R., Gulvik, C. A. & Gonzalez, J. M. Master recyclers: features 683 and functions of bacteria associated with phytoplankton blooms. Nat. Rev. Microbiol. 12, 684 686-698 (2014). 685 39. Sekiguchi, Y. et al. Anaerolinea thermophila gen. nov., sp nov and Caldilinea aerophila 686 gen. nov., sp nov., novel filamentous thermophiles that represent a previously uncultured 687 lineage of the domain Bacteria at the subphylum level. Int. J. Syst. Evol. Micr. 53, 688 1843-1851 (2003). 689 40. Graham, L. & Orenstein, J. M. Processing tissue and cells for transmission electron 690 microscopy in diagnostic pathology and research. Nat. Protoc. 2, 2439-2450 (2007). 691 41. Loman, N. J. & Quinlan, A. R. Poretools: a toolkit for analyzing nanopore sequence data. 692 Bioinformatics 30, 3399-3401 (2014). 693 42. Koren, S. et al. Canu: scalable and accurate long-read assembly via adaptive k-mer 694 weighting and repeat separation. Genome Res. 27, 722-736 (2017).

695 43. Meier-Kolthoff, J. P., Auch, A. F., Klenk, H. P. & Goker, M. Genome sequence-based 696 species delimitation with confidence intervals and improved distance functions. BMC 697 Bioinformatics 14 (2013). 698 44. Afgan, E. et al. The Galaxy platform for accessible, reproducible and collaborative 699 biomedical analyses: 2018 update. Nucleic Acids Res. 46, W537-W544 (2018). 700 45. Trifinopoulos, J., Nguyen, L. T., von Haeseler, A. & Minh, B. Q. W-IQ-TREE: a fast 701 online phylogenetic tool for maximum likelihood analysis. Nucleic Acids Res. 44, 702 W232-W235 (2016). 703 46. Letunic, I. & Bork, P. Interactive Tree Of Life (iTOL) v4: recent updates and new 704 developments. Nucleic Acids Res. 47, W256-W259 (2019).

705

706

707

708

709

710

711

712

713

714

715

716

717

718

719

720 Figures

721

722 Fig. 1. DNA degradation-driven isolation strategy and morphology of

723 Xianfuyuplasma coldseepsis zrk13. a, Diagrammatic scheme of enrichment and

724 isolation of Izimaplasma bacteria. b-d, TEM observation of strain zrk13. e, f, TEM

725 observation of the ultrathin sections of strain zrk13. g, TEM observation of the

726 ultrathin sections of a typical Gram-positive bacterium Clostridium sp. zrk8. CM, cell

727 membrane; PG, peptidoglycan; Scale bars, 50 nm. h, Phylogenetic analysis of

728 Xianfuyuplasma coldseepsis zrk13. Phylogenetic placement of strain zrk13 within the

729 Tenericutes phylum based on almost complete 16S rRNA gene sequences. The tree is

730 inferred and reconstructed under the maximum likelihood criterion and bootstrap

731 values (%) > 80 are indicated at the base of each node (expressed as percentages of

732 1,000 replications). The black dots at a distinct node indicate that the tree was also

733 formed using the neighbor-joining and minimum-evolution methods. Grey names in

734 quotation represent taxa that are not yet validly published. The only two clades

735 (Mollicutes and Izimaplasma) of Tenericutes are indicated on the right. The sequence

736 of Bacillus pumilus SH-B9T is used as an outgroup. Bar, 0.1 substitutions per

737 nucleotide position.

738

739

740

741

742

743

744

745 Fig. 2. Organic nutrient significantly promotes the grwoth of X. coldseepsis zrk13.

746 a, Growth assays of strain zrk13 in the oligotrophic and rich media. b,

747 Transcriptomics based heat map showing all up-regulated genes encoding glycosyl

748 hydrolases. c, Transcriptomics based heat map showing three up-regulated iron

749 hydrogenase maturation genes (hydEFG). d, Transcriptomics based heat map showing

750 all up-regulated genes encoding NADH-quinone/ubiquinone oxidoreductase. e,

751 Transcriptomics based heat map showing all up-regulated genes encoding ATP

752 synthase. Oligo indicates the oligotrophic medium; rich indicates the rich medium.

753

754

755 Fig. 3. Thiosulfate significantly promotes the grwoth of X. coldseepsis zrk13. a,

756 Growth assays of strain zrk13 in the organic medium supplemented with or without

757 100 mM Na2S2O3. b, Transcriptomics based heat map showing all up-regulated genes

758 encoding 2Fe-2S/4Fe-4S-binding proteins. c, Transcriptomics based heat map

759 showing all up-regulated genes encoding glycosyl hydrolases. d, Transcriptomics

760 based heat map showing all up-regulated genes encoding sugar ABC transporter

761 permease.

762

763

764

765

766 Fig. 4. X. coldseepsis zrk13 possesses a significant capability of degrading and

767 utilization of extracellular DNA. a, Gene arrangements of a putative

768 DNA-degradation locus in strain zrk13. The numbers shown in the X-axis indicate the

769 gene number in the gene cluster. b, Detection of the ability of DNA degradation of

770 strain zrk13 by electrophoresis. Lane I, lane IV and lane VII indicate 2 µg E. coli

771 genomic DNA and rRNA without treatment. Lane II, lane V and lane VIII indicate 2

772 µg E. coli genomic DNA and rRNA treated by a Clostridia bacterial supernatant for 5

773 min, 10 min and 15 min at 37 °C , respectively. Lane III, lane VI and lane IX indicate

774 2 µg E. coli genomic DNA and rRNA treated by strain zrk13 supernatant for 5 min,

775 10 min and 15 min at 37 °C , respectively. M, DL4500 molecular weight DNA marker.

776 c, Growth assays of strain zrk13 cultivated in organic medium and inorganic medium

777 supplemented with or without 2 µg E. coli genomic DNA. d, qRT-PCR detection of

778 expression changes of genes shown in panel a when strain zrk13 cultivated in the

779 inorganic medium supplemented with or without 2 µg E. coli genomic DNA. Three

780 biological replicates were performed. The numbers shown in the X-axis indicate the

781 gene number shown in panel a. “C” indicates control.

782

783 Fig. 5. Transcriptomics analysis of X. coldseepsis zrk13 incubated in the deep-sea

784 cold seep. a, b, Distant (a) and close (b) views of the in situ experimental apparatus in

785 the deep-sea cold seep where distributed many mussels and shrimps. c,

786 Transcriptomics based heat map showing all up-regulated genes encoding enzymes

787 degrading nucleic acids after a 10-day incubation of strain zrk13 in the deep-sea cold

788 seep. d, Transcriptomics based heat map showing all up-regulated genes associated

789 with purine catabolism after a 10-day incubation of strain zrk13 in the deep-sea cold

790 seep. e, Diagrammatic scheme of EMP glycolysis pathway. The gene numbers

791 showing in this scheme are the same with those shown in panel f. f, Transcriptomics

792 based heat map showing all down-regulated genes associated with EMP glycolysis

793 pathway after a 10-day incubation of strain zrk13 in the deep-sea cold seep.

794

795

796

797

798

799

800

801

802

803

804

805

806

807 Fig. 6. Muti-omics based central metabolisms model of X. coldseepsis zrk13. In

808 this model, central metabolisms including DNA degradation, EMP glycolysis,

809 oxidative pentose phosphate pathway, hydrogen production, electron transport system,

810 sulfur metabolism and one carbon metabolism are shown. All the above items are

811 closely related to the energy production in X. coldseepsis zrk13. In detail, strain zrk13

812 contains a complete set of genes related to DNA-degradation, which could

813 catabolically exploit various sub-components of DNA, especially purine-based

814 molecules. And further degradation into nucleotides and nucleosides by nucleases

815 were required to facilitate the introduction of DNA sub-components into cells. Once

816 imported into the cytoplasm, purine- and pyrimidine-deoxyribonucleosides were

817 further broken down into different bases, whereby the respective bases may enter

818 catabolic pathways. The metabolites of nucleosides catabolism were finally

819 transformed into phosphoribosyl pyrophosphate (PRPP) and Ribose-1P, thus entering

820 the oxidative pentose phosphate pathway, which is closely related to EMP glycolysis

821 pathway. Sulfide generated by sulfur reduction from thiosulfate works together with

822 the L-serine to form acetate and L-cysteine, which eventually enter the pyruvate

823 synthesis pathway. Furthermore, the formate dehydrogenase linked with NADH may

824 convert formate into CO2/CO with the production of energy. Iron hydrogenases are

825 linked with NADP and thereby generating NADPH for the production of

826 biomolecules. A membrane-bound, Na+-transporting NADH: Ferredoxin

827 oxidoreductase (RNF complex), the H+-transporting NADH: Quinone oxidoreductase

828 (complex I) and F-type ATP synthase required for energy metabolism are present in

829 strain zrk13 genome. Both complex I and the cytochrome bd oxidase interact with the

830 quinone pool, which are associated with energy production.

831

832

833

834

835 42

836 Supplementary methods

837 Transcriptional profiling of X. coldseepsis zrk13 cultured in different conditions

838 (1) Library preparation for strand-specific transcriptome sequencing

839 A total amount of 3 μg RNA per sample was used as input material for the RNA

840 sample preparations. Sequencing libraries were generated using NEBNext® Ultra™

841 Directional RNA Library Prep Kit for Illumina® (NEB, USA) following

842 manufacturer’s recommendations and index codes were added to attribute sequences

843 to each sample. rRNA is removed using a specialized kit that leaves the mRNA.

844 Fragmentation was carried out using divalent cations under elevated temperature in

845 NEBNext First Strand Synthesis Reaction Buffer (5). First strand cDNA was

846 synthesized using random hexamer primer and M-MuLV Reverse Transcriptase

847 (RNaseH-). Second strand cDNA synthesis was subsequently performed using DNA

848 Polymerase I and RNase H. In the reaction buffer, dNTPs with dTTP were replaced

849 by dUTP. Remaining overhangs were converted into blunt ends via

850 exonuclease/polymerase activities. After adenylation of 3’ ends of DNA fragments,

851 NEBNext Adaptor with hairpin loop structure was ligated to prepare for hybridization.

852 In order to select cDNA fragments of preferentially 150~200 bp in length, the library

853 fragments were purified with AMPure XP system (Beckman Coulter, Beverly, USA).

854 Then 3 μL USER Enzyme (NEB，USA) was used with size-selected, adaptor-ligated

855 cDNA at 37 °C for 15 min followed by 5 min at 95 °C before PCR. Then PCR was

856 performed with Phusion High-Fidelity DNA polymerase, Universal PCR primers and

857 Index (X) Primer. At last, products were purified (AMPure XP system) and library

858 quality was assessed on the Agilent Bioanalyzer 2100 system.

859 (2) Clustering and sequencing

860 The clustering of the index-coded samples was performed on a cBot Cluster

861 Generation System using TruSeq PE Cluster Kit v3-cBot-HS (Illumia) according to

862 the manufacturer’s instructions. After cluster generation, the library preparations were

863 sequenced on an Illumina Hiseq platform and paired-end reads were generated.

864 (3) Data analysis

865 Raw data (raw reads) of fastq format were firstly processed through in-house perl

866 scripts. In this step, clean data (clean reads) were obtained by removing reads

867 containing adapter, reads containing ploy-N and low quality reads from raw data. At

868 the same time, Q20, Q30 and GC content the clean data were calculated. All the

869 downstream analyses were based on the clean data with high quality. Reference

870 genome and gene model annotation files were downloaded from genome website

871 directly. Both building index of reference genome and aligning clean reads to

872 reference genome were used Bowtie2-2.2.31. HTSeq v0.6.1 was used to count the

873 reads numbers mapped to each gene. And then FPKM of each gene was calculated

874 based on the length of the gene and reads count mapped to this gene. FPKM, expected

875 number of Fragments Per Kilobase of transcript sequence per Millions base pairs

876 sequenced, considers the effect of sequencing depth and gene length for the reads

877 count at the same time, and is currently the most commonly used method for

878 estimating gene expression levels2.

879 (4) Differential expression analysis

880 Differential expression analysis of two conditions/groups (two biological replicates

881 per condition) was performed using the DESeq R package (1.18.0)3. DESeq provide

882 statistical routines for determining differential expression in digital gene expression

883 data using a model based on the negative binomial distribution. The resulting

884 P-values were adjusted using the Benjamini and Hochberg’s approach for controlling

885 the false discovery rate. Genes with an adjusted P-value < 0.05 found by DESeq were

886 assigned as differentially expressed. (For DEGSeq without biological replicates) Prior

887 to differential gene expression analysis, for each sequenced library, the read counts

888 were adjusted by edgeR program package through one scaling normalized factor.

889 Differential expression analysis of two conditions was performed using the DEGSeq

890 R package (1.20.0)4. The P values were adjusted using the Benjamini & Hochberg

891 method. Corrected P-value of 0.005 and log2 (Fold change) of 1 were set as the

892 threshold for significantly differential expression.

893 (5) GO and KEGG enrichment analysis of differentially expressed genes

894 Gene Ontology (GO) enrichment analysis of differentially expressed genes was

895 implemented by the GOseq R package, in which gene length bias was corrected5. GO

896 terms with corrected P value less than 0.05 were considered significantly enriched by

897 differential expressed genes. KEGG is a database resource for understanding

898 high-level functions and utilities of the biological system, such as the cell, the

899 organism and the ecosystem, from molecular-level information, especially large-scale

900 molecular datasets generated by genome sequencing and other high-throughput

901 experimental technologies (http://www.genome.jp/kegg/)6. We used KOBAS software

902 to test the statistical enrichment of differential expression genes in KEGG pathways.

903 Supplementary references 904 1. Langmead, B. & Salzberg, S. L. Fast gapped-read alignment with Bowtie 2. Nat Methods 905 9, 357-U354 (2012). 906 2. Trapnell, C., Pachter, L. & Salzberg, S. L. TopHat: discovering splice junctions with 907 RNA-Seq. Bioinformatics 25, 1105-1111 (2009). 908 3. Anders, S. & Huber, W. Differential expression analysis for sequence count data. 909 Genome Biol 11 (2010). 910 4. Wang, L. K., Feng, Z. X., Wang, X., Wang, X. W. & Zhang, X. G. DEGseq: an R 911 package for identifying differentially expressed genes from RNA-seq data. 912 Bioinformatics 26, 136-138 (2010). 913 5. Young, M. D., Wakefield, M. J., Smyth, G. K. & Oshlack, A. Gene ontology analysis for 914 RNA-seq: accounting for selection bias. Genome Biol 11 (2010). 915 6. Kanehisa, M. et al. KEGG for linking genomes to life and the environment. Nucleic Acids 916 Res 36, D480-D484 (2008). 917

918

919

920

921

922

923

924

925

926

927

928 Supplementary results

929

930 Supplementary Fig 1. Quantification of the abundance of Tenericutes and

931 Candidatus Izimaplasma in the deep-sea cold seep. a, The community structure of

932 six sampling sites in the cold seep sediments as revealed by 16S rRNA gene amplicon

933 profiling. The relative abundances of operational taxonomic units (OTUs)

934 representing different bacteria are shown at the phylum level. b, Quantification of the

935 abundance of Tenericutes in the bacteria domain based on the metagenomics

936 sequencing. c, Quantification of the abundance of Candidatus Izimaplasma in the

937 phylum Tenericutes based on the metagenomics sequencing.

938

939

940 Supplementary Fig 2. Circular diagram of the genome of X.coldseepsis zrk13. Rings

941 indicate, from outside to the center: a genome-wide marker with a scale of 50 kb;

942 forward strand genes, colored by COG category; reverse strand genes, colored by

943 COG category; repetitive sequences; RNA genes (tRNAs blue, rRNAs purple); GC

944 content; GC skew.

945

946

947

948

949 Supplementary Fig 3. The picture of ‘Xianfuyu’ comes from a strange animal

950 recorded in a very famous Chinese mythology-Classic of Mountains and Rivers.

951 The ‘Xianfuyu’ is the father of all the fishes, which looks like gouramies, with a fish

952 head and a pig body.

953

954

955

956

957 Supplementary Fig 4. Distribution of key genes associated with DNA

958 degradation in the the genomes of typical species of phylums Tenericutes and

959 Firmicutes.

960

961

962

963

964

965

966

967 50

968 Supplementary Table 1. Genomic features of X. coldseepsis zrk13 with other

969 uncultivated Izimaplasma strains HR1, HR2 and ZiA1.

Feature zrk13 HR1 HR2 ZiA1 Gene Bank ID CP048914 CP009415 JRFF00000000 NQYJ00000000 Genome size (bp) 1,958,905 1,878,735 2,115,618 1,884,011 G+C content (%) 38.2 31.3 29.2 29.6 No.scaffolds/contigs 1 1 78 22 No. of genes 1,872 1,846 2,284 1,877 No. of rRNAs 3 4 3 0 No. of tRNAs 42 38 58 34 No. of protein-coding genes 1,762 1,794 2,222 1,803 Completeness (%) 100 100 92.1 98.7 AAI (%) 100 68.8 66.5 64.3 ANIb (%) 100 68.72 67.42 66.57 ANIm (%) 100 85.94 86.53 81.98 Tetra 1 0.85192 0.77694 0.837 isDDH (%) 100 17.50 18.50 15.80 970

971

972

973

974

975

976

977

978

979

980

981

982

983

984

985 Supplementary Table 2. Physiological characteristics of X. coldseepsis zrk13.

Characteristic zrk13

Cell length (µm) 0.2-0.5 Temperature range for growth (°C ) 24-32 Optimum 28 pH range for growth 6.0-8.0 Optimum 7.0 NaCl range for growth (%) 0-8 Optimum 4 Utilization as a sole carbon source: Glucose + Maltose + Butyrate + Fructose + Sucrose + Acetate + Formate + Starch + Isomaltose + Trehalose + Galactose - Cellulose - Xylose - Lactate + Ethanol + D-mannose + Glycerin + Rhamnose + Sorbitol - DNA G+C content (mol%) 38.21 % Isolation source deep-sea sediments

986

987