The Microcystis-microbiome interactions: origins of the colonial lifestyle

Abstract Species of the Microcystis genus are the most common bloom-forming toxic cyanobacteria worldwide. They belong to a clade of unicellular cyanobacteria whose ability to reach high biomasses during blooms is linked to the formation of colonies. Colonial lifestyle provides several advantages under stressing conditions of light intensity, ultraviolet light, toxic substances and grazing. The progression from a single-celled organism to multicellularity in Microcystis has usually been interpreted as individual phenotypic responses of the cyanobacterial cells to the environment. Here, we synthesize current knowledge about Microcystis colonial lifestyle and its role in the organism ecology. We then briefly review the available information on Microcystis microbiome and propose that changes leading from single cells to colonies are the consequence of specific and tightly regulated signals between the cyanobacterium and its microbiome through a biofilm-like mechanism. The resulting colony is a multi-specific community of interdependent microorganisms.


Introduction
It has been since a long time ago that microbiologists have noticed that bacteria do not always live as single cells.Many of the known bacterial species have the ability to grow in a multicellular and coor dinate w a y, the biofilms .Bacterial biofilms are defined as aggr egates of micr obial cells surr ounded by a self-pr oduced pol ymer matrix that can be composed by a single (mono-specific) or sever al species (m ulti-specific) living in a collabor ativ e way.Biofilm gr owth of micr oor ganisms was first defined in medical microbiology, when it was also demonstrated that biofilm-embedded organisms have an increased antimicrobial resistance compared to those growing as planktonic bacteria (Nickel et al. 1985 ).
The classic conceptual model of biofilm formation involves motile planktonic cells that become attached to a surface in response to a variety of environmental signals (Sauer et al. 2022 ).Attac hed cells pr oduce a hydr ated matrix of extr acellular pol ysacc harides (EPS), extr acellular DNA, pr oteins and lipids (Flemming and Wingender 2010 ), changing their structure and functional relationships .After a while , sessile cells arr anged in micr ocolonies fr om wher e some cells can esca pe to r eturn to the planktonic lifestyle and subsequently colonize a new surface (Petrova and Sauer 2016 ).Although biofilm cells encounter stronger gradients of nutrients and waste products than during planktonic life, they are embedded in a more controllable environment (Stewart and Franklin 2008 ).
In the case of aquatic c y anobacteria, despite the increasing amount of information regarding their ecology, the biofilm con-cept is gener all y associated with benthic species, whic h form mats in se v er al aquatic ecosystems (Stal 2012 ).Among the planktonic groups we will focus on Microcystis spp., a complex of cyanobacteria from the Chroococcales order that live in freshwater and br ac kish waters .T hey form dense blooms in eutrophic ecosystems (Paerl 1988, Huisman et al. 2018 ) and can be found as single cells or in colonies floating near the surface, with a size spectrum ranging from ca. 4 μm (single cells) to hundreds of microns (large colonies) (Reynolds et al. 1981 ) that can be detected by naked eye.
Inter estingl y, Microcystis belongs to a phylogenetic group of unicellular c y anobacteria and its ability to form colonies is usually considered as an ecological aggregation strategy to avoid predation or protect from ultraviolet radiation, among others.In this context, colony formation by these organisms has been explained either by cell division (the usual bacterial process to m ultipl y) or cell adhesion (Yang et al. 2008 ).Ho w e v er, r ecent genomic evidence suggests that colonies in Microcystis result from clonal expansion rather than cell aggregation (Carrascal et al. 2021 ).
In spite of the amount of information regarding Microcystis ecology, colony formation and toxicity, little is known about the biological interactions taking place inside the colony and their role in Microcystis biology and e volution.Her e, we focus on (i) the c har acteristics shared by bacterial biofilms and Microcystis colonies; (ii) the current knowledge about colony formation process in Microcystis ; (iii) the evidence on the existence of quorum sensing (QS) in Microcystis and; (iv) the information about community composition and function of the colony-associated microbiota.Based on this, Figure 1.Proposal for the floating biofilm formation of Microcystis .Four phases can be distinguished during the de v elopment of a Microcystis biofilm according to its lifestyle (single celled vs. attached aggregate), EPS and microcystin production, presence of an established microbiome and autoinducers concentration (AHLs).Phase 1 is composed of single cells (4 μm diameter, green circles) having little amount of EPS mucilage, low levels of microcystin production and low levels of AHLs.Phase 2 starts with the initial attachment of dividing cells to each other to form a colony surrounded by a higher amount of EPS mucilage, cells probably mobilize inside the colony and they have low levels of microcystin synthesis while AHLs start to build up and other bacteria (smaller red, blue and black circles) start to be recluted and attached to the EPS.In Phase 3, the pr olifer ation of bacterial cells inside the colony and their interactions with c y anobacterial cells allow the formation of a mature biofilm, with elevated amounts of EPS, high le v els of micr ocystin pr oduction and clearl y differ ent metabolism between inner and outer cells.A micr obiome is well established.The Phase 4 is c har acterized by lar ge, amor phous colonies, low le v els of microcystin production and disaggregation of the mucilage by bacterial degradation of the EPS (typically at the end of a bloom).We propose that the onset of a bloom will depend on abiotic and biotic conditions and on the phase of the Microcystis community, being more likely to develop a high biomass in a short time period during phase 3 (active cells, with high microcystin pr oduction r ates).
we propose that the morphological, functional and microbiome compositional changes occurring from single cells to colonies are consequence of biological and ecological interactions between the c y anobacterium and the heter otr ophic bacteria.These specific and car efull y r egulated inter actions ar e bi-dir ectional and induce the de v elopment of a m ucila ginous env elope that will host the heter otr ophic comm unity thr ough a biofilm-like mec hanism.Taking this into account a conceptual model of emergence and decay of these floating multi-specific biofilms of Microcystis is presented.

Microcystis blooms and microcystins production
Microcystis blooms are composed by a mixture of populations able to produce secondary metabolites called microcystins that are toxic to animals and humans, and by non-toxic populations.It has been shown that high water temper atur e (between 25 and 30 • C) pr omotes the gr o wth of Microc ystis populations able to produce microcystin (to xic), while non-to xic populations seem to have less toler ance to v ariable envir onmental conditions (Davis et al. 2009, Van de Waal et al. 2011 ).Ther efor e, it is very likely that under the current climate warming and worldwide eutrophication scenario a dominance of c y anobacterial blooms containing a higher percentage of toxic Microcystis will occur (Paerl andHuisman 2008 , Kruk et al. 2023 ), making it very relevant to understand the biology and ecology of these organisms.
Until now, studies on the ecology of Microcystis have focused on the determinants of its growth, potential toxicity and diver-sity (Dick et al. 2021 ).More recently, the structure and function of its microbiome and its role in the survival and fitness of the c y anobacterium have started to be included (J anko wiak and Gobler 2020 , Schmidt et al. 2020, Carrascal et al. 2021 ).Ho w ever, there is no consensus on the mechanisms that determine the production of microcystins, the density and persistence of blooms or the microbiome community structure.In this sense, the evidence from different studies is frequently contradictory, since some works are based on axenic cultures of unicellular forms (hard to find in nature), others on environmental samples and others analyse and compare sequences obtained either from isolates , en vironmental DNA or enrichments from blooms, making generalizations difficult (Pimentel and Giani 2014, Martin et al. 2020, Zhou et al. 2020, Yang et al. 2022, Dai et al. 2023, Yin et al. 2024 ).Another possible explanation for the contradictions is that the factors driving bloom formation may be uncoupled from those driving toxicity, perhaps due to complex regulation pathways associated not only to the cyanobacterium, but also to its heter otr ophic partners.

Similarities between Microcystis colonies and biofilms
In biofilms, attached cells produce a hydrated matrix of extracellular pol ysacc harides (EPS), extr acellular DNA, pr oteins and lipids, changing their structure and functional relationships (Stoodley et al. 2002 ).But bacterial biofilms can also exist in the air-liquid interface forming floating biofilms or pellicles .T interface provides access to oxygen and other gasses from the air, as well as nutrients from the liquid phase through opposing gradients (Armitano et al. 2014 ).
Microcystis colonies are extremely buo y ant, commonly forming wind-blown scums .T heir position r elativ e to the surface can be ac hie v ed thanks to the presence of gas v esicles a ggr egations in the c ytoplasm (Šmar da and Maršálek 2008 ), which allo w them to regulate their vertical position in the water column and to form the colony in a suitable position to receive the right amount of light, oxygen, CO 2 and nitrogen, which is necessary to build the pr otein v esicles (Wu et al. 2023 ).The EPS matrix contributes to buo y anc y and has the same composition that has been described for pellicle-forming bacteria (Armitano et al. 2014 ), such as glucose , galactose , rhamnose , mannose or cellulose (Lei et al. 2007(Lei et al. , 2009 ) ).This matrix creates a micr oenvir onment called the phycospher e, wher e complex ecological interactions between phytoplankton and bacteria occur (Seymour et al. 2017 ).
Colony formation in Microcystis can be induced by abiotic factors causing str ess, suc h as low temper atur e (15 • C) and low light intensity (10 μmol photons m −2 s −1 ) (Yang et al. 2012, Li et al. 2013, Xu et al. 2016 ).In the presence of high concentration of calcium (Wang et al. 2011, Sato et al. 2017 ) and lead, the formation of colonies r eac hing mor e than 100 μm diameter can be induced and its EPS acts tr a pping the metal ions (Bi et al. 2013 ).As for bacterial biofilms, the ability of Microcystis to form colonies has also been linked to antibiotic resistance, since low concentrations of aminoglycoside antibiotics induced cell aggregation , suggesting a pr otectiv e r ole for the EPS (Tan et al. 2018 ).Another c har acteristic shared by biofilms and Microcystis colonies is cellular motility.Genes encoding for type IV pili (e .g. pilT ) ha ve been found in Microcystis aeruginosa PCC 7806 (Nakasugi and Neilan 2005 ), which may indicate that cells can move by means of twitching motility during the initial arrangement of the cells inside the growing colony (Maier and Wong 2015 ).As the colony grows and the biofilm starts to mature, water channels develop and a differentiation in physiological processes among cells start to establish in response to conditions in their particular en vironments .
There is growing evidence relating colony size with the amount of microcystin they produce.For example, it has been shown that colonies in the size r ange fr om 60 to150 μm diameter are those producing higher amounts of microcystins compared to single cells or smaller colonies (Gan et al. 2012, Deus Álv ar ez et al. 2020 ).On the other hand, depletion of extracellular microcystin concentrations sho w ed a decr ease in colon y size (Gan et al. 2012 ) .T hus , the evidence suggests that released microcystins could act as an infoc hemical-r elated mec hanism involv ed in the biofilm maintenance.Ho w e v er, if micr ocystins ar e involv ed in a QS-like mec hanism remains uncertain.
Regar ding QS, ac ylated homoserine lactones (AHLs) have been found in cultures of M. aeruginosa PCC-7820 (Zhai et al. 2012 ).Electr on micr oscope photogr a phs of M. aeruginosa supplemented with AHLs sho w ed a shift fr om single fr ee-living cells to a biofilm-like membrane .T his suggests that QS might play an important role in the envir onmentall y-driv en mor phological c hanges of M. aeruginosa, pr oviding str ong e vidence that it r egulates colon y formation through a coordinated multicellular behaviour as that described for biofilms .T his was confirmed more recently, when addition of se v er al AHLs fr om Gr am negativ e bacteria to cultur es of Microcystis induced colony formation (Herrera and Echeverri 2021 ).The fact that AHLs belonging to se v er al species were able to induce a response in Microcystis implies that the QS behaviour leading to colony formation could be triggered by members of the micr obiome.Mor eov er, (Shi et al. 2022 ) sho w ed that se v er al transcripts for pathways involved in biofilm formation were enriched in the Microcystis colonial form compared to single cells.These transcripts belonged mainly to heterotrophic bacteria from the microbiota, meaning that QS in Microcystis is an ability conferred by the cyanobacterium and its microbiome acting cooperativ el y.This kind of m ulti-species, m ulticellular behaviour may hav e ecosystem-le v el effects on se v er al pr ocesses , e .g. nutrient cycling, toxin biosynthesis, bloom stability, etc. (Van Le et al. 2022 ).

The Microcystis holobiont
Current vision of organism´s evolution is increasingly incorporating the concept of holobiont, which recognizes the widespread occurrence of host-associated microbiomes and makes emphasis on the multispecies nature of host-microbiome assemblage (Bordenstein and Theis 2015 ).In the case of Microcystis , the colonial organism is in fact composed of a myriad of different bacterial species interacting and exchanging common goods (nutrients , gasses , carbon, genes) inside the m ucila ginous env elope of the c y anobacterium, whic h confers it an extr emel y high ability to surviv e in differ ent envir onmental conditions (Cook et al. 2020 ).T hus , it seems sound to conclude that the colonial organism we call Microcystis is in fact a holobiont.But, how is this prokaryotic holobiont formed?
It has been reported that the highly diverse microbiome of Microcystis colonies differs markedly from that present in single cells (Wu et al. 2019 ).Co-cultivation of axenic, single-celled cultures of Microcystis with heterotrophic bacteria isolated from Microcystis colonies stimulated cyanobacterial growth and induced the production of EPS, allowing to reconstitute colony formation (Reynolds 2007, Shen et al. 2011, Wang et al. 2016 ).Mor eov er, the existence of a metabolic interdependence between Microcystis and its microbiome has been proposed (Jackrel et al. 2019, Cook et al. 2020 ), suggesting that the ability to compete with other phytoplankton groups would not be determined by the toxin production but by genes from its microbiome (Schmidt et al. 2020 ).
Ther efor e, ther e is evidence of a clear and strong relationship between the presence of an extracellular matrix and the recruitment of heter otr ophic bacteria, whic h stim ulate colonial growth through QS to form a three-dimensional structure where the exchange of common goods occurs .T his constitutes a complex holobiont organism whose formation must have involved the establishment of a symbiotic relationship early in the evolution of the c y anobacterium.As a unicellular c y anobacteria, Microc ystis can onl y ac hie v e a m ulticellular sta ge thr ough its r elationship with the symbiotic partners .T his hypothesis would also explain the r e v ersion fr om colonies to single cells observ ed when isolating Microcystis from environmental samples (Wang et al. 2015 ), probably due to the se v er al dilutions and washing steps that r emov e the attached bacteria.

Microcystis holobiont
The information gathered so far about colony formation in Microcystis spp.suggests that the mec hanisms involv ed in this process are the same as those defined for biofilm formation in a number of bacterial species.Microcystis can switc h fr om single cells to colonies organized into a coordinated functional community that is embedded in an EPS matrix teemed with a div ersity of heter otr ophic bacteria living mainl y in a cooper ativ e manner with the c y anobacterial cells (Fig. 1 ).The change from single cells to m ulticellular or ganization would be triggered by autoinducers molecules (e.g.AHLs) synthesized either by the c y anobacterium, b y the microbiome, or both, in response to environmental cues (e.g.resource-rich conditions).As the population grows, the resources become less available and the AHLs upregulate a number of functional genes allowing the organisms to thrive under conditions that would not be fa vourable , such as nutrients or light shortage, o xidati ve stress , etc. T he main components of the biofilm m ucila ge ar e EPS, DNA fr om l ysed cells, pr oteins, lipids and heter otr ophic bacteria that live embedded in this matrix.This bacterial community has a very constant structure, its functional relationships with the c y anobacterium are closely intertwined and involves the trade of different goods, allowing the holobiont to survive .T he resulting multi-specific biofilm is not built from the attachment of the cells to an abiotic or biotic surface, but on the attachment of cells to each other to form a floating biofilm that thrives in a highly diverse array of environmental conditions.

Future directions
Understanding the mechanism underlying the multispecific biofilm (colony) formation in Microcystis holobiont would help to unveil the role of the microbiome in the evolution and environmental performance of these organisms .T his will be useful to determine not only the biotic or abiotic conditions triggering microcystin production, but also to uncover the role of microcystin in the holobiont ecology and, ther efor e, in blooms de v elopment.We expect that this kind of knowledge would impr ov e curr ent (and sometimes contradictory) models of growth, fitness, dispersal and decay of these c y anobacteria, contributing to water management and risk assessment.