Эколого-генетические механизмы молекулярной эволюции клубеньковых бактерий, определяемой растениями-хозяевами тема диссертации и автореферата по ВАК РФ 00.00.00, доктор наук Андронов Евгений Евгеньевич

• Андронов Евгений Евгеньевич
• доктор наук
• 2022

Оглавление диссертации доктор наук Андронов Евгений Евгеньевич

ВВЕДЕНИЕ

Глава I ОСНОВНЫЕ ЭТАПЫ ЭВОЛЮЦИИ РИЗОБИЙ

Переход свободноживущих азотфиксаторов к симбиозу с растениями

Мобилизация симбиотических генов как эволюционный фактор

Дивергенция корового и акцессорного компонентов ризобиального генома и экологические факторы, её контролирующие

ГЛАВА 2 ГЕНОМНЫЕ МЕХАНИЗМЫ ЭВОЛЮЦИИ ПОЛИТИПИЧЕСКОГО ВИДА

R.LEGUMINOSARUM

Анализ ранних этапов эволюции протосимбионта R.leguminosarum

Анализ геномов ризобиальных изолятов, выделенных из реликтового растения вавиловии красивой

Дупликация кластера_/гхКОР0 генов у штамма Tardiphaga Бр. УаГО7, выделенного из клубенька вавиловии красивой

Детекция следов горизонтального переноса генов при сборке симбиотического кластера R.leguminosarum ьу. viciae

Анцестральные черты в организации геномов R.leguminosarum Ьу. trifoШ

ГЛАВА 3 ЭВОЛЮЦИЯ РИЗОБИЙ, КОНТРОЛИРУЕМАЯ РАСТЕНИЕМ-ХОЗЯИНОМ

Система симбиотических рецепторных генов бобовых растений

Молекулярные основы хозяйской специфичности ризобий афганских и европейских линий гороха

Эффект эволюционного «прессформинга» как следствие микро-коэволюционных процессов в сопряженных популяциях бобового растения и его ризобий

Молекулярные механизмы коэволюции бактерий и растений

Детекция отбора, действующего на ризобии в системе симбиоза

ЗАКЛЮЧЕНИЕ

ВЫВОДЫ

СПИСОК РАБОТ, ОПУЛИКОВАННЫХ ПО ТЕМЕ ДИССЕРТАЦИИ

СПИСОК ЛИТЕРАТУРЫ

БЛАГОДАРНОСТИ

Введение диссертации (часть автореферата) на тему «Эколого-генетические механизмы молекулярной эволюции клубеньковых бактерий, определяемой растениями-хозяевами»

ВВЕДЕНИЕ

Бобово-ризобиальный азотфиксирующий симбиоз на сегодняшний день не нуждается в особом представлении, так как его практическая значимость в сельском хозяйстве очевидна уже более столетия, а именно с тех пор как на рынке появляются первые препараты клубеньковых бактерий. Примерно с того же времени возникает научный интерес к ризобиям. Интересно, что исследователем, впервые выделившим их в чистую культуру из корневых клубеньков бобового растения (1888), был никто иной как Мартин Бейеринк, голландский микробиолог, автор широко известного афоризма «Everything is everywhere, (but) the environment selects»1, ставшего на долгие годы одной из ведущих исследовательских парадигм в области экологии микроорганизмов. В контексте настоящей диссертации это совпадение представляется весьма знаменательным, так как она в значительной степени посвящена именно значению экологических факторов в эволюции ризобий, к которым в случае ризобий, безусловно, относится и растение-хозяин.

Особый интерес к эволюции ризобий вызван еще тем, что специализированный бактероид, возникающий вследствие необратимой дифференциации клеток ризобий в клубеньке растения, в котором и происходит фиксация атмосферного азота, по сути,

1 Сегодня существуют различные мнения об авторстве этого афоризма, включая голландского микробиолога Лоуренса Баас-Бекинга и даже С.Н. Виноградского, однако наиболее часто цитируемым в этом контексте является именно Бейеринк (de Wit R, Bouvier T., 2006).

является временной клеточной органеллои, или, скорее, ее аналогом - «симбиосомой», которую также можно было бы назвать «нитросомой» [Ыауа! е! а1., 2012] или «аммониопластом» [ёе 1а Репа е! а1., 2018]. И если исследования органеллогенеза на примере наиболее известных органелл, таких как хлоропласты и митохондрии чрезвычайно затруднены вследствие исключительной древности этих процессов, скрытых от нас более чем миллиардом лет эволюции, то в случае эволюции бобово-ризобиаль-ного симбиоза мы имеем дело с гораздо более близким к нам событием, начавшимся лишь несколько десятков миллионов лет назад и продолжающимся в настоящее время. Таким образом, в контексте глобальной эволюции азотфиксирующий симбиоз является относительно молодой новацией, оставившей в биологической «летописи» гораздо больше следов, чем возникновение эукариотической клетки. Принимая во внимание это обстоятельство, мы можем надеяться, что раскрытие эволюционных механизмов бобово-ризобиального симбиоза, своеобразных «агро-биотехнологий прошлого», позволит на основе понимания их механизмов разработать «агробиотехнологии будущего», одним из направлений которых уже много лет является попытка переноса симбиотической азотфиксации за пределы круга бобовых растений.

Одним из исторических парадоксов эволюционного учения, на который мы обратим внимание в настоящей работе, является

то, что все базовые эволюционные научные концепции были разработаны для высших организмов, характеризующихся половым размножением, в том числе, представляющие особый интерес в настоящей работе, представления о процессах «макро-» и «микроэволюции», введенные Ю.А. Филипченко [Philiptschenko, 1927]. Собственно, парадокс заключается в том, что в рамках современной синтетической теории эволюции (СТЭ) макроэволю-ционные процессы считаются, скорее, «расширением» микроэволюционных процессов, и общим местом является признание того, что у макро- и микроэволюционных процессов механизмы в основном сходны, а различия определяются лишь временем и кумулятивным масштабом вариаций. В то же время именно сегодня, после расцвета научного скепсиса относительно существенности различий между макро- и микроэволюционными процессами, возникает их «ренессанс», убедительно продемонстрированный Евгением Куниным в недавней обстоятельной работе [Koonin, 2011]. Важным обстоятельством является то, что объектами, в эволюции которых сегодня дифференциально прослеживаются очевидные макро- и микротренды, являются не высшие организмы, обладающие половым размножением, а микроорганизмы, долгое время остававшиеся в тени главных эволюционных путей биологической науки, которые сегодня стали моделями, позволяющими проверять многие эволюционные концепции.

Причиной того, что именно в эволюции микроорганизмов, в частности, клубеньковых бактерий, которым посвящена настоящая работа, очевидны закономерности, слабо выраженные в эволюции высших организмов, вероятно, являются несопоставимо высокие (по сравнению с высшими организмами) скорости эволюции микроорганизмов, определяемые очень коротким циклом размножения, громадными размерами популяций, широчайшими вариациями экологических ниш и соответствующей им генетической и геномной пластичностью, не сдерживаемой таксономическими границами полового процесса, но в то же время, «подстегиваемой» почти неограниченным в таксономическом и функциональном репертуаре горизонтальным переносом генов (ГПГ), на фоне которого традиционный половой процесс у высших организмов, выглядит как весьма ограниченный источник генетической изменчивости. В каком-то смысле мир микробов является «стихией» эволюции, где ее многие проявления не только ярко выражены, но и доступны экспериментальному изучению [ЬешЫ, 2017].

Важным вопросом является пояснение причин, почему микроорганизмы так долго находились в тени эволюционных исследований. И это именно так: первые работы по эволюции ризобий, например, появляются примерно в 80-х годах прошлого столетия, только через сто лет после начала развития эволюционного учения [Парийская, Клевенская, 1979; Проворов,

1985; Young JP, Johnston AW, 1989]. Это связано с тем, что исследование высших организмов гораздо проще - лишь «find-n-grind», поймай и разотри/охарактеризуй, в то время как работа с культурами микроорганизмов требует на порядок более сложной процедуры, обусловленной необходимостью выделения микроорганизмов в чистую культуру. Конечно, это вполне справедливо, но, как нам представляется, есть и более глубокие причины, связанные, прежде всего, с относительной (по сравнению с высшими организмами) наблюдаемой фенотипической «бедностью» микроорганизмов (в данном случае мы ограничиваемся прокариотами и средствами их характеристики в эпоху развития эволюционного учения). Основными маркерами, на исследовании которых было построено здание эволюционной науки, были маркеры морфологические: мир наблюдаемых форм и особенностей высших организмов был фактически неисчерпаем, в то время как разнообразие прокариот долгое время помещалось в один определитель Берджи, хотя, как стало понятно теперь [Woese et al., 1990], масштабы генетического и физиологического разнообразия прокариот в реальности на несколько порядков превосходят разнообразие высших организмов, представляя собой громадное поле для эволюционных исследований, особенно с учетом того, что они имеют маленький и просто организованный геном, доступный для массированного полногеномного анализа.

Тем не менее, необходимо отметить, что один из главных феноменов разнообразия ризобий - хозяйская специфичность -был описан почти сразу после их открытия. Уже в 1927 году была сформулирована актуальная и на сегодняшний день концепция групп перекрестной инокуляции (ГПИ) [Baldwin et al., 1927; Sears & Carroll, 1927], апеллирующая, как мы сегодня понимаем, к одной из центральных проблем эволюционной биологии ризобий. Неудивительно, что одна из ранних фундаментальных работ по эволюции ризобий была посвящена именно этому феномену [Young, 1989].

Пропуская важный, но по сути своей, переходный этап характеристики разнообразия ризобий с использованием спектра физиолого-биохимических, биохимических [Martínez-De Drets and Arias, 1972], серологических [Moawad et al., 1984], связанных с анализом антибиотикоустойчивости [Dakora, 1985], а также уже почти совсем генетических подходов, таких как анализ элек-трофоретической подвижности аллельных вариантов ферментов (MLEE) [Young et al., 1987]), мы перейдем к главному обстоятельству: реальный и мощный прогресс в исследовании эволюции ризобий стал возможен только на генетическом уровне исследований, когда объектами исследований стали (в исторической последовательности) общие характеристики генома (плаз-мидный состав и различного рода геномные фингерпринты)

[Hartmann and Amarger, 1991], нуклеотидные последовательности симбиотических генов [Laguerre et al., 2001], структура сим-биотических кластеров и оперонов [Prakash et al., 1981; Horvath et al., 1987], полные геномы ризобий [Capela et al., 2001; Young et al., 2006; Schmeisser et al., 2009], и, наконец, пангеномы [Schneiker-Bekel et al., 2011; Sugawara et al., 2013] и глубоко про-секвенированные ампликонные библиотеки ризобиальных генов [Igolkina et al., 2019]. Таким образом, и это, скорее, исторический факт - реальный прогресс в области эволюции ризобий стал возможен только при экспансии современных методов генетического анализа, т.е., только в XXI веке.

Таким образом, главной загадкой остается не то, почему микроорганизмы так поздно стали объектами эволюционного учения, а то, почему вопрос о соотношении макро- и микроэволюции, который остался не решенным на основе анализа морфологических признаков высших организмов, получил новые возможности для своего решения и при использовании принципиально иных способов описания биоразнообразия у бактерий?

Далее мы дадим краткий обзор развития эволюционных представлений в биологии ризобий, который, как нам представляется, происходил в тесной связи с развитием понимания природы и особенностей генетических детерминант, контролирующих симбиотические и «домашние» функции клубеньковых бактерий.

Наиболее очевидным и традиционным направлением исследования разнообразия ризобий, в конечном счете отлившимся (и всегда отливавшихся в такого рода исследованиях) в эволюционную концепцию, стала область таксономических исследований. Первым, как уже было сказано выше, свойством ризобий была их хозяйская специфичность по отношению к одному или нескольким видам (родам) растений, что привело к формированию концепции ГПИ. В соответствии с хозяйской специфичностью ризобии получили видовое название по растению-хозяину, например, Rhizobium meliloti, Rhizobium legumi-nosarum, Rhizobium galegae. В отдельный род были выделены медленнорастущие ризобии, получившие родовое название в соответствии с этим признаком Bradyrhizobium. Революционные работы Везе [Woese et al., 1990], основанные на молекулярной таксономии по гену 16S рРНК, привели к радикальному обновлению таксономии ризобий. Стало очевидно, что старый род Rhizobium является сложной таксономической композицией внутри класса альфапротеобактерий (впоследствии последовало открытие видов ризобий за пределами этого класса). В рамках пересмотра таксономии ризобий, только на протяжении нашей научной жизни мы стали свидетелями переклассификации, например, Rhizobium meliloti ^ Sinorhizobium meliloti ^ Ensifer meliloti [Lajudie et al., 1994; Young et al., 2003]. Необходимо от-

метить, что напряженная работа филогенетиков по установлению взаимного родства ризобий и их отношений с более крупными таксонами дала в результате понимание полифилетично-сти происхождения клубеньковых бактерий, относящихся (в лице наиболее крупных и известных ризобиальных таксонов) к трем базовым семействам - Rhizobiaceae, Phyllobacteriaceae, Bradyrhizobiaceae и пониманию отношений родства таксонов более низкого ранга внутри этих семейств. Прогресс в таксономии ризобий отражал стремление исследователей приблизиться к пониманию филогенетических отношений между разными группами бактерий. Как станет понятно позднее, хозяйская специфичность отражает филогению лишь небольшой части генома, а его основная часть эволюционировала вне связи с симбиозом. Сложности исследований в этой области усугублялись тем, что одно и то же растение иногда может инокулироваться неродственными видами ризобий (например, симбионтами сои могут быть представители различных родов - Bradyrhizobium и Sinorhizobium), тогда как некоторые штаммы имеют очень широкий спектр растений-хозяев, включающий неродственные растения (широко известный штамм NGR234 [Schmeisser et 2009]).

Однако, настоящая таксономическая/филогенетическая и, соответственно, эволюционная, революция случилась после открытия генетических детерминант, определяющих основные

симбиотические характеристики ризобий (клубенькообразова-ние - nod-гены и азотфиксацию - nif и/а-гены) и широкого анализа их нуклеотидного разнообразия [Laguerre et al., 2001]. Одним из главных следствий этих исследований явилось понимание того, что филогении 16S рРНК и филогении, построенные по симбиотическим генам, неконгруэнтны [van Berkum et al., 2003]. Объяснением этого факта стало допущение горизонтального переноса симбиотических генов между различными видами/линиями ризобий [Wernegreen et al., 1999; Suominen et al., 2001; Moulin et al., 2004] и относительную независимость эволюции «корового» и «акцессорного» компонентов ризобиального пан-генома, (используя понятийный аппарат современной геномики ризобий [Young et al., 2006]). На уровне биологическом такое независимое поведение компонентов генома соответствует относительной независимости двух фаз жизненного цикла ризобий -эндосимбионта и почвенного сапрофита, каждый из которых контролируется в большой степени непересекающимися наборами генетических детерминант. Таким образом, произошло построение принципиально новой ризобиальной таксономии, но уже на основе понимания эволюционных закономерностей в этой группе микроорганизмов, где генетические детерминанты, традиционно используемые для видовой характеристики, а именно, «гены домашнего хозяйства», эволюционируют под

действием эдафических факторов, в то время как эволюция сим-биотических генов обусловлена, по большей части, растением-хозяином. Одним из примеров такой, почти идеальной таксономии, построенной на понимании этих обстоятельств, на сегодняшний день является таксономия R.leguminosarum, политипического вида, в котором симбиотические гены, обуславливающие формирование эффективного симбиоза либо с ГПИ гороха, либо с ГПИ клевера, локализованные на симбиотической плаз-миде и способные к горизонтальному переносу между хромосомными линиями отнесены к биоварам (симбиоварам) bv. viciae или bv.trifoШ, обладающими специфическими для каждого из биоваров Sym-генами, но общим пулом хромосомных линий [Laguerre е; а1., 1993, 1996].

Другим важным аспектом изучения нуклеотидного разнообразия и структурной организации симбиотических генов ризо-бий явилось понимание довольно глубокой консервативности симбиотических кластеров, состоящих фактически из одних и тех же генов, ответственных за синтез сигнальной молекулы, №ё-фактора и собственно, за фиксацию молекулярного азота [Риеррке е; а1., 1996], причем с весьма сходной организацией симбиотических оперонов, что привело во-первых, к пониманию единства эволюционного происхождения симбиотических систем ризобий [Проворов и др., 2008] (например, общие для всех

видов ризобий гены нодуляции nodABC, пришедшие из актино-бактериальных микросимбионтов Frankia [Persson et al., 2015]), а во-вторых, к некоторым предположениям, о том, что возникновение общего предка симбиотического кластера следует искать в семействе Bradyrhizobiaceae [Проворов и др., 2008].

Следующим прорывом в эволюционной биологии ризобий стала доступность полных геномов ризобий, сначала отдельных [Capela et al., 2001; Young et al., 2006; Schmeisser et al., 2009], а затем, после того, как стали доступны системы высокопроизводительного секвенирования, в особенности, «следующего поколения» (PacBio, Oxford Nanopore), реконструкция ризобиальных пангеномов [Rosselli et al., 2021]. Следует отметить особую важность технологий следующего поколения именно в изучении эволюции ризобий. Доступность полных геномов ризобий в рамках уже не популяционной генетики, а популяционной гено-мики, позволила выяснить, что переход от свободноживущих предков брадиризобий к симбиотическим формам сопровождался существенным увеличением размера индивидуальных геномов, а также и пангеномов, в особенности, их акцессорного компонента (см. Таблица 1) [Provorov, Andronov, 2016]. Это свидетельствовало о радикальном увеличении функционального потенциала симбиотических форм ризобий. Анализ геномной информации продемонстрировал широкое разнообразие структурных вариантов организации и локализации симбиотических

регионов ризобий, находящихся как на мобильных репликонах (Як11оЫит, Ет1/вг), геномных островах (МвзогМ1оЫит), так и на хромосомах (Bradyrhizobium).

Таблица 1. Геномные характеристики свободноживущих (Rhodopseudomonas) и симбиотических (Bradyrhizobium) представителей семейства Bradyrhizobiaceae

Бактерии Количества генов в геномах Доля (%) коровой части в Соотношение инди- Ссылки

Индивидуаль- Коре- Пан геноме Индивидуаль- Пангеноме видуального ге-

ном вом ном геноме нома и пангенома

Rhodopseudomonas 5408 3785 8000 77,5 47,3 0,68 Oda et al.. 2008

Bradyrhizobium (фототрофы)* 7110 4792 12040 67,4 39,8 0,59 Mornico et al., 2012

Bradyrhizobium (гетеро- трофы)** 9821 2750 > 35000 28,0 <8,0 <0,28 Tian et al., 2012

"Штаммы, образующие клубеньки на стеблях бобовых рода АезсЬупотепе без участия липо-хито-олигосахаридных КЫ-факторов (сигнальные молекулы, индуцирующие развитие клубеньков).

"Свыше 15 видов медленнорастущих ризобий, которые относятся к двум группам, представленным видами В. ]аропюит и В. е1капП (образуют клубеньки на корнях широкого круга бобовых с использованием N о с!-фа кто ров).

Кроме того, как было показано еще в ранних исследованиях, особенностью ризобиальных геномов является их исключительно высокая структурная пластичность, обусловленная большим содержанием повторов, в особенности, IS-элементов [Provorov, Vorob'ev, 2000; Lozano et al., 2010], суммарное содержание повторов, в частности, на симбиотических репликонах может составлять до 18% генома [Freiberg et al., 1997]. Эта особенность ризобиальных геномов обуславливает их склонность к инетен-сивным перестройкам, детектируемым даже в чистых культурах [Guo et al., 2003], а в эволюционной перспективе свидетельствует о важности этих процессов в формировании и дальнейших преобразованиях симбиотических регионов ризобий. Таким

образом, к началу XXI века в общих чертах стали понятны механизмы геномной эволюции ризобий, включающие горизонтальный перенос генов, внутригеномные перестройки и вариации нуклеотидного полиморфизма генов.

Однако в большинстве эволюционных процессов необходимо учитывать экологические факторы, контролирующие эволюционные процессы. Эдафический фактор, безусловно важен для понимания эволюции ризобий, однако генетические исследования показали, что при сопоставлении дивергенции природных популяций ризобий, оцениваемой в связи с географическим фактором и фактором растения-хозяина, последний имеет гораздо более глубокое влияние на эволюцию ризобий [Андронов и др., 1999]. Симбиоз - это глубоко интегрированная генетическая система партнеров [Тихонович, Проворов 2012], поэтому и эволюцию ризобий невозможно понять вне эволюции его хозяина. На ранних, филогенетических этапах, исследований было показано, что существует общая тенденция, определяемая филогенетической конгруэнтностью растения-хозяина и микросимбионта, в особенности, когда последняя оценивается с использованием симбиотических генов на популяционном уровне [Ueda et al., 1995]. Из этих представлений возникает идея ко-эволюции хозяина и микросимбионта [Provorov, Vorob'ev, 2009; Martínez-Romero, 2009], которое в некоторых облигатных вариантах эн-

досимбиоза приводит к полной конгруэнтности филогений хозяина и микросимбионта [Chen et al., 2017]. Однако для бобово-ризобиального симбиоза, в силу наличия в жизненном цикле ри-зобий сапрофитной фазы, сходства филогений имеют самый общий характер с заметными отклонениями. Одним из наиболее ярких примеров являются ризобии политипического вида R.le-guminosarum, по большей части вступающие в симбиоз с растениями из ГПИ гороха, однако, один из биоваров этого вида, вступает в симбиоз с клевером, представителем трибы Trifolieae, большинство представителей которой формирует ГПИ, вступающую в симбиотические отношения с ризобиями вида Ensifer meliloti. Понятно, что причину таких отклонений следует искать не столько в традиционных филогениях растения-хозяина, сколько в организации симбиотического генома растения-хозяина. Интенсивный генетический анализ, [Tsyganov, Tsyganova, 2020; Roy et al., 2020], а затем и доступность полных геномов модельных бобовых растений [Young et al., 2011], позволили выяснить в общих чертах структуру и функции растительного сим-биотического геномного компонента, состоящего по большей части из системы рецепторных генов и факторов их регуляции [Berrabah et al., 2018]. Основные и наиболее понятные игроки здесь - растительные рецепторные системы, участники обмена симбиотическими сигналами, непосредственно взаимодейству-

ющие с Nod-фактором ризобий, относящиеся к классу рецептор-ных протеинкиназ [Gough, Jacquet, 2013; Via et al., 2016; Kelly et al., 2017]. Именно здесь следует искать причины таких феноменов, как ГПИ, со времени открытия которых прошло уже почти 100 лет. Здесь открывается совершенно новое поле для ко-эво-люционных исследований бобово-ризобиального симбиоза, где анализируются центральные процессы, связывающие эволюцию тесно интегрированных генетических систем растения-хозяина и микросимбионта. Однако уже сегодня видно, что простых решений и на этом поле ожидать не приходится. Один из очевидных парадоксов заключается в том, что исходное предположение о том, что дивергенция по рецепторным генам растений должна быть сопоставимой с дивергенцией интегрированных генетических систем ризобий или, хотя бы, со структурной дивергенцией ризобиальных Nod-факторов, кажется не верна. В самом деле не сложно убедиться в том, что у некоторых очень близкородственных растений существенно различается хозяйская специфичность по отношению к таким же близким вариантам ризо-бий (история с Афганскими и Европейскими линиями горохов [Lie, 1978; Firmin et al., 1993], вариации в хозяйской специфичности между однолетней и многолетними люцернами [Rome et al., 1996]. В то же время другие ризобии, демонстрирующие такую высокую разборчивость по отношению к незначительным генетическим вариациям одного и того же хозяина, способны к

формированию симбиоза с целым спектром видов и родов других растений, рецепторные гены которых характеризуются десятками аминокислотных замен. Очевидно, что решать такие вопросы невозможно без подходов, позволяющих проводить анализ тонких молекулярных деталей взаимодействий сигнальных систем растения-хозяина и ризобий, например, с использованием молекулярного моделирования и докинга.

Важной особенностью ко-эволюции бобово-ризобиального симбиоза, в особенности, его сигнально-рецепторного комплекса, исследование которой с использованием молекулярного моделирования и докинга стало одной из важных частей настоящей работы, является то, что тесно интегрированные генетические системы симбионтов, несомненно имеют признаки «холо-генома» [Rosenberg, Zilber-Rosenberg, 2018], эволюционирующего как единое целое, в отличие, например, относительно независимой эволюции симбиотической и коровой части ризобиаль-ного генома.

Завершая это краткое введение, необходимо сказать и об одной чрезвычайно важной проблеме, в некотором смысле, центральной для эволюционного учения, а именно, о феномене отбора. Изучение типов отбора, действующих в эволюции бобово-ризобиального симбиоза, пока проводится лишь небольшой группой энтузиастов, среди которых научный консультант дан-

ной работы Н.А.Проворов [Provorov, Vorobyov, 2006] и некоторое количество зарубежных исследователей [Kumar et al., 2015; Cavassim et al., 2020, 2021; Epstein, Tiffin, 2021]. Хотя, пожалуй, мало найдется биологических систем, где экспериментальное исследование механизмов и факторов отбора было бы так удобно, как в бобово-ризобиальном симбиозе. Выбор микросимбионта растением по его «конкурентоспособности» [Bellabarba et al., 2021], селективное размножение в клубеньке эффективных азотфиксаторов и санкции для не фиксирующих азот, но образующих клубеньки «симбионтов-обманщиков» [Provorov, Vorobyov, 2008; Provorov et al., 2017; Pahua et al., 2018], отбор растением из ризосферы редких вариантов ризобий [Provorov, Vorobyov, 2006] и целый спектр родственных феноменов свидетельствуют о том, что почти все типы отбора - Дарвиновский негативный/позитивный, дизруптивный, движущий-стабилизирующий, частотно-зависимый и др. не только представлены в эволюции бобово-ризобиального симбиоза, но и с экспериментальной точки зрения вполне доступны для количественной оценки, в особенности при использовании таких мощных инструментов, как глубокое секвенирование ампликонных библиотек генов, сопряженное с in-vivo, in-vitro и in-silico валидацией полученных результатов и гипотез [Provorov et al., 2017].

Таким образом, вполне очевидно, что комплекс знаний, накопленных к настоящему времени, и научный инструментарий, доступный сегодня, готовы, наконец, исследовать сложнейший вопрос эволюции бобово-ризобиального симбиоза во всей его полноте, как эволюции тесно интегрированной надорганиз-менной системы, иными словами, симбиотического хологенома. В то же время, остаются мало изученными геномные механизмы и определяющие их селективные факторы эволюции даже в наиболее доступной для изучения бактериальной части объединенной генетической системы симбиоза.

В связи с этим целью настоящего исследования явилось комплексное изучение молекулярной и геномной эволюции клубеньковых бактерий (ризобий), направленное на выяснение ее экологических (селективных) механизмов и на анализ соотношения микро-эволюции, видообразования и макроэволюции, определяемых изменениями разных частей бактериального генома и происходящих под действием селективных факторов, индуцируемых растениями-хозяевами. Исследования в рамках этой цели были направлены на решение следующих задач:

Список литературы диссертационного исследования доктор наук Андронов Евгений Евгеньевич, 2022 год

СПИСОК ЛИТЕРАТУРЫ

1. Andronov EE, et al. Symbiotic and genetic diversity of Rhizobium

galegae isolates collected from the Galega orientalis gene center in the Caucasus //

Appl Environ Microbiol. 2003. V.69. №2. P.106V-V4.

2. Andronov E.E., et al. Population structure of the clover rhizobia Rhizobium leguminosarum bv. trifolii upon transition from soil into the nodular niche // Microbiology (Mikrobiologiya). 2014. Vol. 83. № 4. P. 422-429

3.Andronov E.E., et al. Characteristics of natural selection in populations of nodule bacteria (Rhizobium leguminosarum) interacting with different host plants // Russian Journal of Genetics. 2015. Vol. 51. № 10. P. 949-956.

4. Arrighi, J.-F., et al. The Medicago truncatula lysine motif-receptor-lik<

kinase gene family includes NFP, and new nodule-expressed genes [corrected title:

The Medicago truncatula lysin motif-receptor-like kinase gene family includes

NFP, and new nodule-expressed genes] // Plant Physiol. 2006. V.142. P. 265.

5. Baldwin IL, Fred EB, Hastings EG. Grouping of Legumes According to

Biological Reactions of Their Seed Proteins. Possible Explanation of Phenomenon

of Cross Inoculation // Botanical Gazette, 1927.

6. Bellabarba A, et al. Competitiveness for Nodule Colonization in Sinorhi-zobium meliloti: Combined In Vitro-Tagged Strain Competition and Genome-Wide Association Analysis // mSystems. 2021. V.6. №4. :e0055021.

7. Berrabah F, et al. The Multiple Faces of the Medicago-Sinorhizobium Symbiosis // Methods Mol Biol. 2018.V.1822. P.241-260.

8. Broughton WJ, Samrey U, Stanley J, Ecological genetics of Rhizobium

meliloti: symbiotic plasmid transfer in the Medicago sativa rhizosphere // FEMS

Microbiol. Lett. 198V. V. 40. P. 251-255.

9. Capela D, et al. Analysis of the chromosome sequence of the legume sym-

biont Sinorhizobium meliloti strain 1021 // Proc Natl Acad Sci U S A. 2001. V.98.

№1V. P.98VV-82.

10. Cavassim MIA, et al. Recombination facilitates adaptive evolution in

rhizobial soil bacteria // Mol Biol Evol. 2021:msab247.

11. Cavassim MIA, et al. Symbiosis genes show a unique pattern of intro-

gression and selection within a Rhizobium leguminosarum species complex // Mi-

crob Genom. 2020. V. 6. №4.:e000351.

12. Chen R, et al. Insect-bacteria parallel evolution in multiple-co-obligate-

aphid association: a case in Lachninae (Hemiptera: Aphididae) // Sci Rep. 2017. . V.7. № 1.: 10204.

13. Chirak E.R., et al. Search for Ancestral Features in Genomes of Rhizo-bium leguminosarum bv. viciae Strains Isolated from the Relict Legume Vavilovia formosa // Genes. 2019. V.10. P. 990.

14. Chirak E.R., et al. Structural and functional organization of the plasmid regulons of Rhizobium leguminosarum symbiotic genes // Microbiology (Mikrobi-ologiya). 2016. Vol. 85. № 6. P. 708-716.

15. Coba de la Peña T, et al. The Symbiosome: Legume and Rhizobia Co-evolution toward a Nitrogen-Fixing Organelle? // Front Plant Sci. 2018. V.8. P.2229.

16. Colombi E, et al. Comparative analysis of integrative and conjugative mobile genetic elements in the genus Mesorhizobium // Microb Genom. 2021. Oct. V.7. №10.

17. Dakora FD. Use of intrinsic antibiotic resistance for characterisation and identification of rhizobia from nodules of Vigna unguiculata (L) Walp. and Phaseolus vulgaris (L) // Acta Microbiol Pol. 1985. V.34. № 2. P.187-94.

18. de Wit R, Bouvier T. 'Everything is everywhere, but, the environment selects'; what did Baas Becking and Beijerinck really say? // Environ Microbiol. 2006, Apr. V.8. №4. P.755-8.

19. Dénarié, J. , Debellé, F. , & Promé, J.-C. Rhizobium lipo-chitooligosac-charide nodulation factors: Signaling molecules mediating recognition and morphogenesis // Annual Review of Biochemistry. 1996. V. 65. № 1. P.503-535.

20. Dobzhansky, T. Nothing in Biology Makes Sense except in the Light of Evolution // The American Biology Teacher, 1973, Vol. 35, No. 3, pp. 125-129.

21. Epstein B, Tiffin P. Comparative genomics reveals high rates of horizontal transfer and strong purifying selection on rhizobial symbiosis genes // Proc Biol Sci. 2021, Jan 13;288(1942):20201804.

22. Firmin JL, et al.. Resistance to nodulation of cv. Afghanistan peas is

overcome by nodX, which mediates an O-acetylation of the Rhizobium legumi-

nosarum lipo-oligosaccharide nodulation factor // Molecular Microbiology, 1993.

V.10. №2. P.351-360.

23. Freiberg C., et al. Molecular basis of symbiosis between Rhizobium and

legumes // Nature 1997. V.387. №22. P.394-401.

24. Gough C, Jacquet C. Nod factor perception protein carries weight in bio-

tic interactions // Trends Plant Sci. 2013, Oct.V.18. №10. P.566-74.

25. Guo X, et al. Natural genomic design in Sinorhizobium meliloti: novel genomic architectures // Genome Res. 2003, Aug. V.13. №8. P.1810-7

26. Hartmann A., Amarger N. Genotypic diversity of an indigenous Rhizo-

bium meliloti population assessed by plasmid profiles, DNA fingerprinting and in-

sertion sequence typing // Can. J. Microbiol. 1991. V.37. P.600-608.

27. Hayat S, et al. Salicylic acids: local, systemic or inter-systemic regula-

tors? // Plant Signal Behav.. 2012. Jan. V.7. №1. P.93-102.

28. Horvath B, et al. Host-specific regulation of nodulation genes in Rhizobium is mediated by a plant-signal, interacting with the nodD gene product // EMBO J. 1987, Apr. V.6. № 4. P.841-8.

29. Igolkina A.A., et al. Structural insight into the role of mutual polymorphism and conservatism in the contact zone of the NFR5-K1 heterodimer with the nod factor // Frontiers in Plant Science, 2018. V 9. P. 344.

30. Igolkina A.A., et al. Matching population diversity of rhizobial nodA and legume NFR5 genes in plant-microbe symbiosis // Ecol. Evol., 2019. Aug. 30. V.9. № 18. P.10377-10386.

31. Karasev E. S., et al.Evolution of Goat's Rue Rhizobia (Neorhizobium galegae): Analysis of Polymorphism of the Nitrogen Fixation and Nodule Formation Genes // Russian Journal of Genetics. 2019. Vol. 55. № 2. P. 263-266.

32. Kopat V.V., et al. Evolution of fixNOQP genes encoding cytochrome oxidase with high affinity to oxygen in rhizobia and related bacteria // Russian Journal of Genetics. 2017. Vol. 53. № 7. P. 766-774

33. Kelly S, Radutoiu S, Stougaard J. Legume LysM receptors mediate symbiotic and pathogenic signaling // Curr Opin Plant Biol. 2017. V.39. P.152-158.

34. Kimeklis A.K., et al. Rhizobia Isolated from the Relict Legume Vavi-lovia formosa Represent a Genetically Specific Group within Rhizobium legumi-nosarum biovar viciae // Genes. 2019. V. 10. P. 991.

35. Kimeklis A.K., et al. Divergent Evolution of Symbiotic Bacteria: Rhizobia of the Relic Legume Vavilovia formosa Form an Isolated Group within Rhizobium leguminosarum bv. viciae // Russian Journal of Genetics. 2018. Vol. 54. № 7. P. 866-870

36. Koonin E. V. Logic of Chance, The: The Nature and Origin of Biological Evolution. Upper Saddle River: FT Press, 2011.

37. Kumar N, et al. Bacterial genospecies that are not ecologically coherent: population genomics of Rhizobium leguminosarum // Open Biol. 2015. V.5. № 1. P.140133.

38. Laguerre G, et al. Typing of rhizobia by PCR DNA fingerprinting and PCR-restriction fragment length polymorphism analysis of chromosomal and symbiotic gene regions: application to Rhizobium leguminosarum and its different biovars // Appl Environ Microbiol. 1996. V.62. №6. P.2029-36.

39. Laguerre G, et al. Classification of rhizobia based on nodC and nifH gene analysis reveals a close phylogenetic relationship among Phaseolus vulgaris symbionts // Microbiology (Reading). 2001. V.147. № 4. P. 981-993.

40. Laguerre, et al. Conformity and diversity among field isolates of Rhizobium leguminosarum bv. viciae, bv. trifolii, and bv. phaseoli revealed by DNA hybridization using chromosome and plasmid probes // Canadian Journal of Microbiology. 1993. V. 39. P. 412-419.

41. Lajudie, et al.. Polyphasic Taxonomy of Rhizobia: Emendation of the Genus Sinorhizobium and Description of Sinorhizobium meliloti comb. nov., Sino-rhizobium saheli sp. nov., and Sinorhizobium teranga sp. nov. // International Journal of Systematic and Evolutionary Microbiology. 1994. V.44. №4. P715-733.

42. Lenski RE. Experimental evolution and the dynamics of adaptation and genome evolution in microbial populations // ISME J. 2017. V.11. № 10. P.2181-2194.

43. Lie TA. Symbiotic specialisation in pea plants: The requirement of specific Rhizobium strains for peas from Afghanistan // In: Annals of Applied Biology. 1978.

44. Limpens E, et al. LysM domain receptor kinases regulating rhizobial Nod factor-induced infection // Science. 2003. V.302. №5645. P.630-633.

45. Lozano L, et al. Evolutionary dynamics of insertion sequences in relation to the evolutionary histories of the chromosome and symbiotic plasmid genes of Rhizobium etli populations // Appl Environ Microbiol. 2010. V.76. №19. P.6504-13.

46. Macas J, et al. In Depth Characterization of Repetitive DNA in 23 Plant Genomes Reveals Sources of Genome Size Variation in the Legume Tribe Fabeae // PLoS One. 2015. V.10. № 11. P.1-23.

47. Madsen, E. B., et al. A receptor kinase gene of the LysM type is involved in legume perception of rhizobial signals // Nature. 2003. V.425. P.637-640.

48. Makasheva RK, et al. Perennial pea // Bull Appl Bot Genet Plant Breed. 1973. V.51. P.44-56

49. Martínez-De Drets G, Arias A. Enzymatic basis for differentiation of Rhizobium into fast- and slow-growing groups // J Bacteriol. 1972. V.109. № 1. P.467-70.

50. Martínez-Romero E. Coevolution in Rhizobium-legume symbiosis? // DNA Cell Biol. 2009. V.28. № 8. P.361-70.

51. Mayr, E. What Evolution Is // New York: Basic Books, 2001.

52. Mergaert P, Van Montagu M, Holsters M. Molecular mechanisms of Nod factor diversity // Mol Microbiol. 1997. V.25. № 5. P.811-7.

53. Mikic A, et al. Beauty will save the world, but will the world save beauty? The case of the highly endangered Vavilovia formosa (Stev.) Fed. // Planta. 2014. V.240. № 5. P.1139-46.

54. Moawad HA, Ellis WR, Schmidt EL. Rhizosphere Response as a Factor in Competition Among Three Serogroups of Indigenous Rhizobium japonicum for Nodulation of Field-Grown Soybeans // Appl Environ Microbiol. 1984. V.47. № 4 P.607-12.

55. Mornico D, et al.. Comparative genomics of aeschynomene symbionts: insights into the ecological lifestyle of nod-independent photosynthetic bradyrhizo-bia // Genes (Basel). 2011. V.3. № 1. P.35-61.

56. Moulin L, et al. Phylogenetic analyses of symbiotic nodulation genes

support vertical and lateral gene co-transfer within the Bradyrhizobium genus //

Mol Phylogenet Evol. 2004. V.30. № 3. P.720-32.

57. Muntyan A.N., et al. Linked symbiotic populations part I: Analysis of the genetic diversity of the rhizobial component // Russian Journal of Genetics: Applied Research. 2013. Vol. 3. № 2. P. 138-145

58. Nakagawa T, et al. From defense to symbiosis: limited alterations in the kinase domain of LysM receptor-like kinases are crucial for evolution of legume-Rhizobium symbiosis // Plant J. 2011. V. 65. № 2. P.169-80.

59. Nye T. Trees of trees: an approach to comparing multiple alternative phylogenies // Syst. Biol. 2008. V. 57. № 5. P. 785-794.

60. Oda Y, et al. Functional genomic analysis of three nitrogenase isozymes

in the photosynthetic bacterium Rhodopseudomonas palustris // J Bacteriol. 2005.

V.187 . № 22. P.7784-94.

61. Oulhen, et al. English translation of Heinrich Anton de Bary's 1878

speech, 'Die Erscheinung der Symbiose' ('De la symbiose') // Symbiosis. 2016.V.

69. P.131-139

62. Osterman J., et al. Galega orientalis is more diverse than Galega officinalis in Caucasus - whole-genome AFLP analysis and phylogenetics of symbiosis-related genes // Molecular Ecology. 2011. V. 20. № 22. P. 4808-4821.

63. Pahua VJ, et al. Fitness variation among host species and the paradox of ineffective rhizobia // J Evol Biol. 2018. V.31. № 4. P.599-610.

64. Persson, T., et al. Candidatus Frankia Datiscae Dg1, the Actinobacterial

Microsymbiont of Datisca glomerata, Expresses the Canonical nod Genes nodABC

in Symbiosis with Its Host Plant // PloS one, 2015. V.10. № 5. P.1-33.

65. Philiptschenko, Jurii. Variabilität und Variation // (Berlin: Gebrüder Borntraeger). 1927.

66. Porozov Y.B., et al. Conjugate symbiotic populations part II: Analysis of nfr5 receptor gene polymorphisms using molecular docking // Russian Journal of Genetics: Applied Research. 2013. Vol. 3. № 2. P. 146-151.

67. Prakash RK, Schilperoort RA, Nuti MP. Large plasmids of fast-growing rhizobia: homology studies and location of structural nitrogen fixation (nif) genes // J Bacteriol. 1981. V.145. №3. P.1129-36.

68. Provorov NA, Vorob'ev NI. Evoliutsionnaia genetika kluben'kovykh bakterii: molekuliarnye i populiatsionnye aspekty [Evolutionary genetics of rhizobia: molecular and population aspects // Genetika. 2000. V.36. № 12. P.1573-87. (in Russian).

69. Provorov NA, Vorobyov NI. Population genetics of rhizobia: construction and analysis of an "Infection and Release" model // J Theor Biol. 2000. V. 205. № 1. P.105-19.

70. Provorov NA, Vorobyov NI. Interplay of Darwinian and frequency-de-

pendent selection in the host-associated microbial populations. // Theor Popul Biol.

2006. V.70. №3. P.262-72.

71. Provorov NA, Vorobyov NI. Equilibrium between the "genuine mutual-

ists" and "symbiotic cheaters" in the bacterial population co-evolving with plants in a facultative symbiosis // Theor Popul Biol. 2008. V.74. № 4. P.345-55.

72. Provorov NA, Vorob'ev NI. Simulation of bacteria-plant coevolution in the mutualistic symbiosis // Genetika. 2009. V.45. № 5. P.581-94. (in Russian).

73. Provorov N.A., et al. Genetic structure of the introduced and local populations of Rhizobioum leguminosarum in plant-soil systems // Microbiology (Mikrobiologiya). 2012. V. 81. № 2. P. 224-232.

74. Provorov N.A., et al. Comigration of root nodule bacteria and bean plants to new habitats: Coevolution mechanisms and practical importance // Applied Biochemistry and Microbiology. 2013. V. 49. № 3. P. 209-214.

75. Provorov N. A., Andronov E. E. Evolution of root nodule bacteria: Reconstruction of the speciation processes resulting from genomic rearrangements in a symbiotic system // Microbiology (Mikrobiologiya). 2016. V. 85. № 2. P. 131139.

76. Provorov N.A., et al. Forms of natural selection controlling the genomic evolution in nodule bacteria // Russian Journal of Genetics. 2017. Vol. 53. № 4. P. 411-419

77. Provorov N.A., et al. Evolutionary Geography of Root Nodule Bacte-

ria: Speciation Directed by the Host Plants // Microbiology (Mikrobiologiya).

2020. V. 89, № 1, P. 1-12.

78. Pueppke SG. The genetic and biochemical basis for nodulation of leg-

umes by rhizobia // Crit Rev Biotechnol. 1996. V.16. № 1. P.1-51.

79. Radutoiu, S., et al. Plant recognition of symbiotic bacteria requires two

LysM receptor-like kinases // Nature. 2003. V.425. P. 585-592.

80. Rome S, et al.. Sinorhizobium medicae sp. nov., isolated from annual

Medicago spp. // Int J Syst Bacteriol. 1996. V.46. № 4. P.972-80.

81. Rosenberg E, Zilber-Rosenberg I. The hologenome concept of evolution after 10 years // Microbiome. 2018. V.6. № 1. P.78.

82. Rosselli R, et al. Pangenomics of the Symbiotic Rhizobiales. Core and Accessory Functions Across a Group Endowed with High Levels of Genomic Plasticity // Microorganisms. 2021, Feb 16;9(2):407.

83. Roy S, et al. Celebrating 20 Years of Genetic Discoveries in Legume Nodulation and Symbiotic Nitrogen Fixation // Plant Cell. 2020. V.32. №1. P.15-41.

84. Safronova V.I., et al. Genetic diversity of rhizobia isolated from nodules of the relic species Vavilovia formosa (Stev.) Fed. // Antonie van Leeuwenhoek, 2014. V. 105. № 2. P. 389-399.

85. Safronova V.I., et al. Taxonomically different co-microsymbionts of a i

relict legume, oxytropis popoviana, have complementary sets of symbiotic genes

and together increase the efficiency of plant nodulation // Molecular Plant-Microbe

Interactions, 2018.V.31. №8. P. 833-841.

86. Saldi S, et al.. Rhizobium laguerreae sp. nov. nodulates Viciafaba on

several continents // Int J Syst Evol Microbiol. 2014. V.64. Pt1. P.242-247.

87. Shatskaya N.V., et al. The plastid and mitochondrial genomes of Vavilovia formosa (Stev.) Fed. and the phylogeny of related legume genera // Vavilov Journal of Genetics and Breeding. 2019. Vol. 23. №8. P. 972-980.

88. Schmeisser, C., et al. Rhizobium sp. strain NGR234 possesses a remarkable number of secretion systems // Applied and environmental microbiology. 2009. V. 75. № 12. P. 4035-4045.

89. Schneiker-Bekel S, et al. The complete genome sequence of the domi-

nant Sinorhizobium meliloti field isolate SM11 extends the S. meliloti pan-genome

// J Biotechnol. 2011. V.155. № 1. P.20-33.

90. Sears OH, Carroll WR. Cross inoculation with cowpea and soybean nodule bacteria // Soil Science, 1927.

91. Smit, P., et al. Medicago LYK3, an entry receptor in rhizobial nodula-tion factor signaling // Plant Physiol. 2007. V. 145. №1. P.183-191.

92. Solovev Y.V., et al. Towards Understanding Afghanistan Pea Symbiotic Phenotype Through the Molecular Modeling of the Interaction Between LykX-Sym10 Receptor Heterodimer and Nod Factor // Front Plant Sci. 2021. V.12. P.1-12.

93. Suarez, J. The holobiont/hologenome as a level of selection: An approach to the evolution of multispecies systems. THEORIA // An International Journal for Theory, History and Foundations of Science. 2021. V. 36. № 1. P. 81112.

94. Sugawara M, et al. Comparative genomics of the core and accessory genomes of 48 Sinorhizobium strains comprising five genospecies // Genome Biol. 2013. V.14. № 2. R17.

95. Sulima AS, et al. New sources of Sym2A allele in the pea (Pisum sa-

tivum L.) carry the unique variant of candidate LysM-RLK gene LykX // PeerJ.

2019. V.7: e8070.

96. Sullivan JT, Ronson CW. Evolution of rhizobia by acquisition of a 500kb symbiosis island that integrates into a phe-tRNA gene // Proc Natl Acad Sci U S A. 1998. V.95. № 9. P. 5145-9.

97. Suominen L, et al. Identification and structure of the Rhizobium galegae

common nodulation genes: evidence for horizontal gene transfer // Mol Biol Evol.

2001. V.18. № 6. P.907-16.

98. Tikhonovich I.A., et al. The principle of genome complementarity in the enhancement of plant adaptive capacities // Russian Journal of Genetics. 2015. Vol. 51. № 9. P. 831-846.

99. Tsyganov VE, Tsyganova AV. Symbiotic Regulatory Genes Controlling Nodule Development in Pisum sativum L. // Plants (Basel). 2020. V.9. № 12. P.1741.

100. Ueda T, et al.. Phylogeny of Sym plasmids of rhizobia by PCR-based sequencing of a nodC segment // J Bacteriol. 1995. V.177. №2. P.468-72.

101. van Berkum P, et al. Discordant phylogenies within the rrn loci of Rhizobia // J Bacteriol. 2003. V.185. № 10. P.2988-98.

102. Via VD, Zanetti ME, Blanco F. How legumes recognize rhizobia // Plant Signal Behav. 2016. V. 11. № 2: e1120396.

103. Wernegreen JJ, Riley MA. Comparison of the evolutionary dynamics of symbiotic and housekeeping loci: a case for the genetic coherence of rhizobial lineages // Mol Biol Evol. 1999. V.16. № 1. P.98-113.

104. Woese CR, Kandler O, Wheelis ML. Towards a natural system of organisms: proposal for the domains Archaea, Bacteria, and Eucarya // Proc Natl Acad Sci U S A. 1990. V.87. № 12. P.4576-9.

105. Young JM. The genus name Ensifer Casida 1982 takes priority over Si-norhizobium Chen et al. 1988, and Sinorhizobium morelense Wang et al. 2002 is a later synonym of Ensifer adhaerens Casida 1982. Is the combination "Sinorhizobium adhaerens" (Casida 1982) Willems et al. 2003 legitimate? Request for an Opinion // Int J Syst Evol Microbiol. 2003. V.53. (Pt 6). P.2107-10.

106. Young JP, et al. The genome of Rhizobium leguminosarum has recognizable core and accessory components // Genome Biol. 2006. V. 7. №4. R34.

107. Young JP, Demetriou L, Apte RG. Rhizobium Population Genetics: Enzyme Polymorphism in Rhizobium leguminosarum from Plants and Soil in a Pea Crop // Appl Environ Microbiol. 1987. V.53. № 2. P.397-402.

108. Young JP, Johnston AW. The evolution of specificity in the legume-rhizobium symbiosis // Trends Ecol Evol. 1989. V.4. № 11. P.341-9.

109. Young ND, et al. The Medicago genome provides insight into the evolution of rhizobial symbioses // Nature. 2011.V. 480. №7378. P.520-4.

110. Young J.P.W., et al. Defining the Rhizobium leguminosarum Species Complex // Genes (Basel). 2021. V.12. №1. P.111.

111. Zhang XC, et al. Molecular evolution of lysin motif-type receptor-like kinases in plants // Plant Physiol. 2007. V.144. № 2. P.623-36.

112. Zhu H, et al. JM. Tracing nonlegume orthologs of legume genes required for nodulation and arbuscular mycorrhizal symbioses // Genetics. 2006. V.172. № 4. P.2491-2499.

113. Zhukov V, et al. The pea Sym37 receptor kinase gene controls infec-

tion-thread initiation and nodule development // Mol Plant Microbe Interact. 2008.

V.21. . № 12. P.1600-8.

114. Zverev, A.O. et al. Diversity Indices of Plant Communities and Their

Rhizosphere Microbiomes: An Attempt to Find the Connection // Microorganisms.

2021.V. 9. P. 2339.

115. Аксенова Т.С. и др. Выявление анцестральных характеристик генома у Rhizobium leguminosarum ьу. trifoШ // Сельскохоз. Биология. 2020. Т. 55. № 3. С. 489-498.

116. Карасев Е.С. и др. Сравнительный анализ филогений симбиотиче-ских генов клубеньковых бактерий с использованием метадеревьев // Сельскохозяйственная биология, 2017. Т. 52. № 5. С. 995-1003.

117. Кимеклис А.К. и др. Филогенетический анализ штаммов рода Rhi-zobium, выделенных из клубеньков Vavilovia formosa (Stev.) Fed. // Сельскохоз. Биология, 2015. № 5. С. 655-664.

118. Парийская А.Н., Клевенская И.Л. Распространение в природе и

возможные пути эволюции азотфиксирующего симбиоза // Усп. Микробиол..

1979. Т.14 С. 124- 147.

119.Проворов Н.А. Симбиотические модели для реконструкции орга-неллогенеза // Генетика, 2021. Т. 57. №1. С. 15-28.

120. Проворов Н.А. Специфичность взаимодействия клубеньковых бактерий с бобовыми растениями и эволюция бобово-ризобиального симбиоза // С.-х. биология, 1985. Т. 20. № 3. С. 34-46.

121. Проворов Н.А., Воробьев Н.И., Андронов Е.Е. Макро- и микроэволюция бактерий в системах симбиоза // Генетика, 2008. Т. 44. № 1. С. 12-28.

122. Проворов Н. А., Тихонович И. А., Воробьев Н. И. Симбиоз и сим-биогенез. / / Санкт-Петербург : Информ-Навигатор, 2018. - 463 с.

123. Тимофеев-Ресовский Н. В., Воронцов Н. Н., Яблоков А. В. Краткий очерк теории эволюции, М., 1969.

124. Тихонович И.А., Проворов Н.А. Развитие подходов симбиогене-

тики для изучения изменчивости и наследственности надвидовых систем //

Генетика, 2012. Т. 48. № 4. С. 437-450.

БЛАГОДАРНОСТИ

Автор выражает глубокую признательность всем своим коллегам и друзьям, без которых эта диссертация не могла бы появиться на свет. Прежде всего, д.б.н. Проворову Н.А., благодаря которому многолетний интерес автора к генетическому разнообразию ризобий преобразовался в интерес к эволюции симбиоза как такового, а также оказавшему большую помошь на всех этапах подготовки диссертации. Акад. Тихоновичу И.А., определившему на многие годы высочайший уровень научных работ в области симбиогенетики и исключительную творческую атмосферу во ВНИИСХМ. Проф. Симарову Б.В. и к.б.н. Румянцевой М.Л., руководителям моей кандидатской работы, посвященной изучению влияния растения-хозяина на генетическое разнообразие ризобий, прямым продолжением которой является настоящая работа. Моим зарубежным коллегам - проф. Питеру Янгу, послужившему для меня одним из главных образцов исследователя эволюции симбиоза и проф. Кристине Линдстрём, с которой мы тесно сотрудничали многие годы. Покойному проф. Круг-лову Ю.В., оказавшему мне доверие и всестороннюю поддержку, пригласив возглавить созданную им более 60 лет назад лабораторию микробиологического мониторинга и биоремедиации почв ВНИИСХМ, которая стала местом выполнения настоящей работы. Проф. Дзюбенко Н.И. и проф. Вишняковой М.А. - сотрудникам ВИР им. Вавилова Н.И., в тесном контакте с котрыми была выполнена работа по изучению растительного материала. Проф. Пиневичу А.В. явившемуся для автора ярким напоминанием о высоких стандартах

работы в области микробиологии. Моим друзьям и коллегам - д.б.н. Белимову А.А. и к.б.н. Сафроновой В.И., внесшим большой вклад в работу с вавиловией красивой и ее ризобиями, к.м.н. Порозову Ю.Б., с которым в наши исследования пришло молекулярное моделирование, д.б.н. Цыганову В.Е., д.б.н. Борисову А.Ю., проф. Архипченко И.А., д.б.н. Орловой О.В., к.б.н. Онищук О.П., к.б.н. Курчак О.Н., к.б.н. Чижевской Е.П. и всем моим коллегам по лаборатории и институту, которые активно участвовали в исследованиях, ставших основой настоящей работы. Особая признательность Пинаеву А.Г., создавшему уникальную научно-исследовательскую площадку отделения геномных технологий ЦКП ВНИИСХМ, в которой выполнены все исследования настоящей работы и к.ф.-м.н. Иголкиной А.А., продемонстрировавшей исключительно высокий профессионализм в вычислительной биологии и близкое к художственному отношение к исследуемым проблемам. Наконец, и моей семье, долгие годы терпеливо ожидавшей, не займусь ли я диссертацией.

All-Russian Research Institute for Agricultural Microbiology

As a manuscript

Andronov Evgeny Evgenyevich

Ecological and genetic mechanisms of molecular evolution of nodule bacteria defined by host plants

Scientific Specialty 1.5.11. Microbiology

Thesis for the Degree of Doctor of Biological Sciences Translation from Russian

Scientific Advisor:

Provorov Nikolay Alexandrovich Doctor of Biological Sciences

Saint Petersburg, 2021

Table of contents

INTRODUCTION.......................................................................................................................114

Chapter I THE MAIN STAGES OF THE EVOLUTION OF RHIZOBIA................................136

The transition of free-living nitrogen fixers to symbiosis with plants........................................136

Mobilization of symbiotic genes as an evolutionary factor.........................................................138

Divergence of the core and accessory components of the rhizobial genome and environmental factors controlling it....................................................................................................................141

CHAPTER 2 GENOMIC MECHANISMS OF EVOLUTION OF THE POLYTYPICAL SPECIES R. LEGUMINOSARUM..............................................................................................144

Analysis of the early stages of R. leguminosarum protosymbiont evolution..............................144

Analysis of the genomes of rhizobial isolates from the relict plant Vavilovia formosa..............145

Duplication of thefxNOPQ gene cluster in the Tardiphaga sp. Vaf07 isolated from a beautiful vavilovia nodule..........................................................................................................................150

Detection of traces of horizontal gene transfer during assembly of the symbiotic cluster in R.leguminosarum bv. viciae.........................................................................................................154

Ancestral features in the genome organization of R. leguminosarum bv. trifolii........................159

CHAPTER 3 PLANT-CONTROLLED EVOLUTION OF RHIZOBIA....................................160

The system of symbiotic receptor genes of legumes...................................................................160

Molecular basis of the host specificity of rhizobia of Afghan and European pea lines..............162

The effect of evolutionary "molding" as a consequence of micro-coevolutionary processes in conjugated populations of a legume plant and its rhizobia.........................................................166

Molecular mechanisms of coevolution of bacteria and plants....................................................175

Detection of natural selection acting on rhizobia in the symbiotic system.................................178

EPILOGUE.....................................................................................................................................185

CONCLUSION................................................................................................................................193

LIST OF WORKS PUBLISHED ON THE TOPIC OF THE DISSERTATION.......................195

REFERENCES............................................................................................................................199

ACKNOLEDGEMENTS............................................................................................................209

INTRODUCTION

Today the legume-rhizobial nitrogen-fixing symbiosis does not need a special introduction, since its practical significance in agriculture has been obvious for more than a century, namely, since the first preparations of nodule bacteria appeared on the market. From about the same time, scientific interest for rhizobia arises. Interestingly, the researcher who first isolated them into a pure culture from the root nodules of a legume plant (1888) was none other than Martin Beijerinck, a Dutch microbiologist, author of the well-known aphorism "Everything is everywhere, (but) the environment selectsM1, which for years became one of the leading research paradigms in the field of micro-bial ecology. In the context of this dissertation, this coincidence seems to be very significant, since the present work is largely devoted to the importance of ecological factors in the evolution of rhizobia, to which, in the case of rhizobia, the host plant undoubtedly belongs.

Particular interest in the evolution of rhizobia is also caused by the fact that a specialized bacteroid developing as a result of irreversible differentiation of rhizobial cells in a plant nodule, in which atmospheric nitrogen is fixed, is, in fact, a temporary cellular organelle, or, rather, its analogue, a "symbiosome", which could also be called "ni-trosome" [Hayat et al., 2012] or "ammonioplast" [de la Pena et al.,

1 Today, there are various opinions about the authorship of this aphorism, including the Dutch microbiologist Lawrence Baas-Becking and even S.N. Vinogradsky, however, it is Beijerinck that is most often cited in this context (de Wit R, Bouvier T., 2006).

2018]. And if the study of organellogenesis using the example of the most famous organelles, such as chloroplasts and mitochondria, is extremely difficult due to the deep archaism of these processes, hidden from us by more than a billion years of evolution, then in the case of the evolution of legume-rhizobial symbiosis we are dealing with an event much closer to us, as it began only a few tens of millions of years ago and continues at the present time. Thus, in the context of global evolution, nitrogen-fixing symbiosis is a relatively young innovation that has left much more traces in the biological "chronicle" than the emergence of a eukaryotic cell. Taking this into account, we can hope that the disclosure of the evolutionary mechanisms of leg-ume-rhizobial symbiosis, a kind of "agrobiotechnologies of the past", will allow, on the basis of understanding their mechanisms, to develop "agrobiotechnologies of the future", one of the directions of which for many years has been an attempt to transfer symbiotic nitrogen fixation over the limits of the family Fabaceae.

One of the historical paradoxes of evolutionary doctrine, which is our focus point in this work, is that all basic evolutionary scientific concepts were developed for higher organisms characterized by sexual reproduction, including ideas of particular interest in this work, about the processes " macro-" and "microevolutions", introduced by Yu.A. Filipchenko [Philiptschenko, 1927]. Actually, the paradox is that within the framework of the modern synthetic theory of evolution

(STE), macroevolutionary processes are considered, rather, an "extension" of microevolutionary processes, and the common approach is the recognition that the mechanisms of macro- and microevolution-ary processes are identical, while the differences are determined only by time and cumulative scale of variation. At the same time, just today, after the heyday of scientific skepticism about the significance of the differences between macro- and microevolutionary processes, their "renaissance" emerged, convincingly demonstrated by Evgeny Kunin in his recent detailed work [Koonin, 2011]. The important circumstance is that today the objects, in the evolution of which the obvious macro- and microtrends are differentially traced for higher organisms that have sexual reproduction, but microorganisms that have long remained in the shadow of the main evolutionary paths of biological science, which today have become models for the testing of many evolutionary concepts.

The reason that it is in the evolution of microorganisms, in particular, of nodule bacteria, to which this work is devoted, that there are obvious patterns that are poorly expressed in the evolution of higher organisms, probably, are the incomparably high (in contrast to higher organisms) rates of evolution of microorganisms, determined by a very short cycle reproduction, huge population sizes, the broadest variations in ecological niches and the corresponding genetic and genomic plasticity, not constrained by the taxonomic boundaries of

the sexual process, but at the same time, "spurred" by the almost unlimited in the taxonomic and functional repertoire horizontal gene transfer (HGT), against the background which the traditional sexual process in higher organisms looks like a very limited source of genetic variation. In a sense, the world of microbes is the "element" of evolution, where many of its manifestations are not only pronounced, but also available for experimental testing [Lenski, 2017].

An important question is to explain the reasons why microorganisms have been in the shadow of evolutionary research for so long. And this is exactly so: the first works on the evolution of rhizobia, for example, appear approximately in the 80s of the last century, a hundred years after the beginning of the development of the evolutionary doctrine [Pariiskaya, Klevenskaya, 1979; Provorov, 1985; Young JP, Johnston AW, 1989]. This is due to the fact that the study of higher organisms is much easier - just "fmd-n-grind", while working with cultures of microorganisms requires an order of magnitude more complex procedure, due to the need to isolate microorganisms in a pure culture with all the ensuing consequences. Of course, this is quite true, but it seems to us that there are deeper reasons involved, first of all, with the relative (in comparison with higher organisms) observed phenotypic "poverty" of microorganisms (in this case, we are limited to prokaryotes and the means of their characteristics in the era of the development of evolutionary doctrine). The main markers, on the study of which the building of evolutionary science was constructed,

were morphological markers: the world of the observed forms and features of higher organisms was virtually inexhaustible, while the variety of prokaryotes was for a long time placed in one Bergey's volume, although, as it has now become clear [Woese et al., 1990], the scale of the relative diversity of prokaryotes in reality is several orders of magnitude higher than the diversity of higher organisms, representing a huge field for evolutionary research, especially given the fact that they have a small and simply organized genome available for throughout genome-wide analysis.

Nevertheless, it should be noted that one of the main phenomena of rhizobia diversity - host specificity - was described almost immediately after their discovery. Already in 1927, the modern up to this day concept of cross-inoculation groups (CIP) was formulated [Baldwin et al., 1927; Sears & Carroll, 1927], appealing, as we understand today, to one of the central problems of the evolutionary biology of rhizobia. It is not surprising that one of the earliest fundamental works on the evolution of rhizobia was devoted to this very phenomenon [Young, 1989].

Skipping an important, but in essence, a transitional stage in the characterization of the diversity of rhizobia using a spectrum of physiological and biochemical (biochemical [Martinez-De Drets and Arias, 1972], serological [Moawad et al., 1984], associated with the analysis of antibiotic resistance [Dakora, 1985], as well as almost

completely genetic approaches, such as the analysis of the electro-phoretic mobility of allelic variants of enzymes [Young et al., 1987]), we turn to the main circumstance: real and powerful progress in the study of the evolution of rhizobia became possible only at the genetic level of research, when the objects of research were (in historical sequence) general characteristics of the genome (plasmid composition and various genomic fingerprints) [Hartmann and Amarger, 1991], nucleotide sequences of symbiotic genes [Laguerre et al., 2001], the structure of symbiotic clusters and operons [Prakash et al., 1981; Horvath et al., 1987], complete rhizobial genomes [Capela et al., 2001; Young et al., 2006; Schmeisser et al., 2009], and, finally, pan-genomes [Schneiker-Bekel et al., 2011; Sugawara et al., 2013] and deeply sequenced amplicon libraries of rhizobial genes [Igolkina et al., 2019]. Thus, and this is rather a historical fact - real progress in the evolution of rhizobia became possible only with the expansion of modern methods of genetic analysis, i.e., only in the XXI century.

Thus, the main challenge is not why microorganisms became objects of evolutionary doctrine so late, but why the question of the ratio of macro- and microevolution, which remained unresolved on the basis of the analysis of morphological characteristics of higher organisms, received new opportunities for its solution and using fundamentally different ways of describing biodiversity in bacteria?

Next, we will give a brief overview of the development of evolutionary concepts in the biology of rhizobia, which, as we see it, took

place in close connection with the development of an understanding of the nature and characteristics of genetic determinants that control the symbiotic and "house-keeping" functions of nodule bacteria.

The most obvious and traditional direction in the study of the diversity of rhizobia, ultimately molded (and always molded in this kind of research) into an evolutionary concept, has become the area of taxonomic research. As mentioned above, the first property of rhi-zobia was their host specificity in relation to one or several species (genera) of plants, which led to the formation of the concept of CIG. In accordance with the specificity of the host, rhizobia were named after the host plant, for example, Rhizobium meliloti, Rhizobium le-guminosarum, Rhizobium galegae. Slow-growing rhizobia were distinguished into a separate genus, which received a name in accordance with this trait, Bradyrhizobium. The revolutionary work of Woese [Woese et al., 1990], based on the molecular taxonomy of the 16S rRNA gene, led to the complete update of the taxonomy of rhizobia. It became obvious that the old genus Rhizobium is a complex taxonomic composition within the class of Alphaproteobacteria (later the discovery of rhizobia species outside this class followed). As part of the revision of the taxonomy of rhizobia, only during our scientific life did we witness a reclassification, for example, Rhizobium meliloti ^ Sinorhizobium meliloti ^ Ensifer meliloti [Lajudie et al., 1994; Young et al., 2003]. It should be noted that the hard work of phylo-geneticists to establish the mutual relationship of rhizobia and their

relationship with larger taxa resulted in an understanding of the pol-yphyleticity of the origin of nodule bacteria belonging (in the face of the largest and most well-known rhizobial taxa) to three basic families - Rhizobiaceae, Phyllobacteriaceae, Bradyrhizobiaceae and understanding the relationship of kinship of lower-rank taxa within these families. The progress in the taxonomy of rhizobia reflected the desire of researchers to come closer to understanding the phyloge-netic relationships between different groups of bacteria. As it will become clear later, the host specificity reflects the phylogeny of only a small part of the genome, and its main part evolved outside of the relationship with symbiosis. The difficulties of research in this area were aggravated by the fact that the same plant can sometimes be inoculated with different species of rhizobia (for example, representatives of different genera - Bradyrhizobium and Sinorhizobium - can be symbionts of soybeans), while some strains have a very wide range of host plants, including unrelated plants (the well-known strain NGR234 [Schmeisser et al., 2009]).

However, a real taxonomic / phylogenetic and, accordingly, evolutionary revolution took place after the discovery of genetic determinants that define the main symbiotic characteristics of rhizobia (nodulation - nod genes and nitrogen fixation - nif and fix genes) and extensive analysis of their nucleotide diversity [Laguerre et al., 2001]. One of the main consequences of these studies was the understanding

that the phylogeny of 16S rRNA and the phylogeny based on symbiotic genes are incongruent [van Berkum et al., 2003]. This fact was explained by the assumption of horizontal transfer of symbiotic genes between different species / lines of rhizobia [Wernegreen et al., 1999; Suominen et al. 2001; Moulin et al., 2004] and the relative independence of the evolution of the "core" and "accessory" components of the rhizobial pan-genome (using the conceptual apparatus of modern rhizobial genomics [Young et al., 2006]). At the biological level, this independent behavior of genome components corresponds to the relative independence of the two phases of the life cycle of rhizobia -endosymbiotic and soil saprophytic, each of which is controlled to a large extent by non-overlapping sets of genetic determinants. Thus, a fundamentally new rhizobial taxonomy was built, but already on the basis of understanding the evolutionary laws in this group of microorganisms, where the genetic determinants traditionally used for species characteristics, namely, "housekeeping genes", evolve under the influence of edaphic factors, while the evolution of symbiotic genes is driven largely by the host plant. Today one of the examples of such an almost ideal taxonomy, built on the understanding of these circumstances, is the taxonomy of R. leguminosarum, a polytypical species in which symbiotic genes that cause the formation of an effective symbiosis with either pea CIG or clover CIG, localized on a symbiotic plasmid and capable of horizontal transfer between chromosomal lines, are classified as biovars (symbiovars) bv. viciae or bv. trifolii,

which have Sym genes specific to each biovar, but a common pool of chromosomal markers [Laguerre et al., 1993, 1996].

Another important aspect of the study of nucleotide diversity and the structural organization of symbiotic genes of rhizobia was the understanding of the rather deep conservatism of symbiotic clusters consisting in fact of the same genes responsible for the synthesis of a signaling molecule, Nod-factor and for fixing molecular nitrogen [Pueppke et al., 1996], and with a very similar organization of symbiotic operons, which led, first, to an understanding of the common evolutionary origin of symbiotic systems in rhizobia [Provorov et al., 2008] (for example, nodABC nodulation genes common for all rhi-zobia species, which came from actinobacterial microsymbionts Frankia [Persson et al., 2015]), and secondly, to some assumptions that the emergence of a common ancestor of the symbiotic cluster should be sought in the Bradyrhizobiaceae family [Provorov et al., 2008].

The next breakthrough in the evolutionary biology of rhizobia was the availability of complete rhizobial genomes, initially singular [Capela et al., 2001; Young et al., 2006; Schmeisser et al., 2009], and then, after high-throughput sequencing systems became available, especially the "next generation" (PacBio, Oxford Nanopore), the reconstruction of rhizobial pangenomes [Rosselli et al., 2021]. It should be noted that the next generation technologies are of particular importance in studying the evolution of rhizobia. The availability of

complete rhizobia genomes within the framework of population ge-nomics, not population genetics, made it possible to find out that the transition from free-living bradirizobia to symbiotic forms was accompanied by a significant increase in the size of individual genomes, as well as pangenomes, in particular, their accessory component (see Table 1) [Provorov, Andronov, 2016].

Table 1. Genome statistics of free-living (Rhodopseudomonas) and symbiotic (Bradyrhizobium) represetatives of the family Bradyrhizobiacea

Bacteria Number genes in genomes Fraction [%) of the core part in Rate individual genome/pangenome References

Individual Core Pangenome Individual genome Pangenome

Rbodopseudo monas 5403 3785 8000 77,5 47,3 0,68 Oda et al., 2008

Bradyrhizobium (phototrophes)* 7110 4792 12040 67,4 39,S 0,59 Mornico et al., 2012

Bradyrhizobium (heterotrophes)** 9821 2750 > 35000 28,0 < 8,0 <0,28 Tian et al., 2012

^Strains nodulating stems of legumes from the genera Aeschynomerw without lipo-chitooligo saccharide Nod-factors (signal mplecules inducing nodula-

** More than 15 species of slow-growing rhizobia from two species groups of slow-growing rhizobia belonging tQ^^japonicum and B. elkanii species (nodulating the broad spectra of legumes with use of Nod-factors).

This indicated a radical increase in the functional potential of the symbiotic forms of rhizobia. Analysis of genomic information has demonstrated a wide variety of structural organization and localization of symbiotic regions of rhizobia located on mobile replicons (Rhizobium, Ensifer), genomic islands (Mesorhizobium), and on chromosomes (Bradyrhizobium). In addition, as shown in early studies, a feature of rhizobial genomes is their extremely high structural plasticity, due to the high content of repeats, especially of IS-elements [Provorov, Vorob'ev, 2000; Lozano et al., 2010], the total content of repeats, in particular, on symbiotic replicons can account for up to

18% of the genome [Freiberg et al., 1997]. This feature of rhizobial genomes determines the tendency of rhizobia to intense genomic rearrangements, detected even in pure cultures [Guo et al., 2003], and in the evolutionary perspective indicates the importance of these processes in the formation and further transformations of symbiotic regions of rhizobia. Thus, by the beginning of the XXI century, the mechanisms of genomic evolution of rhizobia, including horizontal gene transfer, intragenomic rearrangements, and variations in gene nucleotide polymorphism, have become clear in general terms.

However, in most evolutionary processes, it is necessary to consider the ecological factors that control the evolutionary processes. The edaphic factor is undoubtedly important for understanding the evolution of rhizobia, however, genetic studies have shown that when comparing the divergence of natural populations of rhizobia, assessed in connection with the geographical factor and the factor of the host plant, the latter has a much deeper influence on the evolution of rhizobia [Andronov et al., 1999]. Symbiosis is a deeply integrated system of a plant and a microorganism; therefore, the evolution of rhizobia is difficult to understand outside the evolution of its host. In early phylogenetic stages of research, it was shown that there is a general trend linking the phylogeny of the host plant with the phylogeny of the microsymbiont, especially when the latter is assessed using symbiotic genes [Ueda et al., 1995]. These ideas give rise to the concept

of co-evolution of the host and the microsymbiont [Provorov, Voro-b'ev, 2009; Martinez-Romero, 2009], which in some obligatory variants of endosymbiosis leads to complete congruence of the host and microsymbiont phylogenies [Chen et al., 2017]. However, for leg-ume-rhizobial symbiosis, due to the presence of a saprophytic phase in the life cycle of rhizobia, phylogeny similarities are of the most general nature with noticeable deviations. One of the most striking examples is the rhizobia of the species R. leguminosarum, for the most part entering into symbiosis with plants from the pea CIG (tribe Fabeae); however, one of the biovars of this species enters into symbiosis with clover, a representative of the Trifolieae tribe, most of whose representatives form the CIG, which enters into symbiotic relationship with rhizobia of the species Ensifer meliloti. It is clear that the reason for such deviations should be sought not so much in the traditional phylogenies of the host plant as in the organization of the symbiotic genome of the host plant. Intensive genetic analysis, [Tsy-ganov, Tsyganova, 2020; Roy et al., 2020], and then the availability of the complete genomes of model leguminous plants [Young et al., 2011], made it possible to elucidate in general terms the structure and functions of the plant symbiotic genomic component, which consists mainly of the system of receptor genes and their means of regulation [Berrabah et al., 2018]. The main and most understandable participants here are plant receptor systems, participants in the exchange of

symbiotic signals, directly interacting with the Nod-factor of rhizo-bia, belonging to the receptor protein kinases [Gough, Jacquet, 2013; Via et al., 2016; Kelly et al., 2017]. It is here that one should look for the reasons for such phenomena as the CIG, since the discovery of which almost 100 years have passed. Here, a completely new field opens up for co-evolutionary studies of legume-rhizobial symbiosis, where the central processes linking the evolution of closely integrated genetic systems of the host plant and the microsymbiont are analyzed. However, it is already clear today that no simple solutions can be expected in this field. One of the challenges is that the initial assumption that the divergence in plant receptor genes should be comparable with the divergence of the integrated genetic systems of rhizobia or, at least, with the structural divergence of rhizobial Nod-factors, does not seem to be correct. Indeed, it is clear that in some very closely related plants the host specificity is significantly different in relation to the same close variants of rhizobia (history with the Afghan and European lines of peas [Lie, 1978; Firmin et al., 1993], variations in host specificity between annual and perennial alfalfa [Rome et al., 1996]. At the same time, some rhizobia (e.g., NGR234 S. fredii) are capable of forming symbiosis with an enormously broad spectrum of species and genera of other plants, the receptor genes of which are characterized by dozens of amino acid substitutions. Obviously, such variation of host range can not be analysed without approaches that allow analyzing the fine molecular details of interactions between the signaling

systems of the host plant and rhizobia, for example, using molecular modeling and docking.

An important feature of the co-evolution of legume-rhizobial symbiosis, in particular, its signal-receptor complex, the study of which using molecular modeling and docking has become one of the important parts of this work, is that closely integrated genetic systems of symbionts undoubtedly have signs of "hologenome" [Rosenberg, Zilber-Rosenberg, 2018], evolving as a whole, in contrast, for example, to the relatively independent evolution of the symbiotic and accessory parts of the rhizobial genome.

Concluding this extended introduction, it is necessary to address the extremely important problem, in a sense, central to evolutionary doctrine, namely, the phenomenon of natural selection. The study of the types of selection acting in the evolution of legume-rhizobial symbiosis is still being carried out by only a small group of enthusiasts, including the scientific advisor of this work, N.A. Provorov [Pro-vorov, Vorobyov, 2006] and a number of foreign researchers [Kumar et al., 2015; Cavassim et al., 2020, 2021; Epstein, Tiffin, 2021]. Although, perhaps, there are few biological systems where the experimental study of the mechanisms and factors of selection would be as convenient as in the legume-rhizobial symbiosis. The choice of a mi-crosymbiont by a plant based on its "competitiveness" [Bellabarba et al., 2021], selective reproduction in the nodule of effective nitrogen fixers and sanctions for non-fixing nitrogen, but forming nodules of

"cheater symbionts" [Provorov, Vorobyov, 2008; Provorov et al., 2017; Pahua et al., 2018], the selection of rare genotypes of rhizobia by a plant from the rhizosphere [Provorov, Vorobyov, 2006] and a broad range of related phenomena indicate that almost all types of selection - Darwinian negative / positive, disruptive, driving-stabilizing, frequency- dependent and others are not only represented in the evolution of legume-rhizobial symbiosis, but also from an experimental point of view are quite accessible for quantitative assessment, especially when using such powerful tools as deep sequencing of am-plicon gene libraries, coupled with in-vivo, in-vitro and in-silico validation of the results and hypotheses [Provorov et al., 2017].

Thus, today it is quite obvious that the sum of knowledge accumulated to date and the scientific tools available today are ready, at last, to investigate the most complicated issue of the evolution of the legume-rhizobial symbiosis in its entirety, as the evolution of a closely integrated supraorganism system, in other words, symbiotic hologenome. At the same time, the genomic mechanisms and selective factors of evolution, even the most accessible for studying the bacterial part of the integral genetic system of symbiosis, remain poorly studied.

In this regard, the objective of this study was a comprehensive research of the evolution of nodule bacteria (rhizobia), aimed at elucidating its ecological and genetic mechanisms and at analyzing the trade-off between micro-evolution, speciation and macroevolution,

determined by changes in different parts of the bacterial genome and occurring under the impacts of selective factors, induced by host plants. Research within the framework of this goal was aimed at solving the following tasks:

1. Characterization of the major stages of the evolution of rhizobia associated with their emergence from non-symbiotic nitrogen-fixing forms by genomic rearrangements, as well as with the polyphyletic emergence of the order Rhizobiales by transfer of sym-genes into various soil-born bacteria;

2. Reconstruction of the evolutionary history of the Rhizobium le-guminosarum species associated with the formation of symbiot-ically specialized biovars that enter into symbiosis with plants of the Fabeae tribe (bv. viciae) and the Trifolium genus from the Trifolieae tribe (bv. trifolii);

3. Comparative analysis of the processes of microevolution in rhizobia, determined by the divergence of the symbiotically specialized accessory part of the genome, and speciation, determined by the divergence of housekeeping genes included in the core part of the genome;

4. Analysis of the co-evolution of rhizobia and host plants at the population level (model of evolutionary molding based on the results of deep sequencing of symbiotic genes of bacteria and

plants) and at the molecular level (modeling of symbiotic interactions based on docking of signal and receptor molecules); 5. Identification and characterization of various types of natural selection (positive / negative, disruptive, frequency-dependent, driving / stabilizing) induced by host plants in rhizobia populations.

The novelty of the work:

It is shown for the first time that:

1. The leading role of the host in the evolution of rhizobia is determined by the induction of various forms of selection in their populations;

2. One of the results of this selection is evolutionary moding, defined as bringing the topologies of bacterial genes for the synthesis of signaling Nod-factors in line with the topologies of genes of plant receptors that recognize these signals;

3. Microevolution and speciation in rhizobia are mostly independent processes determined

the action of different environmental factors and genomic mechanisms causing divergence of accessory and core genes, respectively;

4. The divergent evolution of the species Rhizobium leguminosarum is determined, like the macroevolution of rhizobia, by genomic mechanisms designated as "gene gain and loss" and "compaction of sym-gene clusters";

5. New experimental and analytical methods have been developed for the analysis of coevolutionary processes in the symbiotic system, operating with the topology of phylogeny of partners, as well as methods for the analysis of genetic polymorphism of receptor genes in the context of the interaction of signaling systems of partners using molecular modeling and docking (projection of population polymorphism onto the contact surface of receptor dimers simultaneously with the docking of the Nod-factor).

The practical value of the work is determined by the fact that the models of the evolution of legume-rhizobial symbiosis that we have constructed can be used to develop programs and algorithms for the design of supraorganismal systems for agricultural, biomedical and environment-protecting purposes.

The following statements are submitted to the defense:

1. The evolution of rhizobia is divided into two stages, the first of which is associated with the assembly of a system of sym-genes based on genomic rearrangements in free-living nitrogen fixers, which gave rise to "primary" rhizobia, while the second, with

the distribution of these genes among a wide range of soil-born bacteria, which gave rise to "secondary" rhizobia;

2. The main ecological factor in the evolution of rhizobia is leguminous host plants, which induce in bacterial populations various forms of selection (positive / negative, disruptive, frequency-dependent, driving / stabilizing), which determines the evolution of symbiotic traits;

3. The coevolution of rhizobia and leguminous host plants, carried out during the interaction of their populations and described using the model of "evolutionary molding", at the molecular level manifests itself in the form of parallel variation in the primary and spatial structure of signaling Nod-factors and the tertiary structure of receptor proteins;

4. A multilevel classification of evolutionary processes in rhizobia has been proposed, according to which microevolution is determined by changes in the accessory, symbiotically specialized part of the genome, speciation - by changes in the core part of the genome, macroevolution - by horizontal transfer of sym-genes from rhizobia to diverse bacteria.

The results of the dissertation were presented at more than 20 conferences, including congresses of the Vavilov Society of Geneticists and Breeders (2014, 2019), the anniversary conference "50 years of VOGiS: successes and prospects (2015), at specialized workshops

on the genomics of nitrogen fixers at the European Conference on Nitrogen Fixation (2016, 2021), conferences (PLAMIC2018, 2020), Moscow Conference on Computational Biology (MCCMB 2017), Russian Microbiological Congresses (2017, 2019, 2021) and a number of others. The dissertation was presented in full for the first time in a plenary report at the 3rd Russian Microbiological Congress (2021). The results of the dissertation were published in 27 articles. This work was supported by grants from the Russian Science Foundation 14-26-00094, 14-26-00094P and 19-16-00081. All experimental studies were carried out in the laboratory of microbiological monitoring and bioremediation of soils of ARRIAM, as well as in the collective use center "Genomic technologies, proteomics and cell biology" of ARRIAM.

The personal contribution of the author was he supervised almost all published experimental works used in the dissertation research carried out in the laboratory of microbiological monitoring and bioremediation of soils of ARRIAM, over the past 10 years, during which the author has managed this laboratory. The leadership consisted in defining the research topic, project design, choosing experimental technologies and methods for data analysis, participation in the experimental procedures, processing and interpreting data, editing manuscripts, negotiations with editors and reviewers, etc.

The dissertation is presented on 107 pages of text, contains an Introduction, 3 chapters and a conclusion, 20 illustrations, 4 tables, 124 references to literary sources.

For a better orientation in the system of literary references, references to the literature data used are given in regular font, and references to works with the participation of the author, in which the data that are the subject of the dissertation research are published, are given in italics.

Chapter I THE MAIN STAGES OF THE EVOLUTION OF RHIZOBIA The transition of free-living nitrogen fixers to symbiosis with plants

Despite the fact that the details of the evolutionary process that led soil nitrogen fixers to symbiosis with plants, and, as a consequence, to a radical expansion of their adaptive potential [Tikhono-vich et al., 2015] are hidden, apparently, tens of millions of years of evolution, nevertheless the elements of the evolutionary chain of events that have survived to our time, which have preserved, if not the real participants in the process, then at least their close relatives and the logic of their evolution, allow us to put forward a number of plausible hypotheses about how exactly nitrogen fixers switch to a symbiotic lifestyle [Provorov , Andronov, 2016]. These are the following forms: free-living phototrophic nitrogen-fixing agents, close to the genus Rhodopseudomonas (Rhizobiales, Bradyrhizobiaceae), very close to slow-growing rhizobia (Bradyrhizobium) [Oda et al., 2005], phototrophic strains of Bradyrhizobium devoid of Nod-factors (without the formation of intracellular bacteroids) nodulating the stems of the tropical plant Aeschynomene [Mornico et al., 2011], and, finally, modern specialized Bradyrhizobium species possessing a full set of symbiotic factors, including signaling systems and symbiotic nitrogen fixation (in differentiated bacteroids), while completely lost the ability to phototrophy, but recruited part of the genetic systems of photosynthesis to provide electron transport during nitrogen fixation

fx-genes) (Figure 1). The origin of rhizobial genes that control signaling becomes clear from the study of the complete genomes of ac-tinobacterial nitrogen-fixing microsymbionts Frankia [Persson et al., 2015], which belong to the most archaic systems of symbiotic nitrogen fixation, but already possess a set of common symbiotic genes nodA, nodB, nodC, encoding for the parts of the Nod-factor, indicating the actinobacterial origin of these genes.

Figure 1. Evolutionary scheme of the emergence of symbiotic rhizobia from free-living photosynthetic nitrogen fixers with subsequent compaction and mobilization of the symbiotic cluster and its further transfer to a wide range of soil microorganisms [Provorov, Andronov, 2016].

It is important to note that few of the symbiotic Bradyrhizobium species still retain the ability to fix nitrogen ex planta today. In general terms, our proposed scheme of the first phase of the evolution of symbiotic nitrogen fixers is shown in Figure 1. It resulted in the formation of a symbiotic cluster (having chromosomal localization in "primary"

rhizobia), consisting of nitrogen fixation genes nif (proteins and ni-trogenase cofactors) and fix (an electron transport system, borrowed from the phototrophic system (genes ccJNOPQ) and nod genes that provide the synthesis of the Nod-factor. During this evolution, rhizobia lost the ability to phototrophy and nitrogen fixation ex planta. At this stage, the prerequisites arose for the transition to the second phase of evolution - the transfer of the symbiotic gene cassette to diverse soil microorganisms, followed by adaptive radiation in a wide range of leguminous plants.

Mobilization of symbiotic genes as an evolutionary factor

In the course of evolution, rhizobia adapted to a wide range of host plants (Figure 1). Many details of this process are hidden from us by tens of millions of years of evolution, however, post factum we see that for this development, the cluster of symbiotic genes had to change the genomic environment. By means of horizontal transfer of the symbiotic cluster, in accordance with our proposed evolutionary scheme [Provorov et al., 2017], "secondary" rhizobia have arisen (for example, the transfer of the sym-cluster into soil bacteria close to Xanthobacter or Phyllobacterium, which gave rise to the rhizobial genera Azorhizobium and Mesorhizobium) fixing nitrogen in symbiosis with sesbania, astrogalus, acorn, chickpea, lily, etc., and in soil bacteria close to Agrobacterium, which gave rise to the rhizobial genera Rhizobium, Sinorhizobium, Neorhizobium), fixing nitrogen in

symbiosis with the Galegoid complex of legume plants. Thus, it is the horizontal gene transfer that was one of the most important mechanisms of the evolution of rhizobia, and the increase of sym gene "mobility" is an important component of the evolutionary process [Pro-vorov, Andronov, 2016; Provorov et al., 2020].

It is also important to note that there are at least two aspects in the evolution of mobility - "transport" itself, which ensures the physical transfer of genetic material, and "compactness", which minimizes the size of the symbiotic cluster, and, thus, ensures the maximum efficiency of horizontal transfer. The first aspect includes natural "evolutionary experiments" with different types of mobility (mobile genomic islands, transmissible plasmids), the second - the processes of "compaction", including the formation of operons from scattered genetic elements, as well as the exclusion from the sym-cluster of genes that are not essential for symbiosis and are functionally "redundant" genes [Chirak et al., 2016; Chirak et al., 2019]. The result of evolutionary compaction and mobilization of the symbiotic cluster is obvious during the transition from primary rhizobia, in which symbiotic genes can be localized in several unlinked chromosomal clusters, to secondary rhizobia of the second level with a cluster size of several tens of kb, localized on a small, often transmissible plasmid. As will be shown below, all these processes took place in the evolution of pea rhizobia, partially reconstructed by us when analyzing the genomes of symbiont rhizobia of the relict plant Vavilovia formosa.

Analysis of the diverse genomic "architectures" of primary and secondary rhizobia that exist today in nature allowed us to assume that secondary rhizobia are a "masterpiece" of the evolution of rhizo-bia, in which a functionally effective genome structure has been found that allows physically separating the "core" and "accessory" components of the genome, while maximizing the mobility of the latter. This allows, firstly, to make the evolution of the core and accessory components of the genome relatively independent, using all the evolutionary advantages of the rhizobial analogue of the sexual process, which was fully developed in the evolution of eukaryotes. The latest studies in the field of population genomics [Cavassim et al., 2021] demonstrate that in natural populations of secondary rhizobia, it is in the horizontally transferred segments of the rhizobial genome that signs of selection are detected.

From the point of view of the classification of evolutionary processes, it is obvious that the transfer of such a large and functionally rich symbiotic cluster to a new genetic environment undoubtedly leads to the emergence of a new "supraspecific" taxon, since this is not only about obtaining an accessory function that allows occupying a completely new ecological niche, but also to the rearrangement of the entire chromosomal environment to ensure effective (although at first functionally inactive) expression of symbiotic genes. Thus, we are dealing with classical and quite obvious macroevolutionary pro-

cesses. Traces of such processes can be found in many rhizobia, including the species R. leguminosarum, the most probable scenario of origin of which is due to the transfer of a symbiotic plasmid into a free-living species of soil bacteria close to modern Agrobacterium [Provorov et al., 2020], as well as microsymbionts of beans, represented in the Latin American center of origin of beans by the species R. etli and R. tropici, bearing very similar symbiotic clusters, after the introduction to Europe, they were transferred to the local species of soil bacteria, which gave rise to the species R. gallicum, R. giardinii and R. leguminosarum bv. phaseoli [Provorov et al., 2020]. At present, similar events are also detected, for example, the transfer of a symbiotic genomic island from rhizobia Mesorhizobium sp., introduced on the crops of the Lotus in New Zeeland, into local soil bacteria, moreover, a completely new population of rhizobia emerged over few years [Sullivan et al., 1998; Colombi et al., 2021].

Divergence of the core and accessory components of the rhizobial genome and environmental factors controlling it

Rhizobia represent a very interesting object of evolutionary research, since they combine several fundamentally important features. Their life cycle consists of two phases - soil and symbiotic, each is characterized by fundamentally different ecological characteristics, and the genetic determinants that control the life of a microorganism in each phase are, as a rule, located on different replicons (chromosome and symbiotic plasmid). In addition, historically, species diagnostics in

microorganisms has long been based on the divergence of housekeeping genes, i.e. in this case, chromosomal or, more generally, the core and accessory components of the genome. Thus, in the evolution of rhizobia, two directions of divergence can be distinguished: the divergence of the core component controlled by edaphic factors and causing speciation in rhizobia and the divergence of the symbiotic component controlled by the host plant, corresponding to classical microevolutionary processes.

In the variety of rhizobia, we find a lot of evidence of such processes. For example, diversification of housekeeping genes led to speciation in the genus Sinorhizobium, where two sister species S. meli-loti and S. medicae demonstrate a very high level of divergence in chromosomal genes, while demonstrating minor differences in host specificity. What is especially interesting is that both of these "good" species inhabit the same soils and can often be isolated from the nodules of neighboring plants of the same hosts [Muntian et al., 2012; Porozov et al., 2012]. A similar situation is shown for R. legumi-nosarum soil populations of which is represented by a rather large set of "genomic species", among which symbiotic plasmids of different types circulate freely. This spectrum of genospecies is stably maintained not only in the same soil, but, most likely, in many geographically separated regions [Kumar et al., 2015; Young et al., 2021]. Despite a very deep divergence of genospecies in housekeeping genes, only a few of them have received the status of species to date, for

example, R. laguerreae [Sai'di et al., 2014], the reason for which was its predominant isolation from lentil nodules.

We implemented a number of studies that showed that in R. le-guminosarum symbiotically different biovars entering into symbiosis with the CIG of peas and clover, the divergence in symbiotic genes is significantly higher than in housekeeping genes. At the same time, within biovars, the divergence of housekeeping genes is more pronounced than the divergence of symbiotic genes, which indicates that speciation occurs independently of symbiotic diversification [Kimeklis et al., 2019]. We also showed that the diversification of housekeeping genes, which is independent of symbiosis, can be the cause of the appearance of new species in different groups of rhizo-bia. Namely, in Galega rhizobia N. galegae biovars orientalis and officinalis, in which the diversification of housekeeping genes is much more pronounced than for symbiotic genes [Karasev et al., 2019]. Similar patterns were revealed in the Rhizobium leguminosarum populations formed in the boreal zone (northern Karelia), which is characterized by poor soils and a shortage of host plants available for inoculation [Provorov et al., 2012].

The data obtained indicate that the diversification of housekeeping genes and symbiotic genes represent two parallel paths of divergent evolution, the first of which leads to speciation, and the second to intraspecific diversification. The latter occurs under the influence of disruptive selection controlled by the host plant (experimental data

on which will be presented below). In conclusion, it should be noted that it is not clear today which of the edaphic factors control specia-tion in rhizobia, since the soil is a complicated system characterized by many physical, physicochemical and biochemical factors, the differentiation of the influence of which is an extremely difficult task. We also cannot exclude the possibility that speciation in rhizobia can be a largely stochastic process, as a result of which the host plant picks up randomly occurring genotypes from the rare rhizobial biosphere (for example, through frequency-dependent selection, see below), which are then supported due to limitations (for unknown reason) of recombination between genospecies. It is important that the latter assumption can be verified experimentally - and this has become one of the important directions of our further research.

CHAPTER 2 GENOMIC MECHANISMS OF EVOLUTION OF THE POLYTYPI-CAL SPECIES R. LEGUMINOSARUM

Analysis of the early stages of R. leguminosarum protosymbiont evolution

The evolutionary events that led to the emergence of rhizobia, entering into symbiosis with representatives of the Fabeae tribe, are hidden from us by 10-12 million years of evolution. Probably, we will not sin against evolutionary realities if we proceed from the assumption that the beginning of symbiotic interactions in the Fabeae tribe should be attributed to the moment of its appearance. Where can we

look for details of evolutionary events associated with the formation of the symbiotic system at that time?

Analysis of the genomes of rhizobial isolates from the relict plant Vavilovia formosa

The object, the study of which could shed light on the details of evolutionary events at that time, is one of the few legumes that have the status of a "living fossil" - the relict species Vavilovia formosa, a rare plant found in the Caucasus, Turkey and Iran over a rather large, but ruptured area, extremely difficult to study, both because of the inaccessibility and rarity (Figure 2), and because it is one of the few "uncultivated" plants, flowering and fruiting (with very rare exceptions) only in natural conditions.

Figure 2. Vavilovia formosa in natural habitats in the North Caucasus (photo by Ivanov A.L.)

It is believed that beautiful vavilovia is the closest living relative - the last common ancestor of the tribe Fabeae [Mikic et al., 2014], although according to molecular phylogeny vavilovia is a sister species with peas, which, incidentally, was reflected in the early taxonomy, in which this plant was attributed by its discoverer Govorov L.I. [Makasheva et al., 1973] to the genus of peas (Pisum formosum). The resolution of the issue of the evolutionary status of vavilovia requires carrying out genomic studies, in part already begun by us [Shatskaya et al., 2019]. An analysis of preliminary data on genomic evolution in the Fabeae tribe clearly indicates that one of the most important mechanisms of diversification in this group of plants is not so much variation in the gene content, but rather "architectural" transformations of nuclear chromosomes [Macas et al., 2015]. Therefore, vavilovia could well preserve not so much ancestral alleles as ancestral architecture of the genome. We dwelt on this question for the following reason: as will be shown below, rhizobia isolated from vavilovia nodules have a number of ancestral features in the organization of the symbiotic genome; therefore, it is quite logical to conclude that ancestral forms of rhizobia are supported precisely in ancestral lines of legumes. The ongoing work on the assembly and analysis of the vavilovia genome may provide a more detailed answer to the question of which particular features of the vavilovia genome may have an ancestral nature (genome architecture, functional repertoire). However, even if vavilovia turns out to be a genetically simply diverged line of

peas, the mechanism ensuring the preservation of ancestral genotypes of rhizobia in the population of vavilovia may become its relict status and ecological isolation from other plants (it grows in very rare and isolated groups in the remote mountainous regions of the Caucasus on talus due to the very deep root system (Figure 2)).

In a series of our works devoted to the analysis of rhizobia isolated from three independent Caucasian populations of vavilovia [Safronova et al, 2014; Kimeklis et al., 2015, 2018; Kopat et al., 2017; Kimeklis et al., 2019, Chirak et al., 2019], we have shown that the nodules of vavilovia contain a number of microorganisms belonging to the genera Bosea, Tardiphaga, from the ancestral family of rhizobia Bradyrhizobiaceae, as well as from the unrelated to Rhizo-bium genus Phyllobacterium [Safronova et al., 2014]. However, the main symbiont of vavilovia, which has a full set of symbiotic genes, belongs to the species R. leguminosarum and is extremely close to the biovar viciae, although it forms a cluster isolated by symbiotic (but not chromosomal) genes within it (Figure 3).

To search for ancestral features in the genomes of rhizobia of vavilovia, we sequenced the genomes of three strains of R. leguminosarum bv. viciae [Chirac et al., 2016; Chirak et al., 2019] and performed their comparative analysis with the genomes of other representatives of this biovar. Figure 4 shows the main features of the genomes of vavilovia microsymbionts, which, presumably, reflect their evolutionary antiquity.

Figure 3. Phylogeny of a collection of beautiful vavilovia isolates and reference strains of the species R. leguminosarum bv. viciae and bv. trifolii based on concatenated housekeeping genes (left) and symbiotic genes (right) [Kimeklis et al., 2019].

Among the genomic features of vavilovia microsymbionts, presumably having an ancestral character, that is, marking the stage of evolution when a "protosymbiont" was formed in the common ancestor of the Fabeae tribe — the common ancestor of modern strains of R. leguminosarum bv. viciae:

Figure 4. Comparative analysis of the organization of the symbiotic cluster of beautiful vavilovia isolates and the reference genomes of R. legumi-nosarum bv. viciae. The asterisk denotes the TOM strain isolated from the Afghan pea nodules [Chirak et al., 2019]. 3841, 248, WSM1481 - reference genomes of pea ani vicia rhizobia.

1) Extended intergenic regions in the symbiotic cluster, exceeding that of other strains of this species by more than 2 times;

2) Functional redundancy: the obligatory presence of the nodX and fixW genes in the strains isolated from vavilovia;

3) The isolated position of the nodT gene, in "advanced" strains, migrated into a symbiotic cluster, sometimes on a separate replicon;

4) The absence of a chromosomal copy of the fxNOPQ-gene cluster, apparently, also obtained by "advanced" strains in the later evolutionary history

Analysis of the revealed features made it possible to separate the analyzed strains of R. leguminosarum bv. viciae into two groups: "ancestral" and "derived". The first group includes strains isolated from vavilovia, the second - from vetch and peas. Interestingly, the well-known TOM strain, isolated from the Afghan pea line, known primarily for its ability to form nodules on the Afghan pea lines characterized by the Sym2A receptor genotype, is in a sense an "intermediate" evolutionary version, albeit closer to the ancestral group but differs from it in the presence of two additional copies of the /oNOPQ genes, in addition to the copy in the symbiotic cluster.

Duplication of the fixNOPQgene cluster in the Tardiphaga sp. Vaf07 isolated from a beautiful vavilovia nodule

As mentioned above, we attributed the absence of a chromosomal copy of /aNOPQ genes encoding for terminal cytochrome oxidase, which provides respiration and supply of nitrogenase with electrons under microaerophilic conditions of the nodule, to the specific features of the ancestral types of R. leguminosarum bv. viciae. In part, this is dictated by the very fact of identifying these variants in vavilovia, which is close to the common ancestor of the tribe Fabeae, on the other hand, by the absence of any homologues of this group of genes in modern strains of the genus Agrobacterium - the putative evolutionary predecessors of R. leguminosarum, to the genome of

those, apparently, a symbiotic plasmid has been transferred. However, a detailed study of the genomes of the "accompanying" micro-symbionts identified in the nodules of vavilovia made it possible to obtain one more piece of evidence in favor of this hypothesis. When sequencing the Tardiphaga sp. Vaf07, isolated from vavilovia nodules along with its "true" symbionts, revealed two copies of the fixNOPQ genes, separated by a genome region of about 200 kb [Kopat et al., 2017]. In general, the presence of this genetic system in a strain of the genus Tardiphaga is not surprising, since this genus is phylogenetically very close to the genus Rhodopseudomonas, the most likely common ancestor of "primary" rhizobia, whose fxNOPQ genes were involved in the genetic apparatus encoding the photosynthesis system, part of which was later recruited into the genetic network responsible for symbiotic nitrogen fixation apparatus [Provorov et al., 2017]. However, it is quite remarkable that two copies of this operon identified in the Tardiphaga sp. Vaf07 underwent a rather deep divergence, the differences in nucleotide sequences between them exceed 30%. Phylogenetic analysis, including both these copies and the fxNOPQ operons from other rhizobia and free-living soil bacteria (Figure 5), revealed an unexpected circumstance: while one of the copies of this operon, as expected, falls into the same cluster with the closest relatives of Tardiphaga (Bradyrhizobium, Rhodopseudomonas, etc.), then the second falls into the same cluster with chromosomal copies of the rhizobia of alfalfa and pea.

Figure 5. Structural organization of the genome region of Tardiphaga sp. Vaf07 containing two diverged copies of the fixNOPQ operon and phylogeny for fix-concatenates and 16S rRNA gene of Tardiphaga sp. Vaf07 and a group of reference strains [Kopat et al., 2017].

Obviously, the most plausible explanation for this fact would be that we have evidence of one of the well-documented mechanisms of evolution, in which genes that control new functions appear as a result of gene duplications followed by neofunctionalization. Together with the data on the analysis of the symbiotic cluster of R. leguminosarum bv. viciae, demonstrating first the acquisition and then the loss of the nodX and fixW genes in the course of evolution, these data illustrate one of the evolutionary strategies called "gain-and-loss" that manifested itself during evolutionary assembly and subsequent structural and functional compaction of the symbiotic cluster. From a physio-

logical point of view, the presence or absence of an additional chromosomal copy of the fxNOPQ genes may be associated with different aeration of the roots of vavilovia and peas growing on various substrates: the first one on scree which is well aerated, the second one on conventional soils with more limited oxygen access.

Another important aspect of observing the duplication of fix genes in the Thardiphaga strain isolated from the same nodule with the main symbionts of vabilovia is that, in addition to the possible separation of symbiotic genes into different replicons, functional genes in the protosymbiont can be divided according to different types of microsymbionts and represent a kind of symbiotic "consortium". A whole set of data confirming this assumption, obtained by Vera Safronova when studying rhizobial consortia isolated from nodules of relict legume plants [Safronova et al., 2018], confirms the assumption that at early evolutionary stages a range of symbiotic genes can be located and function in concert in the genomes of various rhi-zobia collectively inhabiting the nodules of leguminous plants. Obviously, at the later stages of evolution, symbiotic genes presented on different replicons and even in different genomes are assembled into a single cluster in one genome. The mechanism of this assembly is also obvious - it is horizontal gene transfer which, as shown by model experiments [Broughton et al., 1987], can occur in nodules. Traces of these processes in the analyzed set of genomes can also be detected by their comparative bioinformatics analysis.

Detection of traces of horizontal gene transfer during assembly of the symbiotic cluster in R.leguminosarum bv. viciae

Earlier for the first time we used the method of constructing a "tree of trees" to analyze the contribution of horizontal gene transfer to the formation of a symbiotic cluster of rhizobia [Nye, 2008] or "meta trees" where for the same group of genomes a matrix of pair-wise distances is constructed on the basis of pairwise comparison of topologies of phylogenetic trees of the same group of genes [Karasev et al., 2017]. This is an advanced phylogenetic approach that makes it possible to identify groups of genes with a similar topology of phy-logeny indicating a similar history of horizontal transfers for this group. We used this approach to analyze a group of genomes including strains of R. leguminosarum bv. viciae representing ancestral and derived groups [Chirak et al., 2019], shown in Figure 6.

Figure 6. Metadree built of a group of symbiotic genes from isolates of genomes of R. leguminosarum bv. viciae isolated from beautiful vavilovia and a group of reference genomes of the same species [Chirak et al., 2019].

On the meta-tree, shown in the figure, we can see quite obvious clustering of genes associated first of all with their symbiotic functions. The group of genes responsible for the synthesis of the Nod-factor forms a separate cluster, the second one contains mainly fix genes, and the third one - the fix and nif genes adjacent in the symbiotic cluster. It is important that in this tree we can find a number of exceptions, when functionally completely dissimilar genes form the same cluster on the meta-tree. Nevertheless, the general trend of clus-

tering is quite clear and is in the same vein with the currently understandable ways of assembling a symbiotic cluster - the independent formation of a group of nif/fix and nod genes as well as further evolutionary processes, where these groups of genes were transferred most likely together despite their relative isolation into a symbiotic cluster.

So the topological structure of the meta-tree probably refers us to the time of the formation of the symbiotic cluster of the protosym-biont of the tribe Fabeae when a symbiotic cluster was formed from the rhizobia species existing by that time by recombination for the needs of a new tribe of legumes and subsequent transfer to the ancestral species of soil bacteria that gave rise to the protosymbiont R. le-guminosarum. Since we have cited above evidence in favor of [Saf-ronova et al., 2018] that in the early stages of evolution nitrogen fixation could be performed by a microbial consortium in which a complex of symbiotic genes could be divided into different organisms. It can be assumed that the assembly of a single symbiotic cluster through horizontal transfer was carried out inside the nodule where the selection of promising protosymbionts apparently took place.

Comparative analysis of the genomes of R. leguminosarum bv. viciae made it possible to show that evolution in this group of organisms followed the same patterns that were revealed at the level of the order of Rhizobiales: gain-and-loss of genes as well as compaction of the symbiotic cluster. Here we see the same phenomena: functional

and structural redundancy (size of the symbiotic cluster, nodX and fxW genes), lost at the stage of compaction, ordering of the structure (transfer of the nodT gene into the symbiotic cluster). These processes are subordinated to the main goal - the evolutionary adaptation of rhi-zobia to the conditions of symbiosis and ensuring maximum adaptability to the soil-plant system. An important function is contained in the each gene mentioned by us: the presence of the nodX gene in the ancestral genorypes and its loss in the derived genotypes corresponds to a narrowing of the host specificity in the latter (due to the loss of the ability to enter into symbiosis with Afghan pea lines), accompanied by an increase in the efficiency of symbiotic interaction; the loss of the fixW gene the product of which destroys cysteine-rich bonds in proteins possibly leads to a deepening of bacteroid differentiation which is stimulated by NCR peptides; the transfer of the nodT gene into a symbiotic cluster completes the stage of recruiting a copy of this chromosomal gene which originally encodes one of the transport systems due to the needs of symbiosis (export of the Nod-factor from the cell) integrating it into a single operon with the genes responsible of the synthesis of the Nod-factor. To what was said above about the fixNOPQ chromosome cluster, it should be added that the final stage of its evolution consisted most likely in the transfer of a copy of this operon to the rhizobial chromosome and as a consequence the exclusion of Tardiphaga from symbiosis - a kind of "compactization" of the symbiotic consortium. Finally, the compaction of the symbiotic

cluster by the removal of almost half of the intergenic symbiotic material (often duplicating systems already existing on the chromosome) facilitates the horizontal transfer of the formed symbiotic cassette which is a highly important process in the evolution of rhizobia.

Table 1 summarizes the ancestral and advanced traits in the genomes of R. leguminosarum and describes the putative adaptive effects of the transition from the ancestral to the derivative group.

Differences between ancestral (A) and derivative (D) groups of genotypes of Rhizobium leguminosarum bv. viciae [Chirak et al., 2019].

Differences Group A Group D Possible outcome of the transition A ^ D

An affinity to the host Vavilovia formosa and primitive ("Afghan") genotypes of Pi- sum sativum Advanced ("European") genotypes of P. sativum, Vicia and Lathurus Adaptation to new species that have arisen in the tribe Fa-beae

Extrachromosomal sym-cluster size, kb > 90 < 60 Increased mobility of sym genes in the population

Localization of nodT Outside the nod-cluster Inside a nod-cluster Improved transport efficiency of the Nod-factor

Presence of nodX + — Narrowing host range

Presence of fixW + Deepened differentiation of N2-fixing bac-teroids

Presence of a chromosomal copy of the fxNOPQ operon * + Improving fitness in microaerobic (soil, nodules) niches

* In addition to the plasmid, symbiotically specialized copy

Ancestral features in the genome organization of R. leguminosarum bv. trifolii

We spoke above about the mystery of the evolution of clover rhizobia, the microsymbionts of which belong to the same species as the pea rhizobia (although all other genera of the tribe Trifolieae are nodulated by other rhizobia species). In the course of a preliminary analysis of their evolutionary history we analyzed a sample of clover rhizobia genomes [Aksenova et al., 2020] where it was shown that clover rhizobia like vavilovia rhizobia retain a number of ancestral features including the presence of the nodX and nifW genes, size of symbiotic cluster and functional redundancy. Little is known about the mechanisms that allowed clover rhizobia in contrast to most pea rhizobia to retain ancestral features of the symbiotic genome. However it seems logical to assume that the ancestral organization of the sym-genes bv. trifolii is associated with a much less pronounced diversification of the genus Trifolium compared to the tribe Fabeae. However it is already clear today that the study of clover rhizobia provides further evidence that ancestral features detected in clover and pea rhizobia were characteristic of the protosymbiont R. leguminosarum before its divergence into clover and vetch biovars.

CHAPTER 3 PLANT-CONTROLLED EVOLUTION OF RHIZOBIA The system of symbiotic receptor genes of legumes

For a long time the study of the genetic systems of legume-rhi-zobial symbiosis was limited by the genes of rhizobia a whole spectrum of which (groups nif, fix, nod) was identified and analyzed in detail by the end of the 20th century. At the same time progress in the study of symbiotic plant genes was much more modest. Thus by that time a number of plant mutants were obtained that were defective in various stages of symbiosis development which were designated sym1-sym42 in peas [Tsyganov, Tsyganova, 2020]. However identification, cloning and sequencing of these genes became possible only at the beginning of the XXI century [Madsen et al., 2003; Radutoiu et al., 2003; Arrighi et al., 2006]. It was shown that these genes control a wide range of symbiotic functions from the first stages of signaling to participation in the later stages of the development of intracellular symbiosis. In the framework of this work we focused specifically on the systems of leguminous plants involved in the reception of the symbiotic signal of rhizobia. These are receptor protein kinases containing LysM domains encoding for rhizobial Nod-factor receptors. Among the genes encoding for symbiotic receptors in legumes two groups are distinguished - LYK and LYR, which differ in the presence or absence of an activation loop in the kinase domain: hom-

ologues: LjNFR1-flower / MtLYK3-alfalfa / PsSym37-peas have active protein kinase, and LjNFR -alfalfa / MtLYR3-alfalfa / PsSym10-peas - do not have [Arrighi et al., 2006; Limpens et al., 2003; Madsen et al., 2003; Radutoiu et al., 2003; Smit et al. 2007; Zhukov et al., 2008]. An important feature of receptor genes is that they function in the form of heterodimers consisting of at least two subunits - LYK and LYR. For example the first of the identified lyadvenets receptor heterodimers consists of two NFR1 / NFR5 subunits [Madsen et al., 2003; Radutoiu et al., 2003; Arrighi et al., 2006]. The physical interaction of subunits has been proven experimentally [Fliegmann et al., 2016; Kirienko et al., 2017], although the evidence for the binding of the Nod-factor to individual subunits still presents a serious problem since the binding should be shown not with individual subunits, but with a heterodimer.

An important question is the origin of the symbiotic receptor system in leguminous plants. At present, it is generally accepted that they can originate from plant receptor systems that interact with signaling molecules based on chitin when interacting with mycorrhizal or pathogenic fungi which may be related to the CERK1 kinase in Arabidopsis thaliana [Nakagawa et al., 2011]. Based on the data of the analysis accumulated to date it can be concluded that receptor protein kinases had a single common ancestor in legumes and non-legumes [Zhu et al., 2006] and their further evolution was associated with duplication, neofunctionalization and specialization for the

needs of legumes-rhizobial symbiosis [Zhang et al., 2007]. In legume genomes clusters of LYK and LYR homologues can consist of several genes very similar in sequence which is rather difficult for their functional analysis for example to find out which of the pea NFR1 homologues is involved in the symbiotic receptor heterodimer.

Molecular basis of the host specificity of rhizobia of Afghan and European pea lines

One of the most important functions of receptor genes in leguminous plants is to control the specificity of interaction with rhizobia. A well-known and long-standing example of the divergence of host specificity within the CIG, i.e. the formation of microspecificity groups is the story of Afghan and European peas. More than 40 years ago [Lie, 1978] it was shown for the first time that Afghan pea lines cannot be inoculated with many strains of the species R. legumi-nosarum bv. viciae but only those that (as it became known later carry the nodX gene encoding for acetyltransferase and responsible for decorating the usual Nod-factor of pea rhizobia with an acetyl group at the reducing end of the signaling molecule [Firmin et al., 1993]). In the context of this work this phenomenon is interesting because the loss of the nodX gene as shown above is part of the compaction of the genome of the protosymbiont R. leguminosarum bv. viciae before its further spread to temperate latitudes that is important for understanding the evolutionary patterns of rhizobia as they move to new habitats [Provorov et al., 2013]. In this context the symbiotic features

of Afghan and European peas illustrate the narrowing of the host specificity accompanying the evolution of secondary rhizobia: rhizo-bia that have lost the nodX gene at the same time lose the ability to enter into symbiosis with primitive Afghan pea lines (Figure 7).

R.leg. bv. viciae RCAM1026 R.leg. bv. yiciae TOM

nodX - nodX +

Figure 7. Symbiotic phenotypes of Afghan and European pea lines in interaction with rhizobia with and without the nodX gene [Soloviev et al., 2021].

It is clear that the disclosure of the molecular mechanisms of host specificity in this case is impossible without analyzing the genome of the host plant namely that part of it that controls the signal differential interaction with the acetylated and non-acetylated Nod-factor. For a long time it was believed that there is one plant symbiotic gene responsible for this phenotype called Sym2, respectively, its "Afghan" allele Sym2A causes an effective symbiosis with rhizobia

producing doubly acetylated Nod-factor and the Nod- phenotype with European rhizobia. But for a number of objective reasons it was not possible to identify this gene in the pea genome for a long time. Vladimir Zhukov and his colleagues using the specific features of a number of pea receptor genes managed to identify a candidate for this role namely the gene for the receptor LysM domain containing protein ki-nase presumably constituting an active symbiotic receptor that directly interacts with the Nod-factor of rhizobia in the form of a receptor heterodimer (with the Sym10 receptor) [Sulima et al., 2019].

Using molecular modeling and docking we have shown [Solo-viev et al., 2021] that the interaction of the LykX / Sym10 receptor dimer and the docking of the acetylated and non-acetylated Nod-factor correspond to the classical picture of the interaction of Afghan and European pea lines with rhizobia containing and not containing the gene nodX (Figure 8).

Technically this study demonstrates the heuristic potential of computational methods that significantly complement wet biology methods. But in the context of this work the most interesting thing is that here we are dealing with the molecular mechanisms of co-evolution of the macro- and microsymbiont namely the narrowing of the host specificity during the evolution of symbiotic systems that we talked about above. On the part of rhizobia this mechanism consists in the loss of the whole nodX gene and on the part of the plant only a few amino acid substitutions in the receptor protein.

Figure 8. Molecular modeling of the interaction of acetylated and non-acety-lated Nod-factor with Afghan and European receptor heterodimer LykX / Sym10 in peas. Molecular dynamics is available by QR code [Soloviev et al., 2021].

Apparently the key amino acid positions are affected here since we have many examples when homologous receptors belonging to different legume genera and having dozens of amino acid substitutions nevertheless effectively interact with the same Nod-factors within the same group of cross-inoculation [Igolkina et al., 2019].

The effect of evolutionary "molding" as a consequence of micro-coevo-lutionary processes in conjugated populations of a legume plant and its rhizobia

In a number of our works the presence of a close relationship between the genetic diversity of populations of macro- and micro-symbionts was detected. The first observation was the analysis of genetic analysis of the diversity of galega rhizobia in the North Caucasus [Andronov et al., 2003; Osterman et al., 2011]. To assess the genetic diversity of this coupled symbiotic system we used the AFLP (Amplified Fragment Lengths Polymorphism) genomic fingerprinting method for both plant and rhizobial populations. Figure 9 shows a comparison of the corresponding dendrograms constructed from the results of cluster analysis of genomic fingerprints of micro- and mac-rosymbionts. From the data presented in Figure 9 it is quite obvious that the genetic diversity of the rhizobia of Galega orientalis and G. officinalis is in full agreement with the biodiversity of the host plant populations. The latter can be explained by the history of the settlement of the Caucasus by these species as evidenced by the differences in the ecology of these species: G. orientalis inhabits the most fertile areas both in mountain ecotopes and in the valleys while it is represented by at least 4 morphotypes [Andronov et al., 2003], while the galega occupies mainly ecological "inconveniences" and is often a part of the pioneer flora and is morphologically conservative. This probably testifies to the prescription of the settling of the Caucasus by the G. orientalis in comparison with G. officinalis. Be that as it

may molecular analysis shows significant differences between species of galega in terms of the level of genetic diversity.

Figure 9. Comparison of the genetic diversity of populations of Galega orientalis and G. officinalis with the diversity of their microsymbionts according to the AFLP genomic fingerprinting data [Osterman et al., 2011].

Later when genomic sequencing of rhizobia became available we obtained more detailed data on the comparative divergence of rhi-zobial genomes of galega rhizobia which we discussed above in the section devoted to speciation in rhizobia. Figure 10 shows a diagram of the comparison of nucleotide polymorphisms in the collection of galega rhizobia from the same Caucasian population of both biovars at chromosomal and symbiotic loci [Karasev et al., 2019]. From the data presented it is obvious that there is a correspondence between

the data of genomic fingerprinting of rhizobia and genomic diversity of plants.

Figure 10. Comparison of the nucleotide diversity (p-distance within biovars) of the genomes of the rhizobia of the Galega oriental's and G. officinalis for the core and symbiotic genes (a, b). Divergence (p-distance between biovars) by gene groups (c) [Karasev et al., 2019].

In the functional aspect it is interesting that the divergence in symbiotic genes between strains of different biovars is very uneven and its maximum falls on the genes involved in nitrogen fixation and not on the genes involved in the synthesis of the Nod-factor. This is consistent with the symbiotic features of the galega rhizobia belonging to different biovars - without losing the ability to form nodules

with the "wrong" host they are not capable of effective nitrogen fixation; in this case the divergence peak falls not on signaling systems but on the host specificity of nitrogen fixation.

However the connection between the diversity of macro- and microsymbiont was most clearly demonstrated by us at the next stage of our work [Igolkina et al., 2019], where its objects were populations of wild vetch (Vicia sativa), ranch (Lathyrus pratensis) and clover (Trifolium hybridum) as well as their microsymbionts selected from a small area of a perennial deposit in the Leningrad region. It is important to note that in this experimental set there are representatives of two different groups of cross-inoculation - vetch and rank belong to the pea group, clover is a separate group. Representatives of both groups enter into symbiosis with the same species of rhizobia - R. leguminosarum but represented by two different biovars - bv. viciae and bv. trifolii. Biovars differ genetically from each other only in very restricted group of symbiotic genes (with a common chromosomal background) which is expressed in a clear symbiotic behavior - plants do not form nodules at all with "foreign" biovars (or occasionally form nodules that are abnormal in structure and do not fix nitrogen). A feature of this study was that the objects of analysis were DNA pools that combine either plant material of each species or nodules taken from all plants used for research or soil samples taken in the root zone of all studied plants. For sequencing plant genes were selected that control closely coupled symbiotic signaling systems: for

plant pools a region of the NFR5 gene encoding a receptor that directly interacts with the Nod-factor was sequenced (100 random clones were sequenced for each of the plants); in the case of soil collected from the roots and nodule pools high-throughput sequencing of amplicon libraries of the rhizobial nodA gene involved in the attachment of an unsaturated fatty acid residue to the Nod-factor which is one of the important factors of host specificity was used [Denarie et al., 1996]. Analysis of the resulting libraries revealed two important effects. Figure 11 shows a comparison of the levels of nucleotide polymorphism of plant and rhizobial components. It can be seen from the data presented that the levels of nucleotide diversity of the plant and rhizobial components of the symbiotic signaling system are almost linearly dependent.

However the matter was not limited to detecting the conformity of nucleotide diversity. When comparing the phylogenies of the plant recipe and the rhizobial gene nodA the topological transformation of the phylogenies of rhizobia during the transition from soil to plant was shown. Figure 12 shows the data for this transformation for the Vicia rhizobia pool.

Е

13

0.0040 0.0128 0.0222

Lathyrus Vicia Trifolium

7r diversity in plant NFR5 populations

Figure 11. Comparison of nucleotide polymorphism in the plant recipe libraries (NFR5) and amplicon libraries of the rhizobial nodA gene [Igolkina et al., 2019].

Ризобиальный Ризобиалъный

клубеньковый почвенный

пул пул

nodA nodA

Растительный рецептом NFR5

Figure 12. Transformation of phylogeny by the nodA gene of the rhizobial pool during the transition from soil to plant: phylogeny acquires the features of a plant recipe (NFR5) [Igolkina et al., 2019].

The figure shows that when passing from soil to plant the phy-logeny of the rhizobial pool for the nodA gene begins to acquire topological features that are very similar to the topology of the phyloge-netic tree of the plant receptor gene NFR5. To statistically evaluate this phenomenon a specially developed statistical criterion was used using Gaussian mixture models which made it possible to confirm this effect and give a numerical estimate of the similarity of topologies for trees with different numbers of unmarked branches [Igolkina et al., 2019] namely it was shown that that the AG values, reflecting the differences of rhizobial topologies with the topologies of the plant recipe, significantly and significantly decrease upon transition to the nodule pool (in the case of clover and vetch, it is almost 2 times).

Thus this study showed a clear relationship between the diversity of macro- and microsymbionts, not only at the level of nucleotide polymorphism but also at the level of phylogeny topologies. We called this effect "evolutionary molding" during which the genetic diversity of a microsymbiont is "pressed" into a rigid matrix of genetic diversity of a macrosymbiont leading not only to a correspondence between the general levels of diversity but also their topologies. The latter circumstance allowed us to introduce a new type of diversity "topological P-diversity" to assess and formalize these effects [Igolkina et al., 2019] the main feature of which is that the object of comparison is not nucleotide (allelic) diversity between communities

but diversity topologies of phylogeny, and the basic parameter of which is the measure of topological congruence.

It is interesting to note that this effect was also seen in the study of conjugated populations of the galega and its microsymbionts (Fig. 9) where one can see not only the correspondence of the diversity but also the topologies. It is important to note that there is no molding effect in systems with a low level of functional coupling. Thus in our last work it was shown that the diversity of the rhizosphere microbi-ome was not correlated with the genetic diversity of the root mass [Zverev et al., 2021] where in both cases the genetic diversity was assessed by deep sequencing of amplicon libraries of ribosomal fragments operon.