Do­mes­tic Just for the Sake of it...

The Evo­lu­tion of a Fun­gus with Good Taste

by Daniel P. Haeusser

We know quite a bit about how the wild au­rochs or their ilk evolved into tame, bossy cows and how the in­signif­i­cant grass teosinte be­came nu­tri­tious maize, but what do we know about the evo­lu­tion of mi­crobes in­volved in food and drink pro­duc­tion? For thou­sands of years be­fore the ad­vent of mi­cro­bi­ol­ogy, our an­ces­tors used mi­crobes for this pur­pose with­out knowl­edge of the con­tri­bu­tions or the do­mes­tic ori­gins of these small work­horses. Even to­day, when mi­crobes con­tinue in key roles dur­ing food and bev­er­age pro­duc­tion, we know far less about their evo­lu­tion of do­mes­ti­ca­tion than we do for the process in an­i­mals or plants.

Why should that mat­ter? Ob­vi­ously, we hu­mans would like to en­cour­age mi­crobes that ben­e­fit us and hin­der mi­crobes that harm us. The key to this is un­der­stand­ing what evo­lu­tion­ary routes can lead to one state or the other and how they can be pur­pose­fully al­tered to fa­vor the ben­e­fits.

As­pergillus oryzae, a ver­sa­tile food chemist

A re­cent pub­li­ca­tion by Gib­bons et al. an­a­lyzed the 'evo­lu­tion­ary im­print' of mi­cro­bial do­mes­ti­ca­tion us­ing ge­nomic data from the fil­a­men­tous fun­gus As­pergillus oryzae. This fun­gu­sis used in the food and bev­er­age in­dus­try to hy­drolyze poly­sac­cha­rides dur­ing the pro­duc­tion of miso, rice vine­gar, soy sauce, and no­tably, sake (Fig­ure 1). The fun­gus is such a vi­tal part of Asian cul­tures and cuisines that the Japan­ese manga (comic) Moy­asi­mon and its anime adap­ta­tion have even made it a sup­port­ing char­ac­ter in the se­ries.

Fig­ure 1: The im­por­tance of sake and As­pergillus oryzae to tra­di­tional and mod­ern Japan­ese cul­ture. (A) Sukeroku shi­rozak euri (~1775 – 1781) by Ko­ryü­sai Isoda. Print show­ing sell­ing of white sake. Li­brary of Con­gress. (B) A field of the As­pergillus oryzae char­ac­ter from the Japan­ese manga/anime se­ries Moy­asi­mon: Tales of Agri­cul­ture (via Google Im­ages). (C) A coni­d­ium (asex­ual, non-motile spores) of As­pergillus oryzae. Source

Do­mes­ti­ca­tion of A. oryzae in­volved the se­lec­tion of strains with pleas­ing fra­grances, rapid mycelial growth, high yield of amy­lases, and low pro­duc­tion of pig­ments or tox­ins. In other words, hu­mans of the dis­tant past in­ad­ver­tently en­cour­aged the suc­cess and sur­vival of strains that could break down com­plex car­bo­hy­drates in soy and rice into sim­pler com­pounds that en­hance prod­uct fla­vor. This tamed fun­gus likely de­rived from a strain of A. flavus, an agri­cul­tural pest of seed crops that pro­duces car­cino­gens, the afla­tox­ins, which tar­get the liver. In­ter­est­ingly, afla­toxin is also toxic to yeast, an­other fun­gus in­te­gral to sake pro­duc­tion, par­tially ex­plain­ing why do­mes­ti­ca­tion of A. oryzae se­lected against it.

Gib­bons et al. first looked at the ge­nomic dif­fer­ences be­tween strains of A. oryzae and its wild rel­a­tive A. flavus (95.5% ge­netic sim­i­lar­ity). Six­teen strains split evenly be­tween A. oryzae and A. flavus re­vealed a mono­phyletic (i.e. they all de­scended from the same com­mon an­ces­tor) A. oryzae with only ~25% of the ge­netic di­ver­sity found in the A. flavus iso­lates (Fig­ure 2). Fur­ther­more, while 86 poly­mor­phic (ge­net­i­cally vari­able) sites were iden­ti­fied in A. flavus, only 6 were found in A. oryzae. In other words, do­mes­ti­ca­tion of A. oryzae from a sin­gle an­ces­tor has to a large ex­tent fixed its traits from fur­ther nat­ural evo­lu­tion­ary vari­a­tion, in con­trast to what oc­curs in its wild an­ces­tor.

Fig­ure 2: From Gib­bons et al. (A) Nu­cleotide vari­a­tion across chro­mo­somes for A. oryzae and A. flavus iso­late genomes, as mea­sured by Watterson's test, a sta­tis­ti­cal cal­cu­la­tion of ge­netic vari­a­tion in a pop­u­la­tion. (B) Phy­lo­ge­netic re­la­tion­ship be­tween the six­teen A. oryzae and A. flavus genomes.

Pos­i­tive Se­lec­tion for Food-En­hanc­ing Genes

The re­searchers then searched the A. oryzae genomes for re­gions that show ev­i­dence of re­cent pos­i­tive se­lec­tion. They as­sumed that al­le­les un­der pos­i­tive se­lec­tion should be 'linked' (through syn­teny) to neigh­bor­ing chro­mo­so­mal re­gions that show re­duced vari­a­tion by virtue of prox­im­ity, de­spite be­ing un­der neu­tral se­lec­tive pres­sure them­selves. This is be­cause chro­mo­so­mal genes close to one an­other tend to be in­her­ited to­gether dur­ing meio­sis, so if a given gene is ben­e­fi­cially kept dur­ing evo­lu­tion, its close neigh­bors usu­ally are too. Not sur­pris­ingly, the sixty-one iden­ti­fied pu­ta­tive re­gions of pos­i­tive se­lec­tion in­cluded genes in­volved in pri­mary or sec­ondary me­tab­o­lism and degrada­tive en­zyme pro­duc­tion or se­cre­tion. These gen­eral processes are all im­por­tant dur­ing the break­down of grains into palat­able prod­ucts in food and bev­er­ages. For in­stance, in­cluded within one iden­ti­fied re­gions is a gene for glu­t­a­m­i­nase, an en­zyme that con­verts glu­t­a­mine to glu­ta­mate, a well-known fla­vor en­hancer found at high lev­els in prod­ucts such as sake.

Fig­ure 3: (A) Fig­ure from Gib­bons et al. show­ing the com­par­isons be­tween the sesquiter­pene syn­the­sis clus­ters of A. oryzae and A. flavus. Gray shad­ing rep­re­sents re­gions with high se­quence sim­i­lar­ity. (B) The struc­ture of iso­prene, the build­ing block of ter­penes, a chem­i­cal group that in­cludes the sesquiter­pene class of com­pounds. Source. (C)Tri­cho­derma virens, a sym­bi­otic fun­gus of plants that pro­duces fla­vor-en­hanc­ing sesquiter­pene com­pounds, here shown grow­ing on wheat straw in solid-state fer­men­ta­tion. Source

Fur­ther genome com­par­isons be­tween the two fungi high­lighted ad­di­tional in­stances of pos­i­tive se­lec­tion in A. oryzae of key meta­bolic genes, in­clud­ing an aug­mented gene clus­ter re­lated to sesquiter­pene syn­the­sis. These are chem­i­cals used pri­mar­ily in plants and in­sects for de­fense and as pheromones (Fig­ure 3) that can also be food-fla­vor­ing aro­mat­ics. While con­served and fixed in A. oryzae iso­lates, this gene clus­ter is more vari­able and re­duced in A. flavus genomes. Fi­nally, look­ing at the tran­script lev­els of A. oryzae re­vealed mas­sive up­reg­u­la­tion of car­bo­hy­drate me­tab­o­lism, in­clud­ing a sugar mod­i­fier used in com­mer­cial sweet­ener pro­duc­tion and an en­zyme used to re­duce acry­lamide lev­els in starch-rich foods. Con­sis­tent with A. oryzae's pri­mary do­mes­ti­cated role in starch degra­da­tion, its most abun­dant tran­script is for a‑amylase and is en­coded in mul­ti­ple loci.

The do­mes­ti­ca­tion of A. oryzae thus was dri­ven through spe­cific meta­bolic changes, not via changes in phys­i­cal form, as seen with plants or an­i­mals. This makes sense, given that hu­mans se­lected for bet­ter tast­ing prod­ucts. Im­por­tantly, these meta­bolic changes in­clude both the up­reg­u­la­tion of spe­cific car­bo­hy­drate ca­tab­o­lism, but also in­cluded down­reg­u­la­tion of genes in­volved in toxin pro­duc­tion and of en­zymes in­volved in degra­da­tion of plant poly­sac­cha­rides whose prod­ucts do not en­hance fla­vors that hu­mans ap­pre­ci­ate.

Food for Thought

For me, this study brings up two spec­u­la­tions in why mi­cro­bial do­mes­ti­ca­tion that de­serves fur­ther con­sid­er­a­tion and dis­cus­sion (Com­ments are wel­come!):

1) Lab­o­ra­tory strains are ba­si­cally do­mes­ti­cated for growth in the lab. Re­searchers are aware, at least in the backs of our minds, that what we see in the lab may not be im­por­tant in the wild and vice versa. But how deeply does this di­vide run? Does the re­peated growth of a strain in a lab keep cer­tain poly­mor­phic phe­no­types from ever aris­ing, or is there bias se­lec­tion to­ward phe­no­types we hope for? Does it mat­ter if a genotype/phenotype ap­pear­ing in a lab ex­per­i­ment is un­likely to oc­cur in the wild, and how can we know?

2) If path­o­genic mi­crobes like A. flavus can be do­mes­ti­cated for hu­man ben­e­fit, it is con­ceiv­able that do­mes­ti­ca­tion of other pathogens could be a method of con­trol­ling hu­man dis­ease. The idea has been brought up be­fore, such as in this TED talk by Paul Ewald. But it re­mains un­cer­tain how eas­ily this could be achieved or how broadly ef­fec­tive it could be. Even if a more ben­e­fi­cial, do­mes­ti­cated strain could be pro­duced, this would not nec­es­sar­ily lead to the erad­i­ca­tion of the ex­tant pathogen. Af­ter all, A. flavus still is present through­out the world and is a fre­quent con­t­a­m­i­nant of foods.  A do­mes­ti­cated pathogen would there­fore need to not just be harm­less to hu­mans, but also do one of two things: out­com­pete the wild strain for oc­cu­pa­tion of the host or prime the host im­mu­nity to fight off the wild path­o­genic form.

It will be in­ter­est­ing to see how far the di­rected do­mes­ti­ca­tion of other mi­crobes is pos­si­ble, whether for sci­ence, health, or the cui­sine. Per­haps, fol­low­ing in the steps of Wylie Dufresne, the next gen­er­a­tion of mol­e­c­u­lar gas­tron­o­mists will be com­pos­ing their own unique meta­bolic vari­ants of mi­cro­bial genomes to do­mes­ti­cate mi­crobes for sig­na­ture dish fla­vors. A Top Chef cri­tique of the fu­ture?: "I'm sorry. The fun­gus you used just ex­pressed too much a‑amylase. The dish was just sick­en­ingly sweet. Please pack your strains and go."

 

Ref­er­ence

Gib­bons JG, Sali­chos L, Slot JC, Rinker DC, Mc­Gary KL, King JG, Klich MA, Tabb DL, Mc­Don­ald WH, Rokas A (2012). The evo­lu­tion­ary im­print of do­mes­ti­ca­tion on genome vari­a­tion and func­tion of the fil­a­men­tous fun­gus As­pergillus oryzae. Cur­rent Bi­ol­ogy : CB, 22 (15), 1403−140.9 PMID 22795693

 

Daniel is a post­doc­toral fel­low in the Mar­golin Lab in the De­part­ment of Mi­cro­bi­ol­ogy & Mol­e­c­u­lar Ge­net­ics at the Uni­ver­sity of Texas, Hous­ton Med­ical School. He also teaches as an ad­junct pro­fes­sor at the Uni­ver­sity of Hous­ton-Down­town in the De­part­ment of Nat­ural Sci­ences.