022 w3310 13 - virology › w3310 › 022_w3310_13.pdf · lecture’22 biologyw3310/4310 virology...

Evolu&onLecture 22

Biology W3310/4310Virology

Spring 2013Anything produced by evolu3on is bound to be a bit of a mess.SYDNEY BRENNER

Around here, it takes all the running you can do just to stay in the same place.LEWIS CARROLLAlice in Wonderland

1

• Where did viruses come from?

• Where are they going?

• How are they geLng there?

• How would we know?

Today’s ques&ons

2

Modern virology has provided a window on the mechanisms of evolu&on

• As host populaNons grow and adapt, virus populaNons are selected that can infect them-‐ New viral popula-ons emerge every day

• It also works the other way-‐ Viral popula-ons can be significant selec-ve forces in the

evolu-on of host popula-ons

• If a host populaNon cannot adapt to a lethal virus infecNon, the populaNon may be exterminated

3

The public is constantly confronted with the reality of viral evolu&on (even if they don’t

believe in evolu&on)

• New viral diseases: AIDS, West Nile virus in the US, hepaNNs C, Ebola, Marburg...

• Regular bouts every year with influenza and common cold viruses

• Drug resistant HIV

Simple fact: viruses evolve faster than many can comprehend

4

Viral evolu&on: The constant change of a viral popula&on in the face of selec&on pressures

• If the populaNon can’t change or adapt, it disappears

• Viral populaNons display spectacular diversity -‐ why they are successful

• Sources of diversity:-‐ Muta-on

-‐ Recombina-on

5

• Virus evoluNon is defined in terms of a populaNon, not an individual virus parNcle

• Viral populaNons comprise diverse arrays of mutants that are produced in prodigious quanNNes

• It is misleading to think of an individual parNcle as represenNng an ‘average’ virion for that populaNon -‐ virologists are populaNon biologists whether they know it or not

Viral evolu&on

6

When it comes to viral evolu&on, the median is not the message (Stephen Jay Gould)

• Viral populaNon diversity provides surprising avenues for survival, yet every individual is a potenNal winner

• SomeNmes even the most rare genotype in a populaNon may be the most common a]er a single selecNve event

7

Four main drivers of virus evolu&on

• Large numbers of progeny

• Large numbers of mutants

• Quasi-‐species effects

• SelecNon

8

Virus-‐infected cells produce large numbers of progeny

The interface of host defense and virus replicaNon is ferNle ground for selecNon and evoluNon

Principles of Virology, ASM Press

9

Replica&ng viruses produce large numbers of mutant genomes

• EvoluNon is possible only when mutaNons occur in a populaNon

• MutaNons are produced during copying of any nucleic acid molecule

10

• The masters of error-‐prone replicaNon: RNA polymerases cannot correct errors

• Average error frequency: 1 in 104 or 105 nucleoNdes polymerized

• In a 10 kb RNA virus genome, a mutaNon frequency of 1 in 104 results in about 1 mutaNon per genome

RNA viruses

11

DNA viruses

• Different lifestyle from most RNA viruses

-‐ Narrow host range

-‐ Persistent infecNons common

• Genome replicaNon not as error prone as RNA viruses

• Proofreading

• Most DNA viruses generate less diversity, evolve slower than RNA viruses

12

The quasispecies concept

• Analysis of an RNA bacteriophage populaNon (Qβ):“A Qβ phage populaNon is in a dynamic equilibrium with viral mutants arising at a high rate on the one hand, and being strongly selected against on the other. The genome of Qβ cannot be described as a defined unique structure, but rather as a weighted average of a large number of different individual sequences.” E. Domingo, D. Sabo, T. Taniguchi, C. Weissmann. 1978. NucleoNde sequence heterogenity of an RNA phage populaNon. Cell 13:735-‐744.

• This discovery was far ahead of its Nme, not appreciated by most virologists

• Virus populaNons exist as dynamic distribuNons of nonidenNcal but related replicons, called quasispecies

13

Viral quasispecies


14

• Most viral infecNons are iniNated not by a single virion, but rather by a populaNon of parNcles

• The large number of progeny produced are complex products of intense selecNve forces inside the host

• The survivors that can re-‐infect a new host reflect the intense selecNve forces outside the host

• When small populaNons of virus parNcles replicate (as in laboratories), extreme fluctuaNons in genotype and phenotype are possible

Quasispecies effects

15

• For a given RNA virus populaNon, the genome sequences cluster around a consensus or average sequence, but virtually every genome can be different from every other

• It is unlikely that a genome with the consensus sequence is actually replicaNng in the populaNon

The myth of consensus genome sequences

16

Selec&on

• SelecNon favors both the creaNon of diversity and single dominant mutaNons-‐ Counterintui-ve -‐ the drive for survival must result in diverse

popula-ons as well as selected muta-ons

17

• Survival of the fi?est: A rare genome with a parNcular mutaNon may survive a selecNon event, and this mutaNon will be found in all progeny genomes

• Survival of the survivors: However, the linked, but unselected mutaNons, get a free ride

• Consequently, the product of selecNon a]er replicaNon is a new, diverse populaNon that shares only the selected mutaNons

Selec&on

18

Selec&on for survival inside a host

• At the end stage of AIDS, the billions of virions that survive are T-‐cell tropic, engage CXCr4 receptor

• When infecNon is passed to a new host, the iniNal replicaNng genomes are M-‐tropic, engage CCr5 receptor

• Progeny genomes have passed through a booleneck, only a few pass on

• Virions that ulNmately devastate the immune system are not the most fit for infecNon of new hosts

19

• Increased mutaNon rates are selected during virus evoluNon: mutaNon is good for viral populaNons

• It is possible to isolate mutaNons in viral polymerases that reduce the frequency of incorporaNon errors

• What was learned from the study of these mutants

-‐ They do not have a selecNve advantage when wild type and anN-‐mutators are propagated together

-‐ Lower rates are neither advantageous nor selected in nature

-‐ The mutants are o]en less pathogenic

• High mutaNon rates are a posiNve force in virus evoluNonhop://www.virology.ws/2009/05/15/increased-‐fidelity-‐reduces-‐viral-‐fitness/

Diversity is selected

20

Error threshold

• The capacity to sustain prodigious numbers of mutaNons is a powerful advantage

• There must be a point at which selecNon and survival balance geneNc fidelity and mutaNon rate

• This limit is called the error threshold. If you exceed it, you are dead. If you live too far below it, you cannot produce enough mutaNons to survive selecNon

• RNA viruses: evolve close to their error threshold

• DNA viruses: evolve far below their error threshold21

Error threshold

• Expose a cell culture infected with a DNA virus to a base analog such as 5-‐azacyNdine

• 5-‐azacyNdine is incorporated as a C, but templates as a T (G to A transiNons)

• MutaNon rate among viral progeny increases several orders of magnitude

• When a similar experiment is done with an RNA virus, the error frequency per genome increases only two-‐ to threefold at best -‐ cannot make any more mutaNons

22

Error threshold

anNviral ribavirin and poliovirus

23

Gene&c boNlenecks

• Extreme selecNve pressures on small populaNons that result in loss of diversity, accumulaNon of non-‐selected mutaNons, or both

• A single RNA virus plaque is picked and expanded

• Next, a single plaque is picked from the expanded stock

• The process is repeated over and over


24

Gene&c boNlenecks

• The booleneck arises by restricNng further viral replicaNon to the progeny found in a single plaque

-‐ A few thousand progeny viruses derived from a single founder virus

25

• A]er about 20-‐30 cycles of single-‐plaque amplificaNon, many virus populaNons are barely able to grow

• They are markedly less fit than the original populaNon

• The environment is constant, and the only apparent selecNon is that imposed by the ability of the populaNon of viruses from a single plaque to replicate

• Why does fitness plummet?

Gene&c boNlenecks

26

• The answer lies in a phenomenon dubbed Muller’s ratchet: Small, asexual populaNons decline in fitness over Nme if the mutaNon rate is high

• ReplicaNng RNA viruses produce many mutaNons; they survive close to their error threshold

• By restricNng populaNon growth to serial single founders (the booleneck) under otherwise nonselecNve condiNons, so many mutaNons accumulate (exceed the threshold) that fitness decreases

Gene&c boNlenecks

27

The ratchet metaphor: each of the new mutaNons works like a ratchet, allowing the gear to move forward, but not backward

Each round of error-‐prone replicaNon works like a ratchet, “clicking” relentlessly as mutaNons accumulate at every replicaNon cycle

28


29

Fitness decline compared to ini&al virus clone aPer passage through a boNleneck


30

• InfecNon by a limited virus populaNon and subsequent amplificaNon are o]en found in nature-‐ Small droplets of suspended virus during aerosol

transmission

-‐ Ac-va-on of a latent virus from a limited popula-on of cells

-‐ Small volume of inoculum introduced in infec-on by insect bites

• How do infecNons that spread by these routes escape Muller’s ratchet?

BoNlenecks in the real world?

31

• Subject a more diverse viral populaNon to serial passage-‐ Don’t pick a single plaque, pool several plaques

• More diversity in the replicaNng populaNon facilitates construcNon of a mutaNon-‐free genome by recombinaNon or reassortment, removing or compensaNng for mutaNons that affect growth adversely

Avoiding the ‘ratchet’


32

• Even if such a recombinant is a rare species, it has a powerful selecNve advantage for growth-‐ Its progeny will ul-mately take over the popula-on

• The message is simple: Diversity of a viral populaNon is important for the survival of individual members

-‐ Remove diversity, and the popula-on suffers

Avoiding the ‘ratchet’

33

By exchange of gene&c informa&on

Sick virus 1

Sick virus 2

Healthy viral recombinant

34

• Allows the construcNon of viable genomes from debilitated ones and avoids Muller’s ratchet

• A similar mechanism is reassortment among genomic segments when cells are co-‐infected with segmented RNA viruses

• It is an important source of variaNon, as exemplified by orthomyxoviruses and reoviruses

Exchange of gene&c informa&on

35

Selec&on: Gene&c shiP & driP

• SelecNon of viral mutants resistant to eliminaNon by anNbodies or cytotoxic T cells is inevitable when sufficient virus replicaNon occurs in an immunocompetent individual

• Dri] -‐ diversity arising from copying errors and immune selecNon -‐ may occur each Nme a genome replicates

• Shi] -‐ diversity arising a]er recombinaNon or reassortment -‐ is relaNvely rare

36

Influenza viruses

• Influenza A viruses are classified by anNgenic composiNon, by serologic tesNng of HA and NA

• CombinaNons of H and N are called HxNy

• x = 1-‐17; y = 1-‐9

• H1-‐17 can infect birds; H1, H2, H3 can infect and transmit between humans

37


38

An&genic driP: Influenza virus


40

Exchange of gene&c informa&on


41

Selec&on: Is virulence a posi&ve or nega&ve trait?

• Idea: increased virulence reduces transmissibility because hosts die faster, reducing exposure to uninfected hosts

• ExpectaNon: all viruses evolve to be maximally infecNous and avirulent

• But this is not what is observed

• Persistent infecNons lie dormant for years, then kill host at end stage

• Virulent viruses for one species may be maintained as asymptomaNc infecNons in another

• For some diseases, increased virulence increases transmissibility and is selected for in natural infecNons

42

An experiment in virus evolu&on

• In 1859, the European rabbit was introduced to Australia for hunNng purposes

• Lacking natural predators, it reproduced to plague proporNons in a short Nme

43


• The myxoma leporipox virus was released in Australia in the 1950s in an aoempt to rid the conNnent of the rabbits

• The natural host of myxoma virus is the cooontail rabbit, the brush rabbit of California, and the tropical forest rabbit of Central and South America

• The virus is spread by mosquitoes; infected rabbits develop superficial warts on their ears

• European rabbits are a different species, infecNon is 90-‐99% fatal

44

• In the first year, the released virus was efficient in killing rabbits with a 99.8% mortality rate

• By the second year the mortality dropped to 25%

• In subsequent years, the rate of killing was lower than the reproducNve rate of the rabbits, and hope for 100% eradicaNon was dashed


45


• Both rabbits and viruses produce large numbers of offspring

• The virus evolved to kill fewer rabbits and to extend the life of lethally infected rabbits so that the virus could overwinter and spread in spring mosquitoes

• The rabbits evolved to become more resistant or tolerant of the virus

• As predicted for an evolving host coming to an equilibrium with the pathogen

46

What was learned?

• You always get what you select, but you o]en don’t get what you want

• EliminaNon of rabbits with a virus was flawed idea-‐ Selec-ve forces that could not be controlled were at work

• In Australia, apparently nothing: experiments in biological control of rabbits using a lethal calicivirus are under way

47

The origin of viruses

• Regressive theory: viruses are derived from intracellular parasites that have lost all but essenNal genes

• Cellular origin theory: viruses arose from cells by reducNve evoluNon

• Independent-‐en&ty theory: viruses coevolved with cells from the origin of life (‘pickpockets’)

• But there is no fossil record, and few viral stocks more than 80 years old

• BioinformaNcs has provided insight48

Origin of RNA viruses from RNA cells

Different cell lineages

Biochimie 87 (2005) 793-‐803

49

Very large viral DNA genomes

• Mimivirus (1.2 Mbp), Phycodnavirus (330 kbp) genomes have sequence coherence: are not mosaics

• Homogeneity of genomes within family, lack of homology among families, difficult to explain using model that they arose by the acquisiNon of exogenous genes by a primordial precursor viral genome

• Megavirus


50

Origins of DNA viruses

• By rela*ng *mescale of herpesviral genome evolu*on with that of hosts, believe that three major groups of herpesviruses (alpha, beta, gamma) arose ~180-‐220 million years ago

51

Human viruses

• All known types of viruses likely evolved long before humans appeared on Earth

• All human viruses have therefore evolved from animal viruses

52

Origins of smallpox virus

• Sequence analysis of 45 isolates: genomes can be organized into 3 clades, which cluster geographically (West Africa, South America, Asia)

• Gene content is constant; lack of diversity indicates recent introducNon into humans

• Only member of poxvirus family with credible homology to smallpox is a gerbil poxvirus

• Perhaps human smallpox virus arose a]er a zoonoNc infecNon from infected gerbils

53

Origins of measles virus

• Measles virus is closely related to rinderpest virus, a bovine pathogen

• Probably evolved from an ancestral rinderpest virus when humans first began to domesNcate caole

• Established in the Middle East ~5,000 years ago, when human populaNons began to congregate in ciNes

• Spread around the world by colonizaNon and migraNon, reaching Americas in 16th century and destroying naNve Americans

54

The fundamental proper&es of viruses constrain and drive evolu&on

• Despite many rounds of replicaNon, mutaNon, selecNon, we can recognize a herpesvirus or influenza virus genome by sequence analysis

• Viral populaNons o]en maintain master or consensus sequences, despite opportuniNes for extreme variaNon

• How is stability maintained?

• All viruses share fundamental characterisNcs that define and constrain them

• Viruses that can funcNon within the constraints survive

55

Constraining viral evolu&on

• Extreme alteraEons in viral consensus genome do not survive selecEon

• The viral genome is one constraint

-‐ DNA cannot become RNA, or vice versa

-‐ Replica3on strategy -‐ cannot change; consider interac3on with host proteins

• Physical nature of capsid

-‐ Icosahedral capsids: defined internal space, fixes genome size

• SelecEon during host infecEon

-‐ A mutant too efficient in bypassing host defenses will kill host, suffer the same fate as one that does not replicate efficiently enough

56

Evolu&on of new viruses

• AssumpNon: new viruses can only arise from viruses that are now in existence, not de novo

• What is the number of all possible mutaNons of a viral genome?

• It is so large in fact that all the possibiliNes can never be tested in nature

• Sequence comparisons of several RNA virus genomes have demonstrated that well over half of all nucleoNdes can accommodate mutaNons

57

Evolu&on of new viruses

• For a 10 kb viral genome, there are more than 45000 sequence permutaNons that define all possible mutants

• If one considers deleNons, recombinaNon, and reassortment, the numbers become even larger

• Considering that there are roughly 4135 atoms in the visible universe, this is a large number indeed

58

• New mutants will always arise, and the possibiliNes literally are unimaginable

• Future viruses arise from extant viruses: mutants, recombinants, reassortants of what is out there

• We only know about 1% of the viruses that exist

• As mutaNon rates are probabilisNc and viral quasispecies are indeterminate, the very concept of predicNng the direcNons of viral evoluNon has liole meaning

Virus evolu&on is relentless

59

Food for thought

• It took 8 million years or more for simian primates to change 2% of their genomes to become human

• By contrast, poliovirus can change 2% of its geneNc structure in five days!

• This is about the Nme it takes for the virus to pass from the human mouth to the gut

• Imagine what a virus can do with 8 million years

60

022 w3310 13 - virology › w3310 › 022_w3310_13.pdf · lecture’22 biologyw3310/4310 virology...

Documents