Visualising viruses

Question 1

Are x-rays short or long wavelengths?

Answer 1

Short

Question 2

What is the R factor?

Answer 2

How good the data fits the model

Question 3

Is a high R factor good or bad?

Answer 3

Bad - lower the better

Question 4

Describe MS2 virus structure

Answer 4

T=3 icosahedral particle made of 180x subunits arranged as 90 dimers.

3x conformations of the coat protein - A/B/C where A/B dimerise and C/C dimerise. Major difference between these are in the FG loop region where a proline is - though proline isomerisation (cis and trans) was important but mutants showed it wasn’t.

Crystalised the particle but didn’t get any information about the RNA as the low-resolution data was lost by the beam stop and in the averaging process. Went on to soak crystals with RNA

Question 5

Why was the RNA detail of MS2 lost in the XRC structure?

Answer 5

Lose the low-resolution data due to the beam stop to protect the detector and lose data through averaging. The RNA is poorly ordered therefore will be low resolution.

Question 6

Cryo-EM - Zika virus example

Answer 6

Zika viruses is a Flavivirus like Dengue and yellow fever and it has recently been linked to birth defects. It is spread through mosquitos.

A paper is science reports a 3.8A cryo-EM structure of the mature virus particle which shows the structure is similar to the other flaviviruses except for a ~10aa near the Asn glycosylation site in each of the 180x envelope glycoproteins which make up the icosahedral shell. The carbohydrate group associated with this residue can clearly been seen in the Cryo-EM density and is suggested to function as an attachment site as this region varies among both Zika virus stains and flaviviruses therefore may influence the transmission of the disease. Targeting this entry site may be a possible anti-viral?

Question 7

Cryo-EM - Dengue virus example

Answer 7

+ sense, ssRNA genome with a host-derived lipid bilayer (enveloped) and ~500A in diameter.
24A resolution of dengue - lipid bilayer and genome observed in the layered particle. They fitted the XRC into the Cryo-EM structure and fitted well.
Using Cryo-EM they were able to see conformational changes with pH change which represents the entry via the endosome.

At pH 8.0 spikey virus –> pH 6.0 smooth –> pH 7.5 different spikes

The conformational change makes Furin (host-cell protease) cleavage site accessible on the virus CP. The loss of cleavage product makes the virus competent to fuse with membranes and infect other cells. This loss can only occur when the pH is raised again. Important part of the life-cycle of dengue.

See diagrams

Question 8

How can you visualize conformational change by Cryo-EM?

Answer 8

Can spray things onto the grid directly before plunging for freezing. Adding in ferritin or gold as a control to ensure it has hit the grid

Question 9

What is negative staining?

Answer 9

Virus is embedded in heavy metal salt e.g. uranyl acetate which generates lots of contrast

Question 10

What are the problems of negative staining?

Answer 10

• Non-native conditions - heavy metal salt and the virus is dried before which leads to flattening of the sample and deformation.
• As it is negative stain you are imaging where the sample is not.
• Limited resolution ~20A (envelope only).
• Can have inconsistencies in staining

However it is still routinely done in the lab to check for virus particles.

Question 11

What are the benefits and drawbacks of Cryo-EM

Answer 11

• Unstained and fully hydrated so no deformation of the structure
• Can image the internal features and the particle
• Can study conformational changes
• Solution structures like NMR - more native than XRC
• As using a microscope records an image so no phase problem
• No crystal needed
• Can have problems with radiation sensitivity
• Have to average particles/structures to improve the S/N therefore for non-symmetrical proteins/structures have to use tomography which is at a lower resolution

Question 12

How does Cryo-EM work

Answer 12

Sample is placed on a holey carbon support film and excess is blotted away. The sample is then plunged into liquid ethane cooled to near liquid nitrogen temperatures which causes the sample to be embedded in vitreous ice and can image this in the TEM. Rate of freezing is key for getting vitreous ice and no crystalline ice

Question 13

Describe the single-particle approach

Answer 13

Iterative refinement

Starting model –> reprojections –> aligned images (angle it was taken) –> intial map –> iterative refinment –> refined map.

See diagram

Question 14

Describe the contrast transfer function (CTF)

Answer 14

In Cryo-EM have very low contrast so have to introduce phase contrast. Using siemens start as a model as contains information at different resolutions, if you introduce variable amounts of defocus into an imaging system you see reversal of contrast. Get an oscillation of contrasts when you unfocus and you can do this in EM to get phase contrast as adding defocus forms a halo around the particles. You can computationally correct for this to get high-resolution structures - all structures have this.

Question 15

XRC - Zika example

Answer 15

Zika virus (ZIKV) has been associated with fetal microcephaly and Guillain-Barre syndrome.

Other mosquito-born flaviviruses, such as dengue virus, encode noncoding subgenomic flavivirus RNAs (sfRNAs) in their 3′ untranslated region that accumulate during infection and cause pathology.

ZIKV also produces sfRNAs that resist degradation by host exonucleases in infected cells. The authors solved the structure of one of ZIKV’s sfRNAs by x-ray crystallography and found that the multi-pseudoknot structure that it adopts underlies its exonuclease resistance

Question 16

What are the limitations of XRC?

Answer 16

• Large viruses are hard to crystallize
• Need pure samples which can be difficult for some viruses as all the same protein but the virus is not homogeneous so makes it hard to crystallize
• Need crystals - so may need to truncate or mutate certain regions which can affect capsid assembly
• Large viruses and crystal contacts can be a problem as end up with more water and little contacts which makes the crystal unstable

Question 17

FDMV virus example

Answer 17

Used a combination of XRC and Cryo-EM
Outbreak in 2001
Replication of the positive sense RNA genome via negative intermediates
The replication complexes involve RNA-dependent RNA polymerase (3Dpol - error prone which makes new copies of the genome). The structure of 3Dpol was solved by XRC and showed a general right-hand topology with fingers, thumb and a palm. Negative stain EM showed forms fibrils in the presence of RNA ~10-20nm in diameter and class averages from Cryo-EM showed helical fibres which the XRC structures of the polymerase can be fit into the 11A structure showing how it fits together.

Question 18

Principle of XRC

Answer 18

X-rays are short wavelengths
Take of the crystal in different orientations and spots appear, disappear and reappear - pattern is from constructive interference
You get an e- density map with you can feed in your sequence and measure how well your model fits the data
Isohedral is good for averaging as good symmetry
Crystals pack in a lattice so there are lots of ways they can pack together
Detects the x-rays on a CCD camera now not film ad loose the low-resolution data due to beam stop
Use a goniometer head to precisely rotate the crystal and the amount of data you need to collect depends on the packing of the crystal
Freezing the crystal can reduce the damage and increase the data quality

Question 19

What does the average temperature factor in XRC tell us?

Answer 19

How flexible parts of the molecule are

Question 20

Cryo-EM - Rotavirus example

Answer 20

• It is a bigger virus and was solved at a higher resolution
• dsRNA genome in 11 segments
• Characteristic triple layer structure
• Common cause of severe diarrhea in infants world wide
• Outer layer of triple layered particle is lost on entry leaving a double layer ~700A diameter
• T=13 quasi-icosahedral
• Using single-particle approach were able to determine the de novo structure of major coat protein which forms a spike - 4.1A structure - huge step forwards in resolution
• Solved a 4.1A structure however the density map is nicer than the 3.8A resolution XRC structure as experimental phase in EM is advantageous but the resolution is brought down by other areas of the structure being a lower resolution
• Did lots of averaging of rigid ordered particle to get the good resolution
• The surface consists of VP4 spike and VP7 glycoprotein and VP4 is the cell surface receptor binding proteins which have to undergo tryptic cleavage for efficient infectivity
• Cell penetration occurs via the endosomal compartment and VP4 conformation changes regulate this
• Cryo EM structure revealed N/C terminal ‘arms’ if disordered VP7 protein which XRC did not. VP7 was shown to make contacts with double layer particle and VP7 trimers

Question 21

Cryo EM - Aquareovirus example

Answer 21

Best structure until recently at 3.3 A
Similar to rotaviruses
Were able to solve the individual proteins from de novo
Showed outside chain densities except glycine
The structure and biochemical data revealed that priming involves autocleavage of membrane penetration protein and suggests lysine and glycine may facilitate this autocleavage by nucleophillic attack

Question 22

How to get atomic resolution of virus particles by Cryo EM

Answer 22

• Need the perfect particle
• 600-800A in size
• Need the best microscope and computing
• Biochemical stability needs to be right
• All particles need to be identical for averaging
• Can learn lots from perfect samples and apply this to those which aren’t
• Resolution of Cryo EM structure are increasing as better detectors which detect the e- directly so they are quicker and don’t get point spread - get less mechanical instablity and beam-induced movement as the camera are quicker and can do better movie correction and cherry pick the best particles.