Purification and Proteomic Analysis of Alphavirus Particles from Sindbis Virus Grown in Mammalian and Insect Cells

[Abstract] Current mass spectrometry (MS) methods and new instrumentation now allow for more accurate identification of proteins in low abundance than previous protein fractionation and identification methods. It was of interest if this method could serve to define the virus proteome of a membrane-containing virus. To evaluate the efficacy of mass spec to determine the proteome of medically important viruses, Sindbis virus (SINV), the prototypical alphavirus was chosen for evaluation. This model system was chosen specifically because the alphaviruses contain members which are human pathogens, this virus is well defined biochemically and structurally, and grows to high titers in both vertebrate and non-vertebrate host cells. The SINV proteome was investigated using this method to determine if host proteins are specifically packaged into infectious virions. It was also of interest if the SINV proteome, when grown in multiple host cells representing vertebrate and mosquito hosts, incorporated specific host proteins from all hosts. Observation of recurrent or distinctive proteins in the virus proteome aided in the determination of proteins incorporated into the virion as opposed to those bound to the particle exterior. Mass spectrometry analysis identified the total protein content of purified virions within limits of detection. The most significant finding was that in addition to the host proteins, SINV non-structural protein 2 (nsP2) was detected within virions grown in all host cells examined. This analysis identified host factors not previously associated with alphavirus entry, replication, or egress, identifying at least one host factor integrally involved in alphavirus replication. Key to the success of this analysis is the method of virus purification which must deliver measurably infectious virus free of high levels of contaminants. For SINV and other members of the alphavirus family, this is accomplished by isopycnic centrifugation through potassium tartrate, followed by a high salt wash. Virus Proteome from and Invertebrate Hosts Identifies nsP2 as a Component of the Virion and Sorting Nexin 5


Stock Virus growth in BHK cells or C7-10 cells
The heat resistant SINV (SVHR) strain was used in this study. This strain was isolated by Burge and Pfefferkorn in 1966 by collecting virus that was resistant to heating to 54 °C. The choice of virus strain  The Authors; exclusive licensee Bio-protocol LLC. 4 www.bio-protocol.org/e3239 Bio-protocol 9(10): e3239. DOI: 10.21769/BioProtoc.3239 is important because this strain produces high titers (10 10 PFU/ml) and low particle/PFU (~1 particle/PFU), ratios of highly infectious and physically stable virus. BHK and C7-10 (Aedes albopictus) cells were obtained from internal collections and are the favored cells to grow SVHR. Virus was harvested from 10 T-75 flasks (Corning) which produces enough virus to form a large visible band in a 30 ml potassium tartrate gradient and sufficient material for the mass spec analysis. Cells were infected at an MOI of 10 PFU/ml, for Sindbis virus infections and allowed to replicate for a single cycle and harvested at 18 h post infection to ensure that no cell lysis took place. A single cycle of SINV growth from mammalian cells is ~12 h and from C7-10 cells is about 24 h. The supernatants were clarified by low speed centrifugation (Sorvall RC-5B Super speed centrifuge at 2,000 rpm, 700 x g). Twenty microliter of the resulting virus supernatant was loaded onto a 15-35% linear potassium tartrate gradient and twice purified by isopycnic ultracentrifugation (Beckman SW-28 rotor, 18 h at 10,000 x g). The resulting band of purified virus was collected and washed twice by pelleting the virus in 5 ml 1x PBS in an SW-40 Beckman-40 rotor at 45,000 rpm (12,000 x g) for 30 min and collecting the pellet.

Virus titration by Plaque assay
The assay of virus titer by plaque formation, "plaque assay" is the most accurate method for measuring of the amount of infectious virus. This assay is used to determine the titer, in plaque-forming units ( infecting flasks with dilutions of 10 -5 , 10 -6 , and 10 -7 PFU/ml should give adequate data to make a relatively accurate calculation. If the titer of the virus is completely unknown, it may be necessary to infect flasks or plates with a wide range of dilutions (10 -1 to 10 -8 ). The number and quality of the plaques seen in a given assay can be influenced by a number of factors, including the pH and/or temperature of media, dilution buffer, agarose overlay, or the condition of the cell monolayer. Due to the sensitivity of this assay, it is important to include both positive and negative controls within each assay. The negative control is a flask that is inoculated with diluent only, and the positive control consists of one dilution of SVHR stock virus of known titer sufficient to give ~20 plaques. This number of plaques is significant statistically.
It is important that when going from a high concentration to a low concentration the pipet tip is changed to avoid "carry over" contamination. However, when pipetting from a low concentration to a higher concentration, as is done when the wells are inoculated, a single pipet tip can be used.

Sub-culture of BHK-21 cells
Individual stocks of BHK cells may require different passage schedules and may have different levels of viable passages. It is good practice to keep track of the number of passages that an individual culture can be sub-cultured so that a schedule of cell thawing and storage can be established.
Fresh newly thawed cells can be passaged up to 30 times. If you are not going to split them immediately, cells can be incubated at 28 °C for 2 to 3 days but must be split at least once to recover normal growth before use in experiments.    9. Return flasks to the incubator set no higher than 26 °C. Mosquito cells will go into heat-shock at temperatures higher than 34°C.

Stock virus preparation
It is standard practice to grow a stock of virus from which additional virus stocks will be grown prior to any additional work with the virus. This practice avoids the production of defective interfering particles which will accumulate upon successive serial passage of high concentrations of virus.
Generally, an MOI (multiplicity of infection) of 0.01 PFU (plaque forming units) per cell is required for production of stock virus. To calculate the correct MOI, the number of cells to be infected must be known. Then multiply the number of cells by the required MOI, e.g., 10 6 total cells to be infected x MOI of 0.01 = 10 4 PFU of virus inoculum is needed.
If your stock virus is 1 x 10 9 PFU/ml and you need a 10 4 PFU inoculum, do the following: 1. Make a serial dilution of the stock virus, begin by making a 1:10 dilution of the stock virus to a total of 1 ml. This will be 100 µl of virus supernatant in 900 µl of PBS-D + 3% FBS. This is a 10 -1 virus dilution.
2. Make a serial dilution of this sample so that you get -1 through -4 dilutions in 5 separate tubes (see Figure 3). The amount of virus that you need is in the -4 dilution tube, 1 ml of virus of 10 -4 PFU/ml. Copyright   4. Refreeze your initial stock virus; this is your primary stock from which you will grow subsequent stocks. Virus is quantitated by plaque assay on the cells from which the virus produces CPE which is usually the cell line from which it is produced (in this case BHK). However Sindbis virus does not produce CPE in many insect cells (e.g., U4.4 of C7-10 cells), which cannot be used for the plaque assay.

Virus titration by plaque formation on BHK cells
Preparation of BHK cells www.bio-protocol.org/e3239    the agarose was too hot when it was added. Agarose that is too hot will also kill the cells.

Note: Virus frozen and thawed will lose 1/3 to 1/5 log of titer. This is not a significant loss for use to infect cells. For preservation of the highest infectivity level, virus can be stored at 4 °C for up to 5 days in neutral pH buffer. This method preserves the most infectivity and is used for virus
to be analyzed in structural studies and for the calculation of particle to PFU ratios. The virus supernatant is harvested into conical tubes and cell debris is removed from the supernatant by low-speed centrifugation. This is a critical step and if it is omitted, too much debris will interfere with the following gradient purification steps. Clarify virus supernatant by centrifugation in the appropriate size conical tube for 10 min at 1,800 x g.

Purification and concentration by isopycnic centrifugation
1. Virus supernatant is combined into an appropriate size container omitting addition of the glycerol.
Note: From this step forward, you should not consider your samples sterile. This is not usually a problem and antibiotic is added to the 2x E-MEM media.
2. Determine the number of gradients which will be needed.
Typically, 20 ml of virus is layered onto the initial potassium tartrate step gradient in the 38 ml tubes. This gradient is formed by carefully layering 12 ml of 15% potassium tartrate onto a layer of 6 ml of 35% potassium tartrate, or making a 10% to 35% continuous potassium tartrate  Note: Properly grown and purified SVHR should give a particle/PFU value of ~1.

Concentration of virus by Polyethylene glycol (PEG) precipitation
Note: This is an alternative purification step used to concentrate the virus particles but infectivity will be lost.
2. Additionally, to check for co-purified protein contaminants, check protein content by running ~10 μl of virus on a 4-12% Bis-Tris SDS-PAGE gradient gel (Invitrogen, Novex) as described by the manufacturer.
3. Stain the gel with silver in the method of Wray or with coomassie blue (Wray et al., 1981). Excise the visualized bands (silver stain only) and perform an in-gel digestion prior to LC-MS/MS analysis for protein identification (Glaros et al., 2015).

Protein extraction and digestion
Viral preparations and their respective negative controls, medium from the cell monolayers 5. Trypsin is specified as the protease with a maximum number of missed cleavages set to 2.
6. False discovery rate using PERCOLATOR (Kall et al., 2007) is set to < 1% to score high confidence peptide identifications.
7. Perform grouping and functional analysis using the PANTHER classification system for the human background only with each protein id's accession number (Mi et al., 2016).

Complete E-MEM, 1x and 2x
Note  ii. Leftover 1 M DTT stock can be stored at -20 °C for up to 30 days (it is best make 50-100 μl aliquots to prevent any freeze/thaw cycles). 16. UA buffer (8 M urea, 100 mM Tris-HCl pH 8.5) (10 ml) Note: You will need around 500 μl of UA buffer per sample. Note: this will be an excess of solution, but it is a more manageable to weigh out NaCl for this volume).
Weigh out 146.1 mg NaCl (Sigma) and add mass spec grade water to 5 ml