A lentiviral system for efficient knockdown of proteins in neuronal cultures

2.1. Buffers and chemicals

water:              autoclaved double-deionized water or commercial ultrapure DNase/RNase-Free distilled water

2x annealing buffer:     20 mM Tris pH 7.8

                                        100 mM NaCl

                                        0.2 mM EDTA

1x TE buffer:    10 mM Tris pH 8.0

                           1 mM EDTA

spectinomycin:  set up 10 mg/ml stock in water and filter sterilize, aliquot and store at -20°C

ampicillin:         set up 100 mg/ml stock in water and filter sterilize, aliquot and store at -20°C

LB medium:     25 g/l pre-mixed LB medium

                          dissolve in double-deionized water and autoclave to sterilize

LB plates:       25 g/l LB medium and 15 g/l agar in double-deionized water

                         -    add a stir bar and autoclave to dissolve agar and sterilize the solution

                         -    cool down while spinning at low speed on a magnetic stir plate

                         -    once the solution has cooled down to touch, add sterile-filtered antibiotics, e.g. 50 μg/ml spectinomycin or 200 μg/ml ampicillin, before pouring plates

2M CaCl₂:      dissolve in double-deionized water and filter-sterilize, store at room temperature

0.1x TE:         1 mM Tris pH 8.0

                        0.1 mM EDTA

                        dissolve in double-deionized water, filter-sterilize, aliquot and store at -20°C

2x HBS:         dissolve 32.8 g NaCl, 0.42 g Na2HPO4, and 23.8 g HEPES in double-deionized water

                       -    adjust pH to 6.98 and bring to final volume of 2 l

                       -    set 500 ml aside and adjust the pH of the remaining solution to 7.00

                       -    set 500 ml aside and adjust the pH of the remaining solution to 7.02

                       -    set 500 ml aside and adjust the pH of the remaining solution to 7.04

                       -    filter-sterilize all four solutions, aliquot and store at -20°C

                      To identify the pH/solution with the highest transfection efficiency

                      -    plate 5 × 10⁶ HEK-293-T cells each in four 10 cm dishes

                      -    the next day, prepare four 1.7 ml microfuge tubes each with

                                    660 μl 0.1× TE,

                                    90 μl 2 M CaCl2,

                                    20 μg targeting vector (EmGFP or mRFP)

                                    10 μg pMDLg/pRRE,

                                    6 μg pMD2.g, and

                                    5 μg pRSV-Rev

                      -    mix in one of the four HBS solution into each tube, vigorously pipet up and down to mix

                      -    incubate 20 min at room temperature

                      -    add transfection solutions to cells and return cells to incubator

                      -    the next day, choose the HBS solutions with best transfection efficiency based on the percentage of fluorescent cells in the culture, fluorescence intensity and cell viability.

                          Note: cells and medium contain virus.

                          We usually find that one or two of the four HBS solutions perform significantly better than the others and we test every new batch of HBS solutions made. Only use HBS solutions with the highest transfection efficiency (at least 80+% of EmGFP- or mRFP-positive cells) for virus production (see section 6.1).

2.2. Oligos and miRNA design

2.2.1. Sequencing and PCR primers (in 5’ to 3’ direction)

pcDNA 6.2 antisense oligo              CCTCTAGATCAACCACTTTGT

miRNA-GFP sense BglII                GCGCAGATCTACCGGTCGCCACCATGGTGAGCAAGGGCGAGGAGC

miRNA-RFP sense BglII                GCGCAGATCTACCGGTCGCCACCATGGCCTCCTCCGAGGACGTC

miRNA-XFP antisense XhoI           GCGCCTCGAGTGCGGCCGGATCTGGGCCATTTGTTCCATGTGAGTGC

2.2.2. miRNA design

We use the ThermoFisher Scientific BLOCK-iT RNAi Designer (https://rnaidesigner.thermofisher.com/rnaiexpress/) to select target sequences. Make sure to select the miRNA tab from the Target Design Options (Step 2 on the web form). We usually design the target sequences within the open-reading frame (pre-selected option) and stay within the pre-selected 35–55% GC content (step 4 on the web form). On occasion, we have designed target sequences directed against the UTR’s and those resulted in efficient knockdown as well. Make sure to select the correct species such that the software automatically excludes sequences with predicted off-targets hits (step 3, see section 6.2).

We usually select four to five non-overlapping knockdown sequences for each target protein from the highest ranked sequences and try to cover a range of different GC content percentages. The complete miRNA oligo sequences, including linker, miRNA stem sequences and overhangs for cloning, are designed by the software. For each miRNA, two 64-nucleotide oligos are designed and standard purity oligos (desalted and lyophilized) can be ordered from a multitude of commercial providers.

3.1. Anneal oligos

3.2. Phosphorylate oligos

Make sure to store phosphorylated oligos on ice when in use or at -20°C for long-term storage.

3.3. Digest first vector

This step is best done in parallel with the oligo annealing/phosphorylation to optimize timing.

There are two versions of the first vector, pcDNA6.2/GW-EmGFP-miR and pcDNA6.2/GW-mRFP-miR, which only differ by the fluorescent marker being expressed as part of the shRNAmiR cassette.

Note: This digest will remove one PvuII restriction site, which is located between the two BsaI sites that flank the oligo insertion site.

3.4. Ligate phosphorylated miRNA oligos into BsaI-digested, dephosphorylated vector

3.5. Transformation into bacteria

3.6. Grow over-night cultures

We usually pick 3-4 individual colonies for each construct to have at least two positive clones (based on restriction digest in step 3.7)

3.7. DNA isolation and digest

3.8. Sequence verification

Sequence plasmids with the pcDNA 6.2 antisense oligo to confirm absence of mutations in the miRNA oligo insert and correct integration into the vector.

3.9. PCR amplification of the expression cassette

To transfer the shRNAmiR expression cassette into the vector for virus production, PCR amplify the expression cassette using the miRNA-XFP antisense XhoI oligo in combination with the miRNA-GFP sense BglII oligo (for pcDNA6.2/GW-EmGFP-miR constructs) or miRNA-RFP sense BglII (for pcDNA6.2/GW-mRFP-miR constructs).

                                                    $\begin{array}{c}\text{40 sec}\\ \text{40 sec}\\ \text{60 sec}\end{array}\begin{array}{c}\text{94°C}\\ \text{52°C}\\ \text{72°C}\end{array}]35x$

                                                    10 min      72°C

                                                    ∞             4°C

3.10. Digest PCR products

3.11. Digest virus expression vector pRRLsinPPTeGFP

This step is best done in parallel with the PCR product digest to optimize timing.

The digest removes the eGFP open reading frame from the vector.

3.12. Ligate PCR product into the digested pRRLsinPPT vector

3.13. Transformation into bacteria

3.14. Grow over-night cultures

We usually pick 4-6 individual colonies for each construct to have at least two positive clones (based on restriction digest in step 3.15)

3.15. DNA isolation and digest

3.16. Sequence verification

Sequence plasmids in both directions with the pRRL sense and pRRL antisense oligos to confirm absence of mutations and correct integration of the miRNA expression cassette into the vector. The final constructs are used to express the different miR targeting sequences.

4.1. Prepare high quality DNA preps

The use of high quality DNA preps is highly recommended. We use the Qiagen HiSpeed Plasmid Maxi Kit and usually purify 0.7-1.0 mg of DNA for each bacteria culture. DNA concentrations should be 0.5-2.0 μg/μl and OD_260/280 should be at least 1.8, but purities of 1.9 or higher are preferred. Prepare sufficient amounts of DNA for the scale and number of viruses being produced (see step 4.2), keeping in mind that the packaging mix will be combined with each targeting plasmid. Also, remember to include a non-targeting shRNAmiR control. We routinely use a control sequence that has been designed by Qiagen and which has no target in mammalian cells. However, the control shRNAmiR engages the same enzymes and processes that mediate RNAi upon expression of the shRNAmiR knockdown cassettes and thus controls for more variables than the empty target vector or a miR cassette lacking an shRNA target sequence.

The plasmids required are:    pRSV-Rev (packing mix)

                                                   pMD2.g (packaging mix, encodes VSV-G)

                                                   pMDL/pRRE (packaging mix)

                                                   pRRLsinPPT-EmGFP (or mRFP)-shRNAmiR-control

                                                   pRRLsinPPT-EmGFP (or mRFP)-shRNAmiR-targeting sequence

4.2. Prepare HEK-293-T cells and reagents

We maintain HEK-293-T cells in DMEM high glucose medium supplemented with 10% iron-supplemented calf serum at 37°C in an 5% CO₂-containing, water-saturated atmosphere. To improve cell survival during virus production, the culture medium can be supplemented with 1x non-essential amino acids and 1 mM sodium pyruvate (optional).

We usually plate six 15 cm dishes or T175 flasks with 15 × 10⁶ cells/plate for each virus. Since we often produce six viruses at a time, we grow up eight T175 flasks of HEK-293-T cells to confluency (approx. 80 × 10⁶ cells/plate) to have a large enough cell stock when plating cells for virus production. Also, make sure to have enough incubator space for all dishes/flasks available.

A prep of this size requires approx. 3 l of culture medium to expand the cell stock and for media changes during virus collection as detailed below. For each 15 cm plate/T175 dish, the following DNA amounts are needed:

                            50 μg expression plasmid (shRNAmiR-targeting sequence or control)

                            25 μg pMDLg/pRRE

                            15 μg pMD2.g

                            12.5 μg pRSV-Rev

In our hands, calcium phosphate-mediated transfection provides and efficient and cheap transfection method. For each 15 cm dish/T175 flask, the following amounts are needed:

                            1.875 ml 2x HBS

                            0.225 ml 2M CaCl₂

                            1.65 ml 0.1x TE

4.3. Virus production protocol and timeline

DAY 1: Plate HEK-293-T at a density of 1.2-1.5 × 10⁷ cells/T175 flasks in 25 ml medium. We use six T175 flasks (or 15 cm dishes) per virus and usually produce six viruses at a time (36 flasks/dishes total).

DAY 2: We prepare transfection master mixes for the plates/flasks transfected with the same miRNA construct. For each transfection, prepare 2 tubes (A and B). Volumes are for six dishes/flasks per virus and we use 50 ml conical tubes to set up the transfection mixes.

             In each A tube, add 11.25 ml 2x HBS

             In each B tube, add 9.9 ml 0.1x TE,

                            add DNA for the packaging mix and miRNA vector and mix by shaking,

                            add 1.35 ml 2M CaCl₂ and mix by shaking.

Mix tubes A and B using two pipet guns. Insert a small volume pipet into one pipet gun and continuously bubble the HBS buffer by pushing out air. Using the second gun, dropwise add the content of tube B to tube A, keep bubbling until the complete content of tube B has been added. Start a timer after completion of the first transfection mix. Continue mixing any remaining samples.

Allow for DNA/calcium phosphate precipitate to form for 20–25 min at room temperature, mix by shaking the tube and then add 3.75 ml transfection mix to each of the six 15 cm dishes/T175 flasks. Carefully swirl the plates to mix the transfection solution into the cell culture medium. Return the dishes to the incubator and note the time (time of transfection = 0 hrs).

8 hrs p.t. (post transfection), remove the medium from the flasks/dishes and add 15ml of fresh culture medium. This volume is just enough to cover the cells, ensure plates are level in the incubator to avoid cells loss due to cells falling dry.

DAY 3: Virus-producing cells release virus particles into the medium. Thus, from here on out, consider that cells and medium contain virus. Before collecting the first batch of virus-containing supernatant, check cells for expression of the fluorescent marker (EmGFP or mRFP) to ensure efficient transfection. Combine fluorescent and transmitted light to determine the percentage of transfected cells, the intensity of fluorescent marker expression, and changes in cell morphology (see Section 6.1). At least 80% of cells should express the fluorescent marker protein and the brightest cells should start to round up and may begin to detach. If the transfection efficiency is low, virus titers and purity will be too low to be used in most applications, in particular with primary neurons. The best approach is to start the virus production protocol again.

HEK-293-T cells attach rather weakly to the culture dishes/flasks and virus-producing cells will be in progressively bad shape over time. It is thus important to remove and replace media slowly and without releasing the liquid directly onto the cells. Pipet against the dish/flask side wall instead to limit premature cell loss and elevated levels of cell debris in the virus-containing supernatant.

24 hrs p.t., collect the supernatant from each plate, add 15 ml of fresh medium, and return plates to the incubator. If you have multiple plates for one construct, pool the supernatants and store them at 4°C. Since the medium collection containers need to be sterile and will have to be decontaminated or discarded after use, we collect and store the virus-containing medium in plug-cap T75 flasks (BD Falcon), which are large enough to combine the medium for each virus prep (three collections from six plates/flasks, approx. 270 ml), and discard the collection bottles as part of the solid virus-containing waste.

36 hrs p.t., repeat the collection step and return the plates to the incubator. For each construct, pool the supernatants with the ones from the 24 hr collection point and store at 4°C.

DAY 4: Collect the medium one last time 48 hrs p.t., pool with the earlier collections and discard the plates. Filter the supernatants through 0.45 μm PES filters to remove cell debris. We use disposable, sterile 500 ml-sized filter units consisting of bottle-top filters and receiver bottles.

After filtration, the medium can be aliquoted and stored at -80°C or the virus can be concentrated/purified by centrifugation before being stored at -80°C. Be sure that the tubes/bottles used are rated for the low temperature.

4.4. Virus concentration

We prepare primary hippocampal neurons from pre-natal embryos and have noticed that the calf-serum present in the virus-containing medium is toxic for our cultures. To remove the serum, we isolate the virus particles from the filtered medium by centrifugation and resuspend the virus-containing pellet in DMEM only.

4.5. Virus titration

Freeze/thaw cycles can result in loss of activity of lentiviral particles. We thus recommend to freeze any virus before titration such that all aliquots of a prep will give comparable transduction efficiencies.

Any given virus will transduce different cell lines with a cell line-specific efficiency, i.e. a titer established using one cell line will most likely not yield the same transduction efficiency on a different cell line. However, using a reference cell line for titration does allow to equalize the amount of virus needed for different viruses used in one experiment as well as across multiple virus preps. We use HEK-293-T cells to determine virus titers.

5.1. Transduction of primary hippocampal neurons

5.2. When to transduce primary hippocampal neurons

The best time to transduce primary neurons depends on the parameters of each experiment. For example, for neuronal development studies, neurons should be transduced as early as possible. We have successfully transduced neurons as early as day-in-vitro 1 (DIV 1). In contrasts, our attempts to transduce neurons during plating or immediately after the cells attached to the culture dish resulted in extensive cell death. For analysis of synapse development, we often transduce neurons between DIV 3 and DIV 7, and the time needed for complete turn-over of the target protein should guide timing.

We have also transduced neurons on DIV 14 and DIV 21, e.g. when testing for changes in synaptic function after synapses were allowed to form. We have not seen significant differences in overall transduction efficiency (i.e. percentage of transduced neurons in the culture) for the various transduction time point. However, the shRNAmiR cassette is currently expressed under the CMV promotor, the activity of which is partially suppressed in mature neurons. In case this decreases knockdown efficiency for certain targets, the design of the shRNAmiR cassette could be modified to include the CAG promoter instead.

5.3. When to analyze cells following transduction

The time needed to achieve efficient protein depletion has to be empirically established for each target proteins and depends on the turn-over of the endogenous proteins. In the vast majority of cases, we have achieved maximal knockdown within 2–5 days following transduction and timing appears to be independent of age of the neuronal culture.

The advantage of viral transduction is the stable integration of the shRNAmiR cassette into the cell genome such that no additional transductions are required for knockdown of targets with slow turn-over. For example, we routinely transduce neurons on DIV 4 for analysis of mature neurons on DIV 21 (Figure 1 and Figure 2).

5.4. How much virus to use for transduction

We base MOI on the titer determined by the titration of HEK-293-T cells.

This parameter also needs to be established empirically and we have achieved maximal knockdown with MOIs ranging from 2 to 20 for various target proteins and viruses. We do not use higher MOIs since we have noticed a decrease in cell viability with MOIs of 30 and higher.

5.5. Knockdown analysis

Given that viral transductions allows for the population-wide manipulation of neuronal cultures, we determine knockdown efficiency using protein-based read-outs, such as Western blot (Figure 1) and immunofluorescence imaging (Figure 2). Experimental details are included in Supplementary File 1.

6.1. Transfection efficiency for virus production

To produce virus, each HEK-293-T cells needs to be transfected with all four viruses required for virus production. The expression of the fluorescent marker protein (EmGFP or mRFP) encoded in the shRNAmiR cassette can be used to check transfection efficiency. At least 80% of all cells in the dish should be positive for the fluorescent marker 24 hrs post transfection, and transfection efficiencies of 90+% are desirable. A large population of brightly fluorescent cells that remain single cells and may begin to round up is an encouraging sign. In contrast, low numbers of fluorescent cells and/or low fluorescent intensity indicated suboptimal transfection. In addition, cell fusion indicates an over-abundance of VSV-G expression and a suboptimal plasmid ratio in the transfected cells. Any of these signs indicate sub-optimal virus production and may be cause to abandon this set of cells.

Note that checking for fluorescent protein expression at later time-points in the protocol will skew the result because virus secreted from producing cells can transduce other cells in the dish, thereby increasing the number of intensity of fluorescent cells over time.

Unfortunately, there are multiple reasons for sub-optimal virus production:

- DNA quality. Low DNA concentration or purity, or DNA degradation will impair transfection efficiency, e.g. the formation of white, cloudy precipitates during the 20 min incubation of the transfection solution indicates a contamination with proteins. If we run into any transfection issues, we generally switch to new DNA preps for all plasmids.

- Transfection solutions. Minute changes in pH of the 2x HBS solution has a major impact on transfection efficiency and efficiency decreases if the solution is stored at room temperature over extended periods of time. Alternative approaches are the use of other transfection reagents, e.g. polyethylenimine (PEI) or Lipofectamine 2000. For the latter, the dishes/flasks for the producer cells should be coated with poly-L-lysine to prevent excessive cell loss. While these other transfection reagents work well, we have had the highest transfection efficiencies using the calcium phosphate approach described here.

- Cell health. Contamination of cells, including mycoplasma, can have strong negative effects on virus production even though cells transfect efficiently. In addition, high passage cells eventually stop yielding high virus titers and should be replaced by lower passage cells. Note that the presence of the large T-antigen is essential for virus production and HEK-293 cells cannot replace HEK-293-T cells.

- Collection schedule. We collect media from the producer cells every 12 hrs and as a rough estimate, you will yield 25% of your total titer within the first 24 hrs, 50% between 24 and 36 hrs, and another 25% between 36 and 48 hrs. Some protocols call for virus collection up to 72 hrs but we only retrieve very minimal amounts of virus between 48–72 hours. Also, some protocols call for a single medium collection 48 or 72 hrs after transfection, however, this approach yields lower titers in our hands, likely due to the extended exposure of the viral particles to high temperatures in the incubators. In addition, decreases in medium pH, which can occur over longer times, can also decrease virus activity.

6.2. Oligo design

The BLOCK-iT RNAi Designer runs into problems when designing miRNA sequences directed against targets the sequences areas shared between splice variants. The work around is to omit the species selection in step 3. However, the target sequences predicted by the algorithm need to be manually checked for potential off-target sequences, e.g. by NCBI Blast against the respective species. Sequences with mismatches at the 3’ end, and if possible, with less than 12–14 continuously matched stretches of sequence should be selected.

Also, check if the oligos contain recognition sequences for any of the restriction enzymes used for cloning. If so, adjust expected band patterning accordingly.

6.3. Unwanted recombination during cloning

Use recombination-deficient strains for the transformation of ligations. If recombination persist, make sure not to go over time with the initial growth time during the transformation step. Lowering the temperature to 30°C during the transformation and to overnight growth at room temperature on the plates can help (growth phase may need to be extended to allow colonies to reach normal size).

6.4 Calculating virus titers

To calculate your titer in TU/μl (transforming units = active viral particle μl), you need to know

In this example, you would have transduced 14,000 cells (35% of 40,000) using 0.01953125 μl of your concentrated virus (with 500 μl/well total, at 1:200 equal 2.5 μl virus (well 1); 1: 400 is 1.25 μl (well 2), …).

Multiply the number of transduced cells (14,000) with the ratio of 1/0.01953125 = 51.2 (this is the factor to get to a full μl based on the dilution needed to transduce 14,000 cells). Therefore, the titer of this example virus is 716,800 TU/μl (or 720,000 TU/μl).