A new generation of pPRIG-based retroviral vectors

Abstract

Background: Retroviral vectors are valuable tools for gene transfer. Particularly convenient are IRES-containing retroviral vectors expressing both the protein of interest and a marker protein from a single bicistronic mRNA. This coupled expression increases the relevance of tracking and/or selection of transduced cells based on the detection of a marker protein. pAP2 is a retroviral vector containing eGFP downstream of a modified IRES element of EMCV origin, and a CMV enhancer-promoter instead of the U3 region of the 5'LTR, which increases its efficiency in transient transfection. However, pAP2 contains a limited multicloning site (MCS) and shows weak eGFP expression, which previously led us to engineer an improved version, termed pPRIG, harboring: i) the wild-type ECMV IRES sequence, thereby restoring its full activity; ii) an optimized MCS flanked by T7 and SP6 sequences; and iii) a HA tag encoding sequence 5' of the MCS (pPRIG HAa/b/c).

Results: The convenience of pPRIG makes it a good basic vector to generate additional derivatives for an extended range of use. Here we present several novel pPRIG-based vectors (collectively referred to as PRIGs) in which : i) the HA tag sequence was inserted in the three reading frames 3' of the MCS (3'HA PRIGs); ii) a functional domain (ER, VP16 or KRAB) was inserted either 5' or 3' of the MCS (<< modular >> PRIGs); iii) eGFP was replaced by either eCFP, eYFP, mCherry or puro-R (<< single color/resistance >> PRIGs); and iv) mCherry, eYFP or eGFP was inserted 5' of the MCS of the IRES-eGFP, IRES-eCFP or IRES-Puro-R containing PRIGs, respectively (<< dual color/selection >> PRIGs). Additionally, some of these PRIGs were also constructed in a pMigR MSCV background which has been widely used in pluripotent cells.

Conclusion: These novel vectors allow for straightforward detection of any expressed protein (3'HA PRIGs), for functional studies of chimeric proteins (<< modular >> PRIGs), for multiple transductions and fluorescence analyses of transduced cells (<< single color/resistance >> PRIGs), or for quantitative detection of studied proteins in independently identified/selected transduced cells (<< dual color/selection >> PRIGs). They maintain the original advantages of pPRIG and provide suitable tools for either transient or stable expression and functional studies in a large range of experimental settings.

Figure 1

Schematic representation of the pAP2 and pPRIG vectors. The parental pAP2 retroviral vector harbors the CMV enhancer/promoter (CMVpro) in place of the U3 region of the 5' LTR and eGFP downstream of an ECMV-derived IRES containing two point mutations (hence indicated as IRES*) which strongly weakens its activity. The pPRIG HAa/b/c vectors improve pAP2 by: 1) replacing IRES* by the completely wild-type ECMV-derived IRES sequence (IRES), which restores its full activity; 2) inserting a much more complete multicloning site (MCS) and T7 and SP6 phage promoter on each side of the new MCS; and 3) adding a HA tag sequence in the three reading frames (a/b/c) 5' of the MCS. pPRIG HAa/b/c, as well as a pPRIG (without HA), were previously described and are used here to generate new vectors. For each vector, the unique sites of the MCS are indicated above the MCS.

Figure 2

Schematic representation of the HA3' PRIG vectors. pPRIGp a/b/c HA are three pPRIG derivatives containing a HA tag sequence downstream of the MCS. Each of these derivatives harbors the HA sequence in one of the three reading frames relative to the MCS (hence pPRIG a/b/cHA). The reading frame is therefore open from and throughout the MCS to the stop codon located at the 3' end of the HA sequence. Moreover, the IRES-eGFP cassette is flanked by two PacI sites (Pa) a very rare cutter, hence pPRIGpa/b/c HA, allowing for accurate removal of the cassette. For each vector, the unique sites of the MCS are indicated above the MCS.

Figure 3

Schematic representation of the « modular » PRIG vectors. pPRIGp ER is a pPRIG derivative containing the C-terminal portion of the mouse estrogen receptor bearing the G525V mutation. This mutant does not bind estradiol but still binds the synthetic ligand hydroxytamoxifene. The ER sequence has been cloned at the 3' end of the MCS and the reading frame is thus open from and throughout the MCS to the stop codon ending the ER sequence. In contrast, in pPRIGp VP16HA and pPRIGp KRABHA, the sequence coding the functional module (VP16 transactivating domain from human herpes virus or KRAB transrepressing domain from human KOX1 protein) was cloned upstream of the MCS. The reading frame is thus open from the start codon of the module to the stop codon of the in-frame HA sequence 3' of the MCS. Pa: site for PacI. The two PacI sites of pPRIGp VP16HA and pPRIGp KRABHA are symbolized by vertical bars. For each vector, the unique sites of the MCS are indicated above the MCS.

Figure 4

Schematic representation of the « single color/resistance » PRIG vectors. pPRICp aHA, pPRIYp aHA and pPRIChp aHA and pPRIPu vectors are pPRIG derivatives in which the eGFP sequence from the pPRIG has been replaced by eCFP*, eYFP, mCherry or Puro-R coding sequence, respectively. eCFP* is a derivative of eCFP bearing the H148D mutation which enhances its brightness. pPRICp aHA, pPRIYp aHA and pPRIChp aHA contain an in-frame 3' HA sequence (frame « a ») with respect to the MCS. They also contain the two PacI sites flanking the IRES-XFP cassette (shown as vertical bars in pPRIYp aHA and pPRIChp aHA). By contrast, pPRIPu is devoid of both HA sequence and PacI sites. For each vector, the unique sites of the MCS are indicated above the MCS. Unique sites of the pPRIYp aHA MCS are not listed since they are the same ones as in the immediately above vector.

Figure 5

Schematic representation of the « dual color/selection » PRIG vectors. pPRICp eYFPHA is a pPRICp aHA derivative in which the eYFP sequence has been cloned upstream of the MCS. pPRIGp mChHA is a pPRIGp aHA derivative in which mCherry sequence has been cloned upstream of the MCS. pPRIPu eGFP is a pPRIPu derivative in which the eGFP sequence has been cloned upstream of the MCS. The reading frame is open from the start codon of the upstream fluorescent protein to the stop codon of the HA sequence (for pPRICp eYFPHA and pPRIGp mChHA) or to the in-frame stop codon 3' to the MCS (for pPRIPu eGFP). pPRICp eYFPHA and pPRIGp mChHA, but not pPRIPu eGFP, harbor the two flanking PacI sites. For each vector, the unique sites of the MCS are indicated above the MCS.

Figure 6

FACS analyses of NIH3T3 cells transduced with different PRIG vectors. Cells were analysed using a LSRII cytometer (Becton-Dickinson) 5 to 7 days after exposure to the viral supernatant(s). For each point, 10⁴ events were recorded. A) Non-transduced NIH3T3 cells we analysed for cyan (eCFP*) yellow/green (eYFP/eGFP) or red (Cherry) fluorescences using the lasers and filters described in the Methods section to determine the basal level for each signal. B) Functional test of the pPRIPu eGFP vector. 48 hours after the exposure to a pPRIPu eGFP viral supernatant, cells were plated in a fresh medium containing (right panel) or not (left panel) 10 μg/ml of puromycin, and analysed for eGFP expression after 7 days of culture. All the selected and almost all the non-selected cells strongly express eGFP. Note that at this dose of puromycine, control untransduced NIH3T3 cells died in approximately 24 hours (not shown). C) Functional test of the single color vectors. Cells transduced with the indicated vector (at the bottom of each panel) were analyzed for each fluorescence. Note that under these conditions of transduction and detection, each fluorescent protein (eCFP*, eYFP/eGFP, mCherry) is mainly, if not exclusively, detected in its appropriate channel. D) Feasability of double transductions. Cells were transduced with the indicated (right hand side) mixture of two viral supernatants. In the cases of double transductions, untransduced cells, doubly transduced cells and cells transduced with either one of the two single color vectors can be discriminated from each other. E) Feasability of triple transductions. Cells were transduced with the indicated mixture of the three viral supernatants. Again, because of the minimal overlapping between the three signals, the transduction status of any cell with respect to each vector can be determined. Note however, that for D and E, the quantity of each viral supernatant was not equal in the mixture to correct for either relatively inefficient detection of the signal in our conditions (mCherry) or relatively intrinsic weakness of the protein (eCFP*) (see text). F) Functional test of the doublecolor vectors. Cells were transduced with the indicated vector (at the bottom of the panel). In each case, the vast majority of the cells are positive for the two expected signals.

Figure 7

Fluorescence microscopy analyses of cells transfected or transduced with PRIG vectors. A. pPRIPu confers puro-resistance. HEK293T cells were transfected with either peGFP-N1 + pPRIPu (left) or peGFP-N1 (Clontech) + pRK5 (irrelevant carrier DNA) (right). 24 hours after transfections, cells were further grown for 30 hours in the presence of 10 μg/ml of puromycin. eGFP serves as a control for transfection efficiency. pPRIPu + peGFP-N1 transfected cells are living while most mock-transfected cells are dying. B. Single color PRIGs are expressed in HEK293T transfected cells. HEK293T cells were transfected with the indicated single color PRIG vector. Fluorescence microscopic analyses were performed 24 hours later. The signal emitted by each fluorescent protein was observed and photographed using the appropriate filter. Note that the mCherry signal is relatively weak probably due to the suboptimal excitation light delivered by our filter (546/12 nm). Moreover, a weak portion of the eGFP signal is detectable in the cyan filter, and conversely a residual portion of the eCFP* signal is detectable in the green filter. By contrast, the eYFP (green filter) and mCherry (red filter) signals are only detected in their appropriate filter. C. Single color PRIGs are expressed in transduced rat primary osteoblasts. Rat primary osteoblasts were transduced by the indicated single color PRIG vector and observed using a fluorescent microscope 48 hours later. Signals were photographed using the appropriate filter for each fluorescent protein. D. mCherry localization in HEK293T transfected cells. Left: HEK293T cells were transfected with pPRIGp mChHA. The red signal appears to be predominantly cytoplasmic, which is not observed in pPRIChp transfected cells, where mCherry shows an uniform distribution (see B). Right: HEK293T cells were transfected with a pPRIGp mChHA derivative containing an in-frame deletion of the 5' part of the MCS (PvuII/StuI). This deletion completely restores a normal (uniform) distribution of mCherry. Note that in cells transfected with pPRIGp mChHA (left) and with its PvuII/StuI deleted derivative (right), the eGFP (green, encoded by the 3' cistron) appears normally and uniformly distributed throughout the cell. The regions of interest were enlarged for a better visualization.