Is VP22 nuclear homing an artifact?

To the editor:

The herpes simplex virus protein VP22 is secreted from cells and imported into the nuclei of adjacent cells1. The unique ability of VP22 and VP22 fusion proteins to enter cells makes it a promising tool for gene and protein delivery1,2,3,4,5,6,7,8. The mechanism mediating import of VP22 is unknown. However, it is not dependent on classical endocytosis or energy, as import and nuclear localization of VP22 occurs at 4°C1.

We have studied the import of an N-terminal truncated VP22 protein, containing amino acids 159–301, that can be expressed in Escherichia coli and retain the ability of cellular import9. A commercially available VP22 expression vector (pCRT7/VP22, Voyager TM Protein Production Kit, Invitrogen) was used to express VP22 with a C-terminal Myc-epitope and a poly-histidine tag. Recombinant VP22, purified by metal affinity chromatography, was added to Chinese hamster ovary cells (CHO) cells for 20 min at 37°C. The cells were washed, fixed with methanol and stained with a FITC-conjugated anti-Myc antibody.

Similar to previous reports, virtually every cell in the culture incubated with VP22 exhibited bright nuclear fluorescence (Fig. 1). We incubated cells with VP22 at 4°C for 10 s, 2 min and 20 min to study the kinetics of VP22 import. VP22 was detectable in cell nuclei already after 10 s incubation (Fig. 1). The very rapid nuclear import of VP22 at 4 °C prompted us to investigate an alternative mechanism for this phenomenon, rather than import of VP22 into living cells.

Figure 1

CHO cells incubated with Myc-tagged recombinant VP22 (1 μg) and immunostained with a fluorescein isothiocyanate (FITC)-conjugated anti-Myc antibody.

CHO cells incubated with VP22 at 37°C for 20 min or 4°C for 10 s, 2 min, or 20 min. The cells were washed in phosphate buffered saline (PBS), methanol fixed for 15 min at room temperature, blocked with 10 % fetal calf serum in PBS and stained for 1 h with a FITC-conjugated anti-Myc antibody (Invitrogen). Methanol (MeOH)-fixed cells were incubated with VP22 for 2 min before washing and staining. Cell nuclei were contrastained with DAPI (4′,6 diamidino-2-phenylindole).

Methanol fixation, used for VP22 import studies, results in permeabilization of the cell membrane. We added VP22 to methanol fixed cells and showed that the protein localizes to cell nuclei of the fixed cells (Fig. 1). These findings suggest that import and nuclear localization of VP22 may occur during fixation. VP22 binds to the cell membrane1 and we confirmed this finding by immunostaining the living unfixed cells incubated with VP22 (data not shown). Cell-surface binding of VP22 will allow the protein to remain attached to cells during washing before fixation and it may a constitute a “VP22 reservoir” that is released by methanol fixation.

To directly visualize the location of VP22 in cells, we expressed a fusion protein of VP22 and the green fluorescent protein (GFP) in E. coli. Cells incubated with purified VP22–GFP at 37°C for 20 min showed fluorescence at the cell membrane, but no fluorescence in the nucleus (Fig. 2). We were not able to detect nuclear GFP fluorescence in cells incubated with VP22–GFP for up to 24 h (data not shown). However, directly after methanol fixation and rehydration in PBS, the cells exhibited nuclear GFP fluorescence.

Figure 2

CHO cells incubated with VP22-GFP (1 μg) for 20 min at 37 °C.

GFP fluorescence was visualized in live unfixed cells or after methanol fixation.

These findings show that import and nuclear localization of VP22 occur during fixation. Artificial nuclear localization of soluble proteins during fixation for immunocytochemistry is well documented10. We believe that the VP22 intercellular transport is due to this phenomenon. This conclusion is supported by studies showing that VP22–GFP intercellular trafficking only can be detected after fixation and not in living cells2,3,4,5. Although cells adjacent to cells expressing VP22 fused to p53 or thymidine kinase are reported to show increased cell death, there is no experimental evidence that this effect is due to delivery of functional p53 or thymidine kinase into the adjacent cells7,8. Recently, cell death was not observed in cell cultures incubated with VP22 fused to the diphtheria toxin A-fragment, although import of only a few toxin molecules are sufficient to cause cell death11.

Considering the artificial import and nuclear localization of VP22 during fixation, we find little evidence in the published literature that supports import and nuclear localization of VP22 in living cells.

See Reply to 'Is VP22 nuclear homing an artifact?' by Elliott et al.