Rab14 regulates the transport of human papillomavirus to the trans‐Golgi network for infectious cell entry

Abstract

Human papillomavirus (HPV) utilizes multiple host proteins to infect epithelial cells. Here we identified Rab14, a small GTPase belonging to the Rab family, as a critical host factor in HPV infection. Rab14 knockout (KO) cells exhibited resistance to HPV pseudovirion (PsV) infection. Confocal microscopy revealed that PsV transport to the trans‐Golgi network (TGN) was significantly reduced in Rab14 KO cells. Reintroduction of wild‐type Rab14 restored infection levels, but neither GDP‐bound (S25N) nor GTP‐bound (Q70L) mutants did. S25N preferentially localized to the TGN, while Q70L exhibited similar localization in the early endosome and the TGN. PsV showed reduced colocalization with S25N, whereas the opposite was observed for Q70L. These results indicate that Rab14 GTPase cycling controls HPV transport to the TGN.

This study reveals that the small GTPase Rab14 is necessary for human papillomavirus (HPV) infection and plays an essential role in the transport of virions to the trans‐Golgi network (TGN). HPV in the early endosome (EE), which harbors GTP‐bound Rab14, is transported to the TGN through the switch of Rab14 from its GTP‐bound to GDP‐bound form.

Abbreviations

Cas9, CRISPR‐associated protein 9

CMV, cytomegalovirus

CRISPR, clustered regularly interspaced short palindromic repeat

EdU, 5‐ethynyl‐2′‐deoxyuridine

EEA1, early endosome antigen 1

HA, hemagglutinin

HPV, human papillomavirus

HSPGs, heparan sulfate proteoglycans

KO, knockout

LAMP1, lysosomal‐associated membrane protein 1

luc2, firefly luciferase gene 2

PBS, phosphate‐buffered saline

PsV, pseudovirion

rAAV2, recombinant adeno‐associated virus serotype 2

rLV, recombinant lentivirus

SDS, sodium dodecyl sulfate

sg‐RNA, single‐guide RNA

TGN, trans‐Golgi network

TGN46, trans‐Golgi network glycoprotein 46

Human papillomavirus (HPV), a nonenveloped double‐stranded DNA virus, is the causative agent for cervical cancer [1]. HPV infects epithelial basal cells through microwounded regions and establishes its persistent infection [2]. Initially, HPV virions, which are composed of major capsid L1 and minor capsid L2, bind to heparan sulfate proteoglycans (HSPGs) on the cell surface [3, 4, 5, 6]. The attached virus is supposed to enter the cell through the action of heparinase and metalloproteases, such as metallopeptidase domain 17, from an entry platform containing α6β4‐integrin, annexin A2, epidermal growth factor receptor, and tetraspanins [7, 8, 9, 10, 11, 12]. The binding to HSPGs can induce structural changes in the L1 capsid, facilitating the cleavage at the carboxyl terminus of L1 by kallikrein‐related peptidase eight [13]. Although not yet fully proven, this cleavage may cause the exposure of L2 from the L1 capsid, followed by L2 cleavage by furin [14]. Subsequently, the virions are transported to the endosomal compartment, where L2 penetrates the lipid bilayer of the endosome by γ‐secretase [15]. This process is necessary for transporting the virus to the trans‐Golgi network (TGN) [16] and to the nucleus [17].

Small GTPases belonging to the Rab family regulate intracellular membrane trafficking by coordinating the biogenesis, transport, anchoring, and fusion of organelles and vesicles [18]. Rabs cycle between two states: a GDP‐bound (inactive) form and a GTP‐bound (active) form. The active form localizes to specific organelles or vesicles and recruits various effector proteins to execute each step of membrane trafficking via their GTPase cycles. Regarding the HPV life cycle, Rab5, Rab6a, Rab7b, and Rab9a contribute to the infectious entry pathways of virions [19, 20, 21, 22, 23, 24].

In a previous study, we aimed to identify host genes required for HPV16 infection by performing a genome‐wide CRISPR‐Cas9 knockout screen in HeLa cells. The results identified RAB1A/B, RAB6C, and RAB14 as candidate RAB genes involved in HPV cell entry, among which RAB14 ranked third highest in the screening [25]. Rab14 regulates traffic between the Golgi apparatus and early endosomes, as well as recycling endosomes and the plasma membrane [26]. Regarding virus infection, Rab14 is involved in the endocytosis of the Ebola virus matrix protein, viral protein 40 [27]. Additionally, Rab14 regulates ceramide trafficking to the Golgi apparatus to facilitate classical swine fever virus assembly [28] and is involved in the outward sorting of the envelope glycoprotein complex of human immunodeficiency virus‐1 [29]. In this study, we focus on Rab14 and report its role and mechanism in HPV infectious cell entry.

Materials and methods

Cell culture

HeLa‐mCAT#8, a derivative of HeLa ATCC CCL‐2 (RRID: CVCL_0030), which stably expresses the mouse ecotropic retroviral receptor mCAT1 [30], was a gift from Dr. T. Yamaji. HaCaT (RRID: CVCL_0038) was a gift from Dr. Fusenig, N. E. (German Cancer Research Center, Heidelberg). Lenti‐X 293 T (RRID: CVCL_4401) was purchased from Takara Bio (Shiga, Japan, #632180), and 293FT (RRID: CVCL_6911) was purchased from Thermo Fisher Scientific (Waltham, MA, #R70007). All cell lines used in this study were authenticated by the suppliers and cultured in Dulbecco's modified Eagle's medium containing 10% fetal bovine serum. All experiments were performed with mycoplasma‐free cells.

Plasmids

The CRISPR plasmid and the lentivirus plasmid expressing a Rab14 guide sequence (pSELECT‐CRISPR‐Cas9 and lentiCRISPRv2) were constructed by insertion of the Rab14 single‐guide RNA (sgRNA) sequence, 5′‐CGG CAG GAC AGG AGC GAT TT, into pSELECT‐CRISPR‐Cas9 and lentiCRISPRv2, respectively.

The Rab14 expression plasmid for lentivirus generation (pLVSIN‐EF1α Neo HA‐Rab14) was constructed as follows. First, the synthetic oligonucleotide containing an EcoRI site, Kozak sequence, HA‐tag sequence, Gly/Ser linkers, and a NotI site was inserted into pLVSIN‐CMV Neo (Takara Bio, Shiga, Japan) between the EcoRI and NotI sites to create pLVSIN‐CMV Neo HA‐N. The sequence of the synthetic oligonucleotides was as follows: sense sequence, 5′‐AAT TCG CCA CCA TGT ACC CAT ACG ATG TTC CAG ATT ACG CTC TCG AGG GCG GAG GAG GCT CTG GAG GCG GCG GAT CTG GAG GAG GCG GCT CCG C; antisense, 5′‐GGC CGC GGA GCC GCC TCC TCC AGA TCC GCC GCC TCC AGA GCC TCC TCC GCC CTC GAG AGC GTA ATC TGG AAC ATC GTA TGG GTA CAT GGT GGC G. Next, the RAB14 cDNA was amplified from the RAB14 gene (Clone ID: OHu12700, GenScript, Nanjing, China) by PCR. The sequences of PCR primers were as follows: forward primer, 5′‐TTT TGC GGC CGC TAT GGC AAC TGC ACC ATA CAA CTA C; reverse primer, 5′‐AAA AGC GGC CGC TAG CAG CCA CAG CCT TCT CTC. The amplified DNA was inserted into pLVSIN‐CMV Neo HA‐N at a NotI site to create pLVSIN‐CMV Neo HA‐Rab14. Finally, the HA‐tag‐fused RAB14 gene was amplified from pLVSIN‐CMV Neo HA‐Rab14 by PCR and inserted into pLVSIN‐EF1α Neo (Takara Bio) at XbaI and BamHI sites using the In‐Fusion HD cloning kit (Takara Bio). The primers used for amplification of the HA‐tag fused RAB14 gene were as follows: forward primer 5′‐GGA TTT AAA TTC TAG AGC CAC CAT GTA CCC ATA CG, and reverse primer 5′‐CGG TAG AAT TGG ATC CGC GGC CGC TAG CAG CCA C.

The expression plasmids for Rab14 mutants, Q70L and S25N, were generated from pLVSIN‐EF1α Neo HA‐Rab14 by site‐directed mutagenesis using the KOD‐Plus‐Mutagenesis Kit (TOYOBO, Osaka, Japan). The primers used were as follows: Q70L forward primer, 5′‐TGG AGC GAT TTA GGG CTG TTA CAC G; Q70L reverse primer, 5′‐GTC CTG CCG TAT CCC AAA TCT GCA G; S25N forward primer, 5′‐AAT TGC TTG CTT CAT CAA TTT ACA G; S25N reverse primer, 5′‐TTT TCC TAC TCC CAT GTC CC.

Cloning of Rab14 KO cells

HeLa‐mCAT#8 cells were transfected with the CRISPR plasmid expressing Rab14 sgRNA and screened by puromycin treatment, as previously described [25]. Rab14 KO HaCaT cells were prepared by transduction with the lentivirus expressing Rab14 sgRNA, as previously described [25]. Cells were diluted to less than one cell per well in a 96‐well plate, and the cells that continued to proliferate were isolated as Rab14 KO clones. The disruption of the RAB14 gene was evaluated by sequence analysis of the RAB14 exon 4 and western blotting.

Pseudovirions (PsVs)

HPV16 L1/L2 virion encapsulating the reporter plasmid pCMV‐Luc2 (16PsV) and HPV18 L1/L2 virion encapsulating pCMV‐Luc2 (18PsV) were prepared as follows [25, 31]. Briefly, 293FT cells were transfected with the expression plasmid for HPV16L1 or HPV18L1, the expression plasmid for HPV16L2 or HPV18L2, and pCMV‐Luc2. Two days after transfection, the cells were lysed in phosphate‐buffered saline (PBS) containing 0.35% Brij‐58, 10 mM MgCl2, and 0.9 mM CaCl2 and incubated with benzonase at 37°C for 24 h. The cell lysate was placed on an OptiPrep™ gradient and ultra‐centrifuged with a Beckman SV55Ti rotor. A fraction containing L1, L2, and the reporter plasmid was collected and used as a stock of 16PsV or 18PsV. 16PsV encapsulating pCMV‐Luc2 labeled with 5′‐ethynyl‐2′‐deoxyuridine (EdU) was generated in 293FT cells grown in medium containing EdU [32].

Recombinant viruses

Lentivirus (rLV) expressing Rab14 sgRNA, HA‐tagged Rab14 wild‐type and mutants (Q70L and S25N), and firefly luciferase gene were prepared using the lentiviral high‐titer packaging mix (Takara). Recombinant AAV2 (rAAV2) expressing the firefly luciferase gene was prepared using the AAVpro helper‐free system (Takara).

Reintroduction of Rab14 into Rab14 KO cells

Rab14 KO HeLa cells that were reintroduced with wild‐type and mutant Rab14 were generated by transduction with lentiviruses expressing HA‐Rab14 WT, Q70L, and S25N, as previously described [25].

Western blot analysis

Western blot analysis was performed with rabbit polyclonal anti‐Rab14 antibody (15662‐1‐AP, Proteintech, Rosemont, IL), mouse anti‐HA (clone 16B12, BioLegend, San Diego, CA), and mouse anti‐α‐tubulin (66031‐1‐Ig, Proteintech). Briefly, cells were lysed in an SDS sample buffer containing 1% 2‐mercaptoethanol at 98 °C for 10 min. Proteins in the cell lysate were separated by SDS/polyacrylamide gel electrophoresis and transferred to a polyvinylidene difluoride membrane. After blocking with 10% skimmed milk in PBS, the membrane was incubated with the rabbit anti‐Rab14 antibody diluted at 1:1000, the mouse anti‐HA antibody diluted at 1:1000, or the mouse anti‐α‐tubulin antibody diluted at 1:10000. The membrane was then washed with PBS containing 0.05% Tween 20 and incubated with goat anti‐rabbit or anti‐mouse IgG conjugated with horseradish peroxidase (Santa Cruz Biotechnology, Santa Cruz, CA, USA) diluted at 1:5000. Immunoreactive proteins were detected using the ECL™ Prime Western Blotting Detection Reagent (Cytiva, Marlborough, MA, USA) on a ChemiDoc Touch (Bio‐Rad, Hercules, CA, USA).

Infection assay

Cells were seeded into 96‐well culture dishes (0.5 × 10⁴ cells per well) and infected with PsV, rLV, and rAAV2 at a concentration of 161 vector genomes (vg) per cell, 10 vg per cell, and 10 vg per cell, respectively. After incubation for two days, the luciferase activity in the cell lysate was measured using the luciferase assay system (Promega, Madison, WI, USA) on a Mithras LB940 plate reader (Berthold Technologies, Bad Wildbad, Germany). The activity was normalized to the cell viability, which was estimated using the CellTiter 96 nonradioactive cell proliferation assay (Promega). The experiment was performed with three biological replicates.

Immunofluorescence microscopy

Cells were incubated in an eight‐well chamber slide (1 × 10⁴ cells per well) with EdU‐labeled 16PsV (0.2 μg of L1 per well) at 4 °C for 1 h. The cells were then washed in the growth medium and incubated at 37 °C for 2, 8, and 16 h. Following incubation, the cells were fixed with 4% paraformaldehyde and permeabilized with 0.5% Triton X‐100. EdU was labeled with Alexa Fluor 488 azide through a Click‐iT reaction (Thermo Fisher Scientific, Waltham, MA). HA and organelle marker proteins, EEA1, TGN46, and LAMP1, were labeled with specific antibodies as follows: HA, mouse anti‐HA (clone 16B12, BioLegend, San Diego, CA); EEA1, rabbit monoclonal anti‐EEA1 antibody (clone C45B10, Cell Signaling Technology, Danvers, MA, USA); TGN46, rabbit polyclonal anti‐TGN46 antibody (ab50595, Abcam, Cambridge, UK); LAMP1, rabbit monoclonal anti‐LAMP1 antibody (D2D11XP, Cell Signaling Technology), at a dilution of 1:100. These antibodies were labeled with Alexa Fluor 546 goat anti‐rabbit IgG (H + L) (Thermo Fisher Scientific), Alexa Fluor 555 goat anti‐mouse IgG (H + L) (Thermo Fisher Scientific), or Alexa Fluor 647 goat anti‐rabbit IgG (H + L) (Thermo Fisher Scientific), at a dilution of 1:800. The cells were mounted using ProLong™ Gold Antifade Mountant with DAPI (Thermo Fisher Scientific). The fluorescent images were taken from three different fields (approximately 10 cells per field) using an FluoView FV1000 (Olympus, Tokyo, Japan). The colocalization signals in the merged images from at least 30 cells were quantified using the FluoView software and evaluated using Pearson's correlation coefficient. To visualize the colocalization of viral DNA with mitotic chromosomes, cells were incubated with EdU‐labeled 16PsV (0.2 μg of L1 per well) in an eight‐well chamber slide (1 × 10⁴ cells per well) at 37 °C for 24 h. After incubation, the cells were fixed with 4% paraformaldehyde and permeabilized with 0.5% Triton X‐100. EdU was labeled with Alexa Fluor 488 azide through the Click‐iT reaction. The cells were mounted using ProLong™ Gold Antifade Mountant with DAPI (Thermo Fisher Scientific). Fluorescent images of approximately 30 mitotic cells were taken with a FluoView FV1000 (Olympus). The colocalization signals in merged images from more than 30 cells were quantified using the FluoView software and evaluated using Pearson's correlation coefficient.

Proximity ligation assay (PLA)

Cells were seeded into an eight‐well chamber slide (1 × 10⁴ cells per well). The next day, the cells were infected with 16PsV (1 μg of L1 per well). Sixteen h postinfection, the cells were fixed with 4% paraformaldehyde in PBS for 15 min and permeabilized with 0.5% Triton X‐100 in PBS for 20 min at room temperature. PLA was performed using Duolink™ (Merck, Darmstadt, Germany) according to the Duolink^® PLA Fluorescence Protocol. Briefly, the permeabilized cells were blocked with Duolink^® Blocking Solution at 37 °C for 1 h. The cells were incubated at 37 °C for 1 h with 100‐fold diluted mouse and rabbit antibodies as follows: mouse anti‐HPV16L1 antibody (554 171, BD Biosciences, San Jose, CA, USA) and rabbit anti‐TGN46 antibody (ab50595, Abcam), or rabbit anti‐GM130 antibody (ab52649, Abcam). The cells were then incubated with Duolink^® PLA Probes (anti‐mouse PLUS and anti‐rabbit MINUS) at 37 °C for 1 h 15 min, followed by incubation with a ligation mixture at 37 °C for 45 min and an amplification mixture at 37 °C for 3 h. Finally, the cells were mounted with Duolink^® In Situ Mounting Media with DAPI. The fluorescent images were taken from three to five different fields (approximately 10 cells per field) using a FluoView FV1000. The fluorescence intensity from at least 30 cells in the images was quantified using the FluoView software.

Statistical analysis

All statistical analyses were performed using Microsoft Excel. Graphical representations were created using Microsoft Excel and GraphPad Prism10.

Results

Rab14 is required for efficient HPV infection

We generated Rab14 KO HeLa cells by transfection of the CRISPR‐Cas9 expression plasmid and examined their susceptibility to HPV16/18 PsV infection. Sequence analysis of the genomic DNA from a Rab14 KO clone demonstrated that the coding sequence in exon 4 of RAB14 was frameshifted in all alleles. Western blot analysis confirmed that Rab14 expression was abolished entirely in the Rab14 KO cells (Fig. 1A). The proliferation of the KO cells showed no change compared to the parental cells. In infection assays, these cells showed approximately 80% reduced susceptibility to each PsV infection compared to the control cells. Notably, the reintroduction of wild‐type Rab14 fused with a hemagglutinin (HA) tag into the Rab14 KO cells restored the levels of PsV infection (Fig. 1A). This observation confirmed that the reduction in HPV infection was attributable to the loss of Rab14 expression. The effect of Rab14 KO was also examined with another cell line, immortalized human keratinocytes HaCaT. Similar to HeLa cells, PsV infection was significantly reduced in Rab14 KO HaCaT cells (Fig. 1B).

Fig. 1.

Effect of Rab14 knockout on PsV infection. (A) Construction of Rab14 knockout (KO) and Rab14‐reintroduced HeLa cells. (Top) The sequences of Rab14 KO exon 4, in which red hyphens indicate deleted nucleotides that cause frameshift mutations. The sequence below the arrow shows the Rab14 sgRNA sequence. The boxed sequence indicates the protospacer‐adjacent motif sequence. (Bottom left) Western blotting of the lysates of parental (Ctl), knockout (KO), and Rab14‐reintroduced KO (KO/HA‐Rab14) cells using anti‐Rab14 and anti‐α‐tubulin antibodies. (Bottom right) Infection levels of 16PsV and 18PsV in Ctl, KO, and KO/HA‐Rab14 cells. (B) Construction of Rab14 KO HaCaT cells. (Top) The sequences of Rab14 KO exon 4, in which red letters and red hyphens indicate inserted and deleted nucleotides, respectively. (Bottom left) Western blotting of the lysates of parental (Ctl) and knockout (KO) cells. (Bottom right) Infection levels of 16PsV and 18PsV in Ctl and KO cells. (C) Infection levels of lentivirus pseudo‐typed with vesicular stomatitis virus glycoprotein (rLV) and recombinant adeno‐associated virus serotype 2 (rAAV2) in Ctl and KO HeLa cells. The average and standard deviation of three biological replicates are shown. ****, P < 0.001; ns, not significant (Student's t test).

Next, we tested whether infection with other viruses was also affected by Rab14 KO. For this purpose, we inoculated the Rab14 KO cells with a recombinant lentivirus pseudo‐typed with the glycoprotein of vesicular stomatitis virus (rLV) or a recombinant adeno‐associated virus serotype 2 (rAAV2). As shown in Fig. 1C, Rab14 KO HeLa cells retained their susceptibility to infection by rLV, similar to that of the parental cells. Meanwhile, the KO cells showed slightly reduced susceptibility to infection with rAAV2. These findings suggest that the loss of Rab14 does not cause broad defects in cellular functions that support viral infection.

The loss of Rab14 reduces the transport of incoming HPV virions to the TGN

We investigated the effect of Rab14 KO on the subcellular localization of HPV16 PsV that contained the reporter plasmid labeled with EdU, which enables the visualization of PsV by the Clik‐iT reaction. Rab14 KO and parental HeLa cells were incubated with the EdU‐labeled 16PsV at 37 °C for 2, 8, or 16 h and subjected to PsV visualization using the Click‐iT reaction. The subcellular localization of early endosome antigen 1 (EEA1) (an early endosome marker), TGN46 (a TGN marker), and lysosome‐associated membrane protein 1 (LAMP1) (a late endosome/lysosome marker) was also visualized with the corresponding antibodies. As shown in Fig. 2, the colocalization of the 16PsV and TGN46 was significantly reduced in the KO cells compared with the control cells after 8 and 16 h incubation, indicating that Rab14 is involved in the transport of incoming HPV virions to the TGN. In contrast, the colocalization of 16PsV and LAMP1 was significantly increased in the KO cells compared to the control cells at 16‐h postinfection, suggesting that virions accumulate in the late endosome or lysosome.

Fig. 2.

Subcellular localization of 16PsV in Rab14 knockout cells. (Left panels) Control (Ctl) and Rab14 knockout (KO) HeLa cells were incubated with EdU‐labeled 16PsV at 37 °C for 2, 8, or 16 h. EdU was detected using the Click‐iT reaction. EEA1 (an early endosome marker), TGN46 (a trans‐Golgi marker), and LAMP1 (a late endosome/lysosome marker) were stained with corresponding specific antibodies. Fluorescence signals were visualized by confocal microscopy. EdU staining is shown as a green signal. EEA1, TGN46, and LAMP1 staining is shown as a magenta signal. The boxed area is enlarged in the upper right corner. (Right panels) The colocalization signals in the merged images from at least 30 cells were quantified using FluoView FV1000 software and evaluated with Pearson's correlation coefficient (Coeff). Each dot represents the Coeff value of individual cells (n > 30), and the solid horizontal lines indicate the average values. *, P < 0.05; ***, P < 0.005 (Student's t test).

We further examined the transport of HPV virions to the TGN using proximity ligation assay (PLA), which can detect in situ protein–protein interactions within 40 nm distance as fluorescent puncta using specific antibodies against two target proteins raised in different species. Rab14 KO and control HeLa cells were incubated with 16PsV at 37 °C for 16 h, followed by PLA with mouse anti‐HPV16 L1 and rabbit anti‐TGN46 antibodies. As shown in Fig. 3A, PLA puncta were significantly reduced in the Rab14 KO cells compared to the control cells. A similar reduction was also observed when PLA was performed using mouse anti‐L1 and rabbit anti‐GM130 (a cis‐Golgi marker) antibodies (Fig. 3B). These results strongly support the involvement of Rab14 in the transport of HPV virions to the Golgi apparatus.

Fig. 3.

Impaired localization of 16PsV to the Golgi apparatus in Rab14 knockout cells. (A and B, left panels) Control (Ctl) and Rab14 knockout (KO) HeLa cells were incubated with 16PsV at 37 °C for 16 h. Proximity ligation assay (PLA) was performed using specific antibodies against HPV16L1 and TGN46 (A) or GM130 (B). (−), uninfected; (+), infected. PLA signals are shown as green puncta. (A and B, right panels). The intensity of fluorescence signals from at least 30 cells was quantified using the FluoView FV1000 software. Each dot represents the fluorescence intensity value of individual cells (n > 30), and the solid horizontal lines indicate the average values. ***, P < 0.005; ****, P < 0.001; ns, not significant (Student's t test).

The GTPase cycle of Rab14 is required for the transport of HPV virions to the TGN

To understand the mechanisms of Rab14‐dependent transport of HPV virions to the TGN, we generated Rab14 KO HeLa cells that retrovirally express HA‐tagged Rab14 Q70L and S25N, constitutively active (GTP‐bound) and inactive (GDP‐bound) mutants of Rab14, respectively [26]. Western blot analysis verified the expression of these mutants in the Rab14 KO cells (Fig. 4A, left panel). In infection assays, unlike wild‐type Rab14, neither mutant restored the levels of 16PsV infection (Fig. 4A, right panel), indicating that GTPase cycling of Rab14 is required to support HPV infection.

Fig. 4.

Subcellular localization of 16PsV in cells expressing Rab14 mutants. (A) (Left panel) Western Blotting of the lysates of Rab14 KO HeLa cells (KO), HA‐fused Rab14 (HA‐Rab14)‐reintroduced KO cells (WT), HA‐Rab14 GTP‐bound form mutant‐reintroduced KO cells (Q70L), and HA‐Rab14 GDP‐bound form mutant‐reintroduced KO cells (S25N) using anti‐HA and anti‐α‐tubulin antibodies. (Right panel) Infection levels of 16PsV in WT, Q70L, S25N, and KO cells. The average and standard deviation of three biological replicates are shown. ****, P < 0.001 (Student's t test). (B) (Left panels) Subcellular localization of Rab14 wild‐type (WT) and mutants (Q70L and S25N). HA‐tag, EEA1, and TGN46 were stained with corresponding specific antibodies and visualized by confocal microscopy. EEA1 and TGN46 staining is shown as a green signal, while HA staining is shown as a magenta signal. The boxed area is enlarged in the upper right corner. (Right panels) The colocalization signals in the merged images from at least 30 cells were quantified using the FluoView FV1000 software and evaluated with Pearson's correlation coefficient (Coeff). Each dot represents the Coeff value of individual cells (n > 30), and the solid horizontal lines indicate the average values. ****, P < 0.001 (Student's t test). (C) (Left panels) Subcellular localization of 16PsV in HeLa cells expressing each type of Rab14. WT, Q70L, and S25N cells were incubated with EdU‐labeled 16PsV at 37 °C for 1, 4, or 8 h. EdU was detected using the Click‐iT reaction, and HA was stained with anti‐HA antibody. EdU and the anti‐HA antibody were visualized by confocal immunofluorescence microscopy. EdU staining is shown as a green signal, while HA staining is shown as a magenta signal. (Right panels) The colocalization signals in the merged images from at least 30 cells were quantified using the FluoView FV1000 software and evaluated with Pearson's correlation coefficient (Coeff). Each dot represents the Coeff value of individual cells (n > 30), and the solid horizontal lines indicate the average values. ****, P < 0.001 (Student's t test).

The subcellular localization of wild‐type Rab14, Q70L, and S25N was investigated using confocal microscopy. As shown in Fig. 4B, compared to cells expressing wild‐type Rab14, cells expressing S25N showed reduced colocalization of S25N and EEA1, while S25N more frequently showed colocalization with TGN46. Thus, the GDP‐bound Rab14 relocates from the early endosome to the vicinity of the TGN. These results indicate that Rab14 changes its intracellular localization depending on whether it is bound to GTP or GDP, as reported previously [26].

Nest, we examined the colocalization of incoming virions and the Rab14 mutants. After inoculating the cells with 16PsV for 4 and 8 h, the colocalization of PsV and Q70L was significantly increased compared to that of PsV and wild‐type Rab14 (Fig. 4C), suggesting that HPV virions remain within cellular organelles or vesicles containing GTP‐bound Rab14, presumably the early endosome. In contrast, the colocalization of PsV and S25N was significantly reduced after 4‐h inoculation compared to that of PsV and wild‐type Rab14. Overall, these results indicate that incoming HPV is efficiently transported from the early endosome to the TGN in a manner dependent on Rab14 GTPase cycling.

Loss of Rab14 impairs the colocalization of viral DNA with mitotic chromosomes

Finally, since HPV infection requires the delivery of viral DNA to the host cell's chromosomes during mitosis [33], we examined the effect of Rab14 KO on the colocalization of viral DNA with mitotic chromosomes. Rab14 KO and parental HeLa cells were incubated with EdU‐labeled 16PsV at 37 °C for 24 h, and viral DNA was visualized using the Click‐iT reaction, while nuclear DNA was counterstained with DAPI. As shown in Fig. 5, the colocalization of viral DNA with metaphase chromosomes was severely impaired in Rab14 KO cells when compared to control cells. These results suggest that Rab14 is also involved in delivering HPV DNA to mitotic chromosomes.

Fig. 5.

Colocalization of EdU‐labeled viral DNA with chromosomes in mitotic cells. (Left panels) Control (Ctl) and Rab14 knockout (KO) HeLa cells were incubated with EdU‐labeled 16PsV at 37 °C for 24 h. EdU was detected using the Click‐iT reaction. The nucleus was stained with DAPI. Fluorescence signals were visualized by confocal microscopy. EdU staining was shown as a green signal. Nucleus staining was shown as a blue signal. (Right panels) The colocalization signals in the merged images of at least 30 mitotic cells were quantified using the FluoView FV1000 software and evaluated with Pearson's correlation coefficient (Coeff). Each dot represents the Coeff value of an individual cell (n > 30), and the solid horizontal lines indicate the average values. ****, P < 0.001 (Student's t test).

Discussion

During HPV cell entry, incoming virions must be transported to the TGN [22]. In this study, we found that Rab14 is required for HPV infection and plays an essential role in supporting the transport of the virion to the TGN. In Rab14 KO cells, the virion showed reduced localization in the TGN (Fig. 2). Rab14 colocalized with the early endosome and the TGN at equal levels; however, GDP‐bound Rab14 (Rab14‐GDP) significantly colocalized with the TGN (Fig. 4B). Additionally, the incoming virion was more frequently colocalized with GTP‐bound Rab14 (Rab14‐GTP), but less associated with Rab14‐GDP (Fig. 4C). Collectively, these results suggest that the HPV virion is transported from organelles or vesicles containing Rab14‐GTP to the TGN, depending on the conversion from Rab14‐GTP to Rab14‐GDP. To our knowledge, this is the first report demonstrating that Rab14 is an essential host factor for HPV infection.

It has already been reported that several Rabs are involved in the intracellular trafficking of HPV virions [19, 20, 21, 22, 23, 24], among which Rab7 plays a crucial role in transporting the virus to the TGN. The interaction of L2 with the retromer complex recruits a Rab7 GTPase‐activating protein, TBC1D5, to stimulate hydrolysis of Rab7‐GTP at the endosome membrane, which induces retromer disassembly from L2, thereby delivering the virus to the retrograde pathway for transport to the TGN [34]. Notably, this trafficking requires the cycling of Rab7 between GTP‐ and GDP‐bound forms [34], as does Rab14 GTPase cycling observed in our study.

Rab14 participates in various intracellular transport pathways, including the transport of multiple cargos from the Golgi to the endosome [26], endocytic recycling [35, 36], and retrograde transport [37, 38]. Furthermore, Rab14 has been linked to Rab5‐ and Rab7‐independent endocytic pathways [39] and the degradation of mitochondria by mitophagy [40]. It has also been reported that Rab14 interacts with the Golgi‐associated retrograde protein (GARP) complex [41], which is necessary for tethering endosome‐derived vesicles to the TGN [42]. Furthermore, Rab14 was shown to interact with the lipid transfer protein SHIP164 [38, 41], which binds to the GARP complex and is required for the retrograde transport of cellular cargos [43, 44]. Therefore, Rab14 may be involved in retrograde transport of HPV by supporting the tethering of virus‐carrying endosomal vesicles to the TGN. At this point, it remains unclear whether Rab14 directly or indirectly regulates HPV transport to the TGN; however, no interaction between Rab14 and L2 was detected in co‐immunoprecipitation assays (data not shown).

Additionally, it has been shown that Rab14 promotes lysosomal acidification by facilitating the delivery of V‐ATPase to the lysosome [45]. Inhibition of V‐ATPase affects disassembly of HPV virions within the endosome, leading to reduced infection [46]. Similarly, reduced endosomal acidification in Rab14 KO cells may result in insufficient disassembly of the L1 capsid, which in turn could inhibit the exposure of L2 from the capsid and indirectly impede HPV transport to the TGN.

Although a statistically significant decrease in PsV transport to the TGN was observed in Rab14 knockout cells, this effect alone may not fully account for the marked reduction in PsV infection in these cells. The significant reduction in viral DNA delivered to the mitotic chromosomes could further exacerbate the decline in infection in Rab14 knockout cells. To establish infection, HPV migrates from the Golgi apparatus to microtubule‐organizing centers during prometaphase and reaches the host's metaphase chromosomes [17, 47, 48]. Indeed, viral DNA was colocalized with the mitotic chromosomes of control cells, whereas its chromosome localization was severely impaired in Rab14 KO cells. HPV DNA transport to the mitotic chromosomes involves Ran‐binding protein 10 and karyopherin alpha2, and the dynein light chain DYNLT3 is also required for spindle minus‐end microtubule transport [49]. Moreover, cyclin‐dependent kinase 1 and Polo‐like kinase 1 function as kinases required for the phosphorylation of L2, thereby regulating the proper delivery of HPV DNA to mitotic chromatin [50]. How these nuclear processes might be regulated by Rab14 is an important question that requires further investigation.

In conclusion, we demonstrated that Rab14 plays a critical role in facilitating the intracellular transport of HPV to the TGN. Furthermore, Rab14 may also be involved in post‐Golgi trafficking of viral DNA in the nucleus. A more detailed investigation of the HPV transport mechanism regulated by Rab14, as well as the effects of Rab14 depletion on cellular function, is likely to yield new insights into HPV infection mechanisms. Such analysis will also deepen our understanding of intracellular protein transport in general.

Author contributions

Conceptualization: YI; Funding acquisition: YI, IK; Methodology: YI; Project administration: YI, IK; Resources: YI, IK; Investigation: YI; Visualization: YI; Writing‐original draft: YI; Writing‐review and editing: YI, IK.

Data accessibility

The data that support the findings of this study are available from the corresponding author [ishii.yo@jihs.go.jp] upon reasonable request.