Journ

alof

Cell

Scie

nce

Neisseria gonorrhoeae infection causes DNA damageand affects the expression of p21, p27 and p53 innon-tumor epithelial cells

Katarina Vielfort1, Niklas Soderholm1,*, Linda Weyler1,*, Daniel Vare1, Sonja Lofmark2 and Helena Aro1,`

1Department of Genetics, Microbiology and Toxicology, Svante Arrhenius vag 20C, Stockholm University, SE-106 91 Stockholm, Sweden2Swedish Institute for Communicable Disease Control, Nobels vag 18, SE-171 82 Solna, Sweden

*These authors contributed equally to this work`Author for correspondence (helena.aro@gmt.su.se)

Accepted 8 October 2012Journal of Cell Science 126, 339–347� 2013. Published by The Company of Biologists Ltddoi: 10.1242/jcs.117721

SummaryThe constant shedding and renewal of epithelial cells maintain the protection of epithelial barriers. Interference with the processes ofhost cell-cycle regulation and barrier integrity permits the bacterial pathogen Neisseria gonorrhoeae to effectively colonize and invadeepithelial cells. Here, we show that a gonococcal infection causes DNA damage in human non-tumor vaginal VK2/E6E7 cells with an

increase of 700 DNA strand breaks per cell per hour as detected by an alkaline DNA unwinding assay. Infected cells exhibited elevatedlevels of DNA double-strand breaks, as indicated by a more than 50% increase in cells expressing DNA damage-response protein53BP1-positive foci that co-localized with phosphorylated histone H2AX (cH2AX). Furthermore, infected cells abolished their

expression of the tumor protein p53 and induced an increase in the expression of cyclin-dependent kinase inhibitors p21 and p27 to 2.6-fold and 4.2-fold of controls, respectively. As shown by live-cell microscopy, flow cytometry assays, and BrdU incorporation assays,gonococcal infection slowed the host cell-cycle progression mainly by impairing progression through the G2 phase. Our findings show

new cellular players that are involved in the control of the human cell cycle during gonococcal infection and the potential of bacteria tocause cellular abnormalities.

Key words: DNA damage, N. gonorrhoeae, VK2/E6E7

IntroductionThe mammalian cell cycle plays significant roles in many

biological processes including immune response, the maintenance

of epithelial barrier functions and cellular differentiation (Jacinto

et al., 2001; Nougayrede et al., 2005). An active cell cycle in the

mucosal epithelium is central to maintaining the fidelity of the

mucosal barrier against invasive microorganisms. Eukaryotic cell

cycle progression is driven by cyclin-dependent kinases (CDKs)

and cyclins. The regulation of CDK-cyclin complex activity occurs

through cyclin-dependent kinase inhibitors (CKIs), such as p21 and

p27, at checkpoints that halt the cell cycle in response to un-

replicated or damaged DNA, which allows time for repair or

apoptosis (Vermeulen et al., 2003). Cellular damage has been

recognized to occur during bacterial infections (Kundu and Surh,

2008), and microbial pathogens have developed a variety of

strategies to manipulate host cell functions, including the release of

toxins that inhibit or promote cell cycle progression (Lara-Tejero

and Galan, 2000; Nougayrede et al., 2005; Oswald et al., 2005),

associate with host cell microtubules (Hu and Kopecko, 1999), or

block multiple checkpoints in the cell cycle (Scanlon et al., 2000).

Neisseria gonorrhoeae, which causes gonorrhea, infects over

80 million individuals worldwide each year (Department of

Reproductive Health and Research, World Health Organization,

2011; see Emergence of multi-drug resistant Neisseria

gonorrhoeae – Threat of global rise in untreatable sexually

transmitted infections. Fact Sheet WHO/RHR/11.14). Many

infected individuals show no symptoms of illness, leading to

long-term asymptomatic infections and low-grade inflammation

over the course of several months. Moreover, repeated

symptomatic infections, such as urethritis, prostatitis and

epididymitis in males, affect the host mucosa. Gonococcal

infections can lead to severe complications, such as pelvic

inflammatory disease, ectopic pregnancy, and infertility and

perhaps in the worst cases, cervical, prostate or bladder cancer

(Michaud, 2007; Sutcliffe et al., 2006a; Sutcliffe et al., 2006b).

The initial attachment of the bacteria to the apical surface of

epithelial tissues is mediated by type IV pili (Kallstrom et al.,

1997; Merz et al., 1999; Merz et al., 1996; Swanson, 1973). The

adherence mediates host cell-signaling events and elicits a

multitude of cellular responses, including cortical plaque

formation (Merz et al., 1999), the release of intracellular

calcium (Kallstrom et al., 2000; Kallstrom et al., 1998), and

the induction of anti-apoptotic factors (Binnicker et al., 2003;

Follows et al., 2009). During N. gonorrhoeae infection, several

signaling pathways are induced by virulence factors such as

porins, lipooligosaccharides, iron receptors, outer membrane

vesicles, and pilus biogenesis-associated proteins (for reviews,

see Popp et al., 2001; Koomey, 2001).

As previously shown, N. gonorrhoeae arrests HeLa cells in

the early G1 phase of the eukaryotic cell cycle and causes the

degradation of cyclins (Jones et al., 2007). In ME-180 cells,

the infection upregulates, alternatively cleaves, and changes the

Research Article 339

Journ

alof

Cell

Scie

nce

subcellular distribution of the human growth factor amphiregulin,a protein known to be upregulated in different forms of cancer(Lofmark et al., 2011). A bacterial infection may not simply

impact the mucosal and epithelial cell barrier but may also affecthost cell DNA integrity; indeed, infections of Helicobacter pylori

and certain Escherichia coli strains have been shown to directly

cause DNA breaks in host cells (Cuevas-Ramos et al., 2010;Toller et al., 2011). In this study, we aim to explore whether N.

gonorrhoeae can induce DNA damage. To this end, we have

chosen a non-tumor cell line from the human vaginal mucosa thatexpresses characteristics of stratified squamous non-keratinizingepithelia. Piliated, but not non-piliated, N. gonorrhoeae binds toand invades these cells (Fichorova et al., 2001).

ResultsNeisseria gonorrhoeae causes DNA strand breaks

An alkaline DNA unwinding assay (ADU) was performed todetermine whether N. gonorrhoeae colonization caused DNAdamage to host cells. The ADU method enables quantification of

strand breaks in the DNA; based on the well-establishedstatement that 1 Gy of c irradiation causes 1000 SSBs (single-strand breaks) and 40 DSBs (double-strand breaks) per cell (Nias,

1998; Nikjoo et al., 1994).

Healthy cells constantly experience both SSB and DSB that are

continuously repaired. Therefore, to detect and measure DNAbreaks DNA repair inhibitors have to be added during the assay(supplementary material Fig. S1). Cells may use different

pathways for DNA repair in response to different types ofDNA lesions, and in an initial screening, we determined that aPARP inhibitor, which mainly inhibits base excision repair

(BER), inhibited the repair pathway in VK2/E6E7 cells whileHU/AraC, which inhibits nucleotide excision repair (NER)inhibited the repair pathway in ME-180 cells (data not shown).

Cells were infected with N. gonorrhoeae for 6 hours as an

early time-point and for 24 hours as a late time-point. DNArepair inhibitors were present during the last hour of the 6-hourassay. For the later time-point, the inhibitors were present for thelast four hours. Post infection, DNA was unwound and sonicated,

and ssDNA and dsDNA was separated by hydroxyapatitechromatography. In uninfected control VK2/E6E7 cells 19% oftotal 3H-labeled DNA was eluted as ssDNA (Fig. 1A). After

6 hours and 24 hours of infection the percentage of ssDNAeluted increased to 26% and 28%, respectively. In uninfectedcontrol ME-180 cells 12% of total 3H labeled DNA was eluted as

ssDNA (Fig. 1B). After 6 and 24 hours of infection thepercentage of ssDNA eluted increased to 18% and 34%,respectively. Each time-point was performed three times.

Based on the well-established statement that 1 Gy of cirradiation causes 1040 strand breaks per cell, a c irradiation

standard curve was made for both cell lines (supplementarymaterial Fig. S2A) and the linear correlation between the 2LOGvalues of dsDNA (total DNA-ssDNA) (supplementary material

Fig. S2B) allowed us to quantify the number of strand breaksinduced by the bacterial infection. Since the DNA repairinhibitors were present for 4 hours during the 24-hour time-

points, the values were divided by a factor of 4 to generate anaverage of the number of strand breaks per hour. Taken together,elevated numbers of strand breaks per cell were present in the

infected cells compared to the non-infected control cells. At6 hours, control VK2/E6E7 cells accumulated an average of 660strand breaks per cell/hour while the infected VK2/E6E7 cells

exhibited 1350 strand breaks per cell/hour (Fig. 1C). At 24 hours,

the infection caused an average of 640 strand breaks per infected

cell/hour. For uninfected ME-180 cells, the 6-hour time-point

accumulated 410 strand breaks per control cell/hour (Fig. 1D). At

6 hours and 24 hours, the infection increased the average to 1260

and 580 strand breaks per cell/hour, respectively.

The inhibitors used for either VK2/E6E7 or ME-180 cells did

not affect the viability of the bacteria. In addition, bacterial DNA

did not interfere with the ADU assay (data not shown).

The expression of 53BP1 foci was increased during

gonococcal infection

To verify the DNA damage observed during gonococcal

infection, the cells were stained for p53 binding protein 1

(53BP1), which specifically localizes to DNA double-strand

breaks. Cells were infected for 24 hours, fixed, and stained with

an antibody against 53BP1. The number of 53BP1-positive foci

was counted, and any cell containing more than four positive foci

per nucleus was deemed a positive cell. Gonococcal infection

increased the number of 53BP1-positive cells by 1.55-fold in the

case of VK2/E6E7 cells and 1.45-fold in the case of ME-180 cells

compared with non-infected control cells (Fig. 2A).

Following DNA damage and DNA replication, Histone 2A

(H2A) becomes phosphorylated at serine 139 and is then termed

cH2AX. An anti-cH2AX antibody was used to further analyze the

Fig. 1. Gonococcal infection of VK2/E6E7 and ME-180 cells causes DNA

strand breaks. An alkaline DNA unwinding assay showing DNA strand

breaks in infected cells. The 3H-TdR-labeled DNA from infected or control

cells was unwound and eluted by hydroxyapatite chromatography. DNA

repair inhibitors were added for the last 4 hours of infection in the 24-hour

time-points and for the last 1 hour in the 6-hour time-points as well as in the

standard curve. Graphs show the percentage of induced 3H-labeled single

stranded DNA (ssDNA) of total 3H-labeled DNA eluted from (A) VK2/E6E7

cells and (B) ME-180 cells. The numbers of strand breaks per cell per hour

caused by 6 and 24 hours of gonococcal infection were calculated from the

equations from supplementary material Fig. S2 and are shown in (C) for

infected VK2/E6E7 cells and (D) for infected ME-180 cells. A paired 2-tailed

Student’s t-test was used for statistical analysis (*P,0.05 unless stated

otherwise).

Journal of Cell Science 126 (1)340

Journ

alof

Cell

Scie

nce

DNA damage in infected cells; however, high-background caused

by scattered noise and sporadic, undetermined pan-nuclear staining

was observed for both infected and control cells, making the foci

unable to be quantified in the microscope. This is in agreement with

Cleaver (Cleaver, 2011) who showed that cH2AX plays roles in

other cellular events and is not a specific marker for DNA damage.

However, increased fluorescent signals upon cH2AX staining was

clearly visible in infected cells as well as a clear co-localization with

the 53BP1-positive foci (Fig. 2B). Taken together, these data

demonstrate that N. gonorrhoeae causes DNA damage to host cells.

Neisseria gonorrhoeae induces and redistributes p21

and p27

We next investigated the expression of p21 and p27, two DNA

damage-response proteins of the Cip/Kip family. A 24-hour

gonococcal infection in VK2/E6E7 cells resulted in a 2.6-fold

increase in p21 protein levels (Fig. 3A) and a 2.2-fold increase in

ME-180 cells (Fig. 3B). Gonococcal infection also induced the

expression of p27. Infected VK2/E6E7 cells showed a 4.2-fold

increase in p27 expression (Fig. 4A), and infected ME-180 cells

exhibited a 2.7-fold increase (Fig. 4B). An immunofluorescence

assay on VK2/E6E7 and ME-180 cells revealed that both p21 and

p27 are homogeneously distributed throughout the cells, whereas

in infected cells, a pronounced nuclear localization of both p21

(Fig. 3) and p27 (Fig. 4) was observed at 24 hours post-infection.

Because DNA damage was already detected at 6 hours

post-infection, the localization of p21 and p27 was also

investigated at this early time-point. Infected VK2/E6E7 and

ME-180 cells both showed a redistribution of p21 and p27 to the

nucleus at 6 hours. The p21 and p27 proteins are cell cycle

inhibitors, and their nuclear localization is associated with a

slowing of the cell cycle (Rodrıguez-Vilarrupla et al., 2002; Sherr

and Roberts, 1995). Hence, these data indicate a possible

deceleration of the cell cycle during infection. To summarize,

gonococcal infection elevates both p21 and p27 protein levels,

and both p21 and p27 were redistributed to the cell nucleus,

indicative of a slowing in cell cycle progression.

N. gonorrhoeae infection downregulates p53 in VK2/E6E7cells but not in ME-180 cells

The observed induction of p21 may be either p53-dependent or

p53-independent. Therefore, to further understand the DNA

damage response, we sought to determine the role of the tumor

repressor protein p53 during a gonococcal infection. p53 is

important both during DNA repair and for arresting the cell cycle

as a consequence of DNA damage. In a western blot analysis,

antibodies detected the presence of p53 in both VK2/E6E7 and

ME-180 control cells. Surprisingly, p53 protein was undetectable

in 24-hour infected VK2/E6E7 cells (Fig. 5A) but unchanged in

24-hour infected ME-180 cells (Fig. 5B). These data were also

Fig. 2. The proportion of 53BP1-positive cells increased during

gonococcal infection. In three independent experiments, 300 cells were

counted, and a cell nucleus containing .4 positive foci was considered

positive. (A) Graph shows the proportion of 53BP1-positive VK2/E6E7 cells

and ME-180 cells after 24 hours of infection (black bars). Values were

normalized against the number of 53PB1 positive uninfected control cells

(white bars) set to 1. Representative images of cells infected for 24 hours and

control cells are shown in right panel. Cells were stained with DAPI (blue)

and for 53BP1 (red). Scale bar: 10 mm. (B) Fluorescence staining of cH2AX

and 53BP1 in control and infected VK2/E6E7 cells. The 53BP1-positive foci

co-localize with cH2AX-positive foci. Scale bar: 10 mm. A paired 2-tailed

Student’s t-test was used for statistical analysis (*P,0.05).

Fig. 3. N. gonorrhoeae infection of both VK2/E6E7 and ME-180 cells

elevates p21 protein levels. Cells were infected with N. gonorrhoeae for

24 hours and analyzed for p21 expression by western blot. The graph shows

band intensities normalized to a tubulin. The expression of p21 at 6 and

24 hours post-infection was visualized by immunofluorescence. Scale bars:

10 mm. (A) The expression of p21 protein and its subcellular localization in

VK2/E6E7 cells. (B) The expression of p21 protein and its subcellular

localization in ME-180 cells. A paired 2-tailed Student’s t-test was used for

statistical analysis (*P,0.05).

Gonococcal infection causes DNA damage 341

Journ

alof

Cell

Scie

nce

confirmed by an immunofluorescence assay using the same

antibody after both 6-hour and 24-hour infection. These data

show that a gonococcal infection decreases p53 levels in VK2/

E6E7 cells but not in ME-180 cells.

Further, quantitative PCR analysis using p53 specific primers

showed that the mRNA was downregulated more than 5-fold

during 24 hours of gonococcal infection in VK2/E6E7 cells in

comparison to uninfected control cells. The mRNA levels was

normalized against Alpha tubulin (TUBA1A) (Fig. 5C) and

housekeeping GAPDH (data not shown).

Gonococcal infection decelerates the cell cycle

Following DNA damage, the CDK inhibitors p21 and p27 arrest

the progression of the cell cycle. Therefore, we next studied

infected and control cells at different stages of the cell cycle.

Cells were infected and transferred to a live-cell microscope.

Randomly chosen positions of infected and non-infected cells

were selected, and images were captured every 10 minutes from

between 3 and 24 hours post-infection. Individual cells were

followed, and the time-point when each cell had completed

cytokinesis was logged. Shortly following 12 hours post-

infection, the deceleration of the cell cycle in VK2/E6E7 cells

was observed, and following 18 hours, 36% of the infected

cells had completed cytokinesis, compared to 43% of the control

cells. By the end of the assay, 60% of the non-infected VK2/

E6E7 cells had completed cytokinesis, whereas only 50% of the

N. gonorrhoeae-infected cells had divided (Fig. 6A). In the case

of ME-180 cells, 67% of non-infected cells had completed

cytokinesis after 24 hours, whereas only 60% of infected cells

had divided (Fig. 6B). The variance within the large number of

ME-180 cells counted is greater. This is most likely due to the

fact that it is a tumor cell line. No increase in apoptosis was

observed in the infected cells (data not shown).

A BrdU proliferation assay was used to further verify the

reduction in proliferation caused by N. gonorrhoeae infection.

Bacteria were added to cells, and the relative BrdU incorporation

compared to non-infected cells was quantified. A 24-hour

infection caused a 48% reduction in BrdU incorporation in

VK2/E6E7 cells (Fig. 7A) and 30% in ME-180 cells (Fig. 7B).

Therefore, infection reduces the number of cells progressing

through S phase. In order to see if any released products in the

cell supernatant caused the cell cycle deceleration, we used

sterile-filtrated supernatants from a 24-hour ME-180 cell culture,

a 24-hour infection, and a bacterial culture grown in cell culture

Fig. 4. N. gonorrhoeae infection of both VK2/E6E7 and ME-180 cells

elevates the expression of p27. Cells were infected with N. gonorrhoeae for

24 hours and analyzed for p27 expression by western blot. The graph shows

western blot band intensities normalized to a tubulin. The expression of p27

at 6 and 24 hours post-infection was visualized by immunofluorescence.

Scale bar: 10 mm. (A) The expression of p27 protein and its subcellular

localization in VK2/E6E7 cells. (B) The expression of p27 protein and its

subcellular localization in ME-180 cells. A paired 2-tailed Student’s t-test was

used for statistical analysis (*P,0.05).

Fig. 5. N. gonorrhoeae infection of VK2/E6E7 but not ME-180 cells

abrogates the expression of p53. Cells were infected with N. gonorrhoeae for

24 hours and analyzed for p53 expression by western blot relative to a tubulin

expression. The expression and localization of p53 at 6 hours and 24 hours

post-infection were visualized by immunofluorescence. Images were taken

from three independent experiments. Shown are representative western blots

and fluorescent images from infected and control (A) VK2/E6E7 cells and

(B) ME-180 cells. Scale bar: 10 mm. (C) The mRNA levels of p53 in infected

VK2/E6E7 cells were quantified by qPCR and compared with uninfected

control cells and normalized to mRNA of a-tubulin (TUBA1A). A paired 2-

tailed Student’s t-test was used for statistical analysis (*P,0.05).

Journal of Cell Science 126 (1)342

Journ

alof

Cell

Scie

nce

media. None of the supernatants were able to reduce BrdU

incorporation, suggesting that the S phase reduction is mediated

continuous bacterial presence (supplementary material Fig. S3).

Cells infected with N. gonorrhoeae show impaired G2

phase progression

To determine whether the cell cycle delay occurs at a specific

phase, we analyzed cell cycle profiles by flow cytometry. Cells

were synchronized by a double thymidine block and infected

1 hour after release. Cells were harvested at 0 hours (at the start

of the infection) and at 6, 12, 18 and 24 hours post-infection.

Both control and infected VK2/E6E7 cells showed similar cell

cycle profiles at 6 hours, with 63% of the control cells and 60%

of the infected cells having entered G2 phase (Fig. 8A).

However, after 12 hours, 69% of the control cells were in G1

phase, whereas only 40% of the infected cells had progressed to

G1. Therefore, the majority of the infected cells were impeded in

G2 phase. The reduced ability of infected cells to progress

through G2 phase was still observed at 18 hours, with a greater

number of infected cells still in G2 phase (30%) compared to

control cells (14%). As the infection proceeded, the delay in G2

phase resulted in a greater number of infected cells (67%) being

in G1 phase at the 24-hour time-point compared to control cells

(35%) (Fig. 8A).

Following their release from the thymidine block, the ME-180

cells progressed through the cell cycle more slowly than VK2/

E6E7 cells. After 6 hours of infection, most of the control ME-

180 cells (71%) and infected cells (68%) had entered S phase. At

the 12-hour time-point, both control ME-180 cells (65%) and

infected cells (73%) had reached G2 phase. A prominent delay in

progressing through G2 phase for infected cells was observed at

the 18-hour time-point, with 47% of infected cells in G2 phase

compared to 28% of the control cells (Fig. 8B). Similar to

infected VK2/E6E7 cells, this delay in cell cycle progression

caused a significant increase in the number of infected ME-180

cells in G1 phase (47%) at the 24-hour time-point compared to

control cells (41%). The infected cells were not totally arrested in

any one cell cycle phase; however, they showed a delayed

progression through G2 phase and through the G2/M checkpoint,

thus spending a longer time in G2 phase.

DiscussionIt is becoming increasingly apparent that the human cell cycle is

affected by bacterial infections. Many bacterial pathogens

manipulate the host cell cycle to benefit bacterial attachment,

survival and growth within the host at the cost of eukaryotic

DNA integrity. A healthy human cell experiences both SSB and

DSB every hour along with other types of DNA lesions

(Friedberg, 1995; Lindahl and Barnes, 2000). Here, we reveal

that gonococcal infection more than doubles these strand breaks

per cell per hour in two different epithelial cell lines, to levels

comparable to 1 Gy of c irradiation. But in contrast to instant

doses of irradiation, gonococcal infection can proceed for a very

long time (weeks or months), and although cells efficiently repair

strand breaks during persistent infection, increased DNA damage

intensifies the pressure on the DNA repair system.

The number of strand breaks per infected cell was quantified

using the ADU technique, and an inhibitor of DNA repair was

Fig. 6. Gonococcal infection decelerates cell cycle progression. Non-

confluent cells were infected with N. gonorrhoeae. Multiple images were

captured 3–24 hours post-infection in a live-cell microscopy assay. The

experiment was repeated three times, in which a total of 450 individual cells

were followed until they successfully completed cytokinesis. The

percentages of cells that had completed cytokinesis after 12, 18, and

24 hours in infected and control (A) VK2/E6E7 cells and (B) ME-180 cells

are shown. Data from the 24-hour time-point were statistically analyzed. A

paired 2-tailed Student’s t-test was used for statistical analysis (* indicates

that P,0.05).

Fig. 7. Gonococcal infection reduces BrdU incorporation. Non-confluent

VK2/E6E7 and ME-180 cells were incubated for 24 hours with N.

gonorrhoeae, and BrdU incorporation was determined. The average of at least

four independent experiments performed in triplicate is shown, and BrdU

uptake in control cells was set to 100%. A paired 2-tailed Student’s t-test was

used for statistical analysis (* indicates that P,0.05).

Gonococcal infection causes DNA damage 343

Journ

alof

Cell

Scie

nce used to accumulate the strand breaks at different time-points. The

DNA repair mechanism of the vaginal non-tumor cell line VK2/

E6E7 cells seem to be different from the cervical tumor cell line

ME-180. While ME-180 cells repaired the damage using NER,

the VK2/E6E7 cells repaired the damage using BER, indicating

either two different kinds of lesions in the two cell lines, or cell

line specific preferred pathways for the same kind of DNA lesion.

The cause of this will need further investigation.

The number of strand breaks determined here includes mostly

SSB but most likely also DSB. In order to determine whether a

gonococcal infection also causes DSB we used antibodies against

53BP1 and cH2AX. The accumulation of both cH2AX and 53BP1

at DNA repair foci is a well-established hallmark for DNA double-

strand breaks. 53BP1 interacts with cH2AX (Ward et al., 2003),

and gonococcal infection increased the number of cells positive for

53BP1 by 1.5-fold, with this expression co-localizing with cH2AX

staining. Indeed, a gonococcal infection causes both SSB and DSB

in both VK2/E6E7 cells and ME-180 cells.

We further show that gonococcal infections induce the

expression of p21 and p27. These two proteins have roles both

during DNA damage and repair and in the regulation of cell cycle

progression. Following DNA damage, p21 can arrest the cells in

both G1/S and G2/M (Gartel and Tyner, 2002). The p21

expression can be induced independently of p53 in the

presence of DNA damaging agents (Russo et al., 1995). This is

most likely the case for gonococcal infections in VK2/E6E7 cells,

because infection in these cells also resulted in greatly reduced

levels of p53. The stable expression of p53 in VK2/E6E7 cell

lines have previously been shown by Deva et al. where they

interestingly also could detect a degradation of p53 upon another

microorganism, Candida albicans (Deva et al., 2010). This has

recently been shown to be the case also for H. pylori infections in

vivo, in which the bacteria increased the ubiquitination and

proteasome degradation of p53 via the AKT1 pathway, which led

to the inhibition of apoptosis and increased survival of those cells

that suffered DNA damage (Wei et al., 2010). Gonococci infect

both vaginal and cervical cells and the differences seen between

cervical ME-180 cells and vaginal VK2/E6E7 cells in terms of

DNA repair pathways and cell cycle regulation reflect the unique

tissue tropism seen in gonococcal infections. Also, p53

expression may be differently regulated in tumor cell lines in

comparison to non-tumor cell lines. We suggest that the VK2/

E6E7-specific dys-regulation of p53 is a potential mechanism

that may be utilized by gonococci to keep the host cell alive

despite DNA damage. This observation is consistent with the

anti-apoptotic features of N. gonorrhoeae (Binnicker et al., 2004;

Howie et al., 2008).

In the regulation of the cell cycle, both p21 and p27 are potent

CDK inhibitors, and the induced expression and nuclear shuttling

of these proteins during a gonococcal infection could explain the

deceleration observed during our live-cell time-lapse microscopy

and cell proliferation assays, which clearly showed that the

infected cells were delayed in one or several phases of the cell

cycle. Using flow cytometry assays on synchronized cells, we

showed that this delay primarily occurs during G2 phase

progression, which results in an accumulation of infected cells

in G1 after 24 hours of infection. This correlates to previous data

by our group in which unsynchronized HeLa cells were

sequestered in early G1 after 24 hours of gonococcal infection

(Jones et al., 2007). Also, gonococcal infection in ME-180 cells

also leads to the induction of transforming growth factor beta

(TGFb) (Naumann et al., 1997). TGFb has been shown to possess

anti-proliferative properties and could further promote the

survival of gonococcal infections.

In summary, gonococci adhere to epithelial cells of the

urogenital tract, where they can persist for a long time and cause

damage to the reproductive system. The ability to invade and

multiply within the cell cytoplasm enables bacteria to evade

extracellular immune responses and find a protected niche in

which to survive within the epithelial cell layer. Prolonged local

bacterial existence and survival in the epithelial cell lining can be

achieved by slowing down the progression of the cell cycle,

leading to reduced epithelial renewal and exfoliation (Muenzner

et al., 2010) to the detriment of mucosal barrier fidelity.

Therefore, the protected niche created by gonococcal infections

is a favorable environment for bacteria, but its persistence also

Fig. 8. The decelerated cell cycle progression in infected cells

is due to sequestration in G2. A DNA profile of 10,000 cells

was generated by flow cytometry and analyzed. The assay was

performed twice in triplicate. The percentages of infected and

control cells in G1, S and G2 phases of the cell cycle for

(A) VK2E6E7 cells and (B) ME-180 cells at 0, 6, 12, 18 or

24 hours post-infection are shown. Data from the 24-hour time-

point was repeated an additional time for statistical analysis

where the G1 cell-fraction was compared with the total amount

of cells gated (10,000 cells) in a paired 2-tailed Student’s t-test

(* indicates that P,0.05).

Journal of Cell Science 126 (1)344

Journ

alof

Cell

Scie

nce

causes DNA damage to the reproductive tract. Undoubtedly, our

findings will prove useful in further dissecting the pathways

involved in microbial pathogenesis and contribute to a greater

understanding of whether N. gonorrhoeae infections could

predispose host cellular malignancies.

Materials and MethodsCell lines and growth conditions

VK2/E6E7 (ATCC CRL-2616, LGC Standards, London) non-tumor epithelialcells, originating from human vaginal mucosa, were cultured in keratinocyte

serum-free medium supplemented with 0.1 ng/ml human recombinant epidermal

growth factor (Invitrogen, Carlsbad, CA, USA), 0.05 mg/ml bovine pituitary

extract (Invitrogen, Carlsbad, CA, USA), and 44.1 mg/l calcium chloride (Sigma-Aldrich Inc., St. Louis, MO, USA). The ME-180 (ATCC HBT-33, LGC Standards,

London) human cervical epithelial-like adenocarcinoma cell line was cultured in

Dulbecco’s modified Eagle medium (DMEM) containing GlutaMAX (Invitrogen,

Carlsbad, CA, USA) and supplemented with 10% fetal calf serum (FCS)(Invitrogen, Carlsbad, CA, USA). Cells were maintained at 37 C in 5% CO2. In

all assays, a monolayer of low-confluence cells was used. Cells were never grown

to high confluence to avoid disturbing the progression of the cell cycle.

Bacterial strain

N. gonorrhoeae MS11 (P+) has been described (Swanson et al., 1987). Bacteria

were grown on gonococcal medium base GCB (NEOGEN, Lansing, MI, USA)

agar plates containing Kellogg’s supplement (Kellogg et al., 1968) at 37 C in 5%CO2. Piliated, non-opaque phenotypes were distinguished by morphology under a

binocular light microscope and used in all experiments. For the infection assays,

18- to 20-hour-old piliated bacteria were collected from GCB agar plates and re-suspended in DMEM. The optical density was measured at 600 nm to calculate the

number of bacteria/ml, and throughout the study, a multiplicity of infection (m.o.i.)

of 50 bacteria/cell was used.

Alkaline DNA unwinding assay

Cells were sub-cultured in a 75-cm2 flask and labeled with 0.1 ml/ml 3H-TdR (37

MBq/ml, Perkin Elmer, San Jose, CA, USA) in DMEM/10% FCS for one day.

Cells were then seeded in a 24-well plate and incubated overnight for furtherlabeling. The labeling medium was replaced with fresh DMEM/10% FCS

following three washes with PBS. For the c irradiation standard curve, the cells

were maintained on ice and irradiated with 137Cs c rays at a dose rate of 0.4 Gy/

min (Gammacell 1000, Scanditronix, Ottawa, Canada), directly followed by DNAunwinding as described below.

For the infection assays, the bacteria were added to the non-confluent monolayer

of labeled cells (m.o.i. of 50) and allowed to adhere for 2 hours, after which theunbound bacteria were removed and the cells incubated further. Cells were

infected with gonococci for a total of either 6 or 24 hours. To increase the

sensitivity, DNA repair inhibitors were added for the last hour of the early time-point and for the last four hours of the later time-point to both infected cells and

control cells. The poly ADP ribose polymerase (PARP) inhibitor 4-amino-1,8-

napthalimide (4ANI) (Acros Organics, Geel, Belgium) was added to VK2/E6E7

cells to a final concentration of 9 mM to prevent the sealing of strand breaks duringbase excision repair. For the inhibition of DNA incorporation during nucleotide

excision repair and the consequent accumulation of strand-break intermediates,

2 mM hydroxyurea (HU) (Sigma-Aldrich, St. Louis, MO, USA) and 20 mM

cytosine arabinoside (AraC) (Sigma-Aldrich, St. Louis, MO, USA) were added toME-180 cells. Following an incubation of 1 or 4 hours with inhibitors, the cells

were washed three times with ice-cold PBS followed by 0.5 ml/well of DNA

unwinding solution (0.03 M NaOH in 1 M NaCl). The DNA unwinding solution

was added to the wells for 30 minutes in a dark box at room temperature. DNAunwinding was stopped with a forceful injection of 1 ml/well of 0.02 M NaH2PO4,

and the DNA was sheared by sonication for 15 seconds. After the addition of 50 ml

of 7.5% sodium dodecyl sulphate (SDS), the samples were stored at 220 C until

elution. Thawed samples were separated into single-strand and double-strand DNAfractions by hydroxyapatite chromatography as previously described (Erixon and

Ahnstrom, 1979).

A c irradiation standard curve was made for both cell lines (supplementary

material Fig. S2A) and the number of induced strand breaks was calculated. Cells

were irradiated with increasing doses of c rays followed by the same ADU assay.

Based on the well-established statement that 1 Gy of c irradiation causes 1040strand breaks per cell, the linear correlation between the 2LOG values of dsDNA

(total DNA-ssDNA) (supplementary material Fig. S2B) allowed us to quantify the

number of strand breaks induced by the bacterial infection. Since the DNA repair

inhibitors were present for 4 hours during the 24-hour time-points, the values weredivided by a factor of 4 to generate an average of the number of strand breaks per

hour.

Quantitative real time PCR analysis

Bacteria were added to the non-confluent monolayer of VK2/E6E7 cells (m.o.i. of50) and allowed to adhere for 2 hours, after which the unbound bacteria wereremoved and the cells incubated further. Cells were infected with gonococcifor a total of 24 hours. RNA from control cells and infected VK2/E6E7 cellswas isolated with the RNeasy Mini kit according to the manufacturer’srecommendations (Qiagen, Hilden, Germany). Total RNA (up to 1 mg) wasreverse transcribed using Superscript III First strand synthesis super mix(Invitrogen, Carlsbad, CA, USA), using oligo-dT primers according to themanufacturer’s recommendations. Primers used were TP53 forward: 59-CCA TGTGCT CAA GAC TGG CGC T-39, TP53 reversed: 59-GCA GTC TGG CTG CCAATC CA-39, TUBA1A forward: 59-ACA TCG ACC GCC TAA GAG TCG C-39,TUBA1A reversed: 59-TGC ACT CAC GCA TGG TTG CTG-39. PCRamplification was performed using a Light Cycler 480 (Roche, Basel,Switzerland) and SYBR Green I Master kit (Roche, Basel, Switzerland) with0.1 mM of each gene specific primer and 5 ml of cDNA. PCR program was asfollows: initial denaturation for 300 s, amplification for 30 cycles withdenaturation at 94 C for 30 s, annealing at 55 C for 15 s and extension at 71 Cfor 30 s. The transition rate was 2.2–4.4 C/s. Relative expression of mRNA of thetarget genes were calculated, normalized, and compared between control cells andinfected cells, using alpha tubulin (TUBA1A) as an internal standard andGlyceraldehyde 3-phosphate dehydrogenase (GAPDH) as an additional standard.

Immunofluorescence

Epithelial cells were sub-cultured on poly-D-lysine-coated 0.17-mm coverslips(Gerhard Mentzel GmbH, Braunschweig, Germany). Cells were infected and, atthe given time-points, fixed for 10 minutes in 3.7% paraformaldehyde (PFA).Cells were permeabilized for 10 minutes in 20 mM Tris-HCl pH 7.4, 50 mMNaCl, 3 mM MgCl2, 0.5% Triton X-100, and 300 mM sucrose and blocked with1% skim milk powder (SMP) for 1 hour at room temperature. The followingprimary antibodies were used: rabbit anti-p21 (H-164, Santa Cruz BiotechnologyInc., Santa Cruz, CA, USA, 1:500 dilution), mouse anti-p27kip1 (BD Biosciences,San Jose, CA, USA, 1:500 dilution), mouse anti-p53 (DO-1, Santa CruzBiotechnology Inc., Santa Cruz, CA, USA, 1:500 dilution), rabbit anti-53BP1(Nordic Biosite, Taby, Sweden, 1:500 dilution), and mouse anti-phosphorylatedH2AX serine 139 (cH2AX, clone JBW301, Merck Millipore, Billerica, MA, USA,dilution 1:200). Secondary antibodies included goat anti-rabbit IgG and goat anti-mouse IgG conjugated to either Alexa Fluor 488 nm or Alexa Fluor 633 nm(Molecular Probes, Life technologies, Paisley, UK) and were used at a dilution of1:500. Coverslips were mounted in Vectashield containing 49-6-Diamidino-2-phenylindole (DAPI, Vector laboratories, Burlingame, CA, USA). Fluorescentimages of infected and control cells were captured with a CCD camera connectedto an inverted fluorescence microscope (Cell Observer, Carl Zeiss, GmbH,Gottingen, Germany). Images were further processed using ImageJ (NIH,Bethesda, MA, USA) and Photoshop CS5 (Adobe).

SDS-PAGE and western blotting

Infected and control cells were washed three times with PBS and harvested in0.2% Igepal (Sigma-Aldrich Inc., St. Louis, MO, USA)/16protease inhibitorcocktail (Roche, Mannheim, Germany)/PBS. Cell lysates were resolved in samplebuffer (87% Glycerol, 1.5 M Tris-HCl pH 6.8, 10% 2-mercaptoethanol, and 0.1%SDS) and boiled for 10 minutes at 95 C. SDS-PAGE was performed according tothe manufacturer’s protocol (Bio-Rad, Hercules, CA, USA). Protein gels wereblotted onto Imobilon-FL PVDF membranes (Merck Millipore, Billerica, MA,USA) according to the manufacturer’s protocol (Bio-Rad, Hercules, CA, USA).Membranes were blocked for 1 hour in 5% SMP and incubated with primaryantibodies overnight at 4 C with agitation. The following antibodies were used forimmunoblotting: rabbit anti-p21 (H-164, Santa Cruz Biotechnology Inc., SantaCruz, CA, USA, 1:200 dilution), mouse anti-p27kip1 (BD Biosciences, 1:200dilution), and anti-p53 (DO-1, Santa Cruz Biotechnology Inc., Santa Cruz, CA,USA, 1:200 dilution). Secondary antibodies included goat anti-rabbit IgG and goatanti-mouse IgG conjugated to IRdye800CW (Li-COR, Lincoln, Nebraska, USA) orIRdye680 and were used at a dilution of 1:10,000. The membranes were visualizedand analyzed using an Odyssey IR scanner (Li-COR, Lincoln, Nebraska, USA) at700 or 800 nm. The immunoblotting band intensities were quantified using ImageJ(NIH, Bethesda, MA, USA). A polyclonal antibody raised against a-tubulin(MBS316320, Mybiosource, 1:1000 dilution) was used for normalizing the totalamount of protein loaded in each well.

Live-cell time-lapse imaging

Cell monolayers were grown in 12-well poly-D-lysine-coated glass bottom dishes(MatTek corp. Ashland, MA, USA) to 30–40% confluence. Cells were washedonce in PBS before new media and bacteria were added (m.o.i. of 50). After2 hours, the unbound bacteria were carefully washed away, and the dish wastransferred to a live-cell incubator at 37 C in 5% CO2 connected to an invertedfluorescence microscope (Cell Observer, Carl Zeiss, Gottingen, Germany). Afteran additional 1-hour acclimatization period in the microscope, randomly selectedfields-of-view (containing 10–20 cells) were observed for 21 hours with a 206

Gonococcal infection causes DNA damage 345

Journ

alof

Cell

Scie

nce

objective. Differential interference contrast (DIC) images were captured every10 minutes for each of the positions chosen. Data were collected from threeindependent experiments, and 150 cells were monitored during each experiment.Images were further processed using ImageJ (NIH, Bethesda, MA, USA) andPhotoshop CS5 (Adobe).

BrdU assay

Non-confluent cells were incubated for 24 hours with N. gonorrhoeae (m.o.i. of50). Bromodeoxyuridine (BrdU) was added 2 hours prior to the termination of theassay to allow the nucleoside to be incorporated into the replicating DNA of cellsin S phase. We quantified the incorporation of BrdU into S-phase cells by ELISA(QIA58, Calbiochem, San Diego, CA, USA) according to the manufacturer’sinstructions. Absorbance was measured at 450 nm. BrdU proliferation assays wererepeated at least four times, and each sample was assayed at least in triplicate. Inthe assay belonging to supplementary material Fig. S3, we collected supernatantswith three different experimental set ups: (I) The supernatant from a 24-hour ME-180 cell culture, (II) the supernatant from a 24-hour infection, and (III) thesupernatant from a bacterial culture grown in cell culture media. All three differentsupernatants were collected, sterile filtrated and 0.1 ml was added to a new set ofME-180 cells (maintained in 0.1 ml DMEM/10%FCS) for 24 hours. Also here,BrdU was added 2 hours prior to the termination.

Flow cytometry

Cells were first synchronized using 2.5 mM thymidine for 28 hours at 37 C and5% CO2. The cells were subsequently released from the block by substitutingDMEM for thymidine for 12 hours. Then, the cells were again blocked with2.5 mM thymidine for 28 hours at 37 C and 5% CO2. One hour after this secondrelease, infections were initiated by adding N. gonorrhoeae (m.o.i. of 50) to thenon-confluent cells. The unbound bacteria were removed after 2 hours ofincubation, and samples were collected after a total infection time of 0, 6, 12,18 or 24 hours. Cells were harvested by trypsinization (0.2% trypsin-EDTA,Invitrogen, Carlsbad, CA, USA) and collected by centrifugation at 800 g for10 minutes at 4 C. The supernatant was decanted, and the cells were fixed in ice-cold 70% ethanol at 4 C. Prior to analysis, the cells were washed in PBS, treatedwith RNase (40 mg/ml, Sigma-Aldrich, St. Louis, MO, USA), and stained withpropidium iodide (40 mg/ml, BD Biosciences, Franklin Lakes, NJ, USA) for30 minutes at 37 C. DNA profiles were generated by flow cytometry (FACSCalibur), and the data were analyzed using Cell Quest Pro (BD Biosciences,Franklin Lakes, NJ, USA). Flow cytometry experiments were performed twice intriplicate, and 10,000 cells were recorded from each sample. The 24-hour time-point was recorded three times to permit statistical analysis.

Statistical analyses

All experiments were performed at least in triplicate unless otherwise stated. Inwestern blot and qPCR values were adjusted using an internal control, alphatubulin. Data were analysed using a paired 2-tailed Student’s t-test. Error barsrepresent s.e.m.

FundingThis work was supported by Cancerfonden [grant number CAN2011/671]; Stockholm University; Carl Tryggers Stiftelse; AkeWibergs Stiftelse; Harald Jeanssons Stiftelse; Petrus och AugustaHedlunds Stiftelse and Granholms Stiftelse.

Supplementary material available online at

http://jcs.biologists.org/lookup/suppl/doi:10.1242/jcs.117721/-/DC1

ReferencesBinnicker, M. J., Williams, R. D. and Apicella, M. A. (2003). Infection of human

urethral epithelium with Neisseria gonorrhoeae elicits an upregulation of host anti-apoptotic factors and protects cells from staurosporine-induced apoptosis. Cell.

Microbiol. 5, 549-560.

Binnicker, M. J., Williams, R. D. and Apicella, M. A. (2004). Gonococcal porin IBactivates NF-kappaB in human urethral epithelium and increases the expression ofhost antiapoptotic factors. Infect. Immun. 72, 6408-6417.

Cleaver, J. E. (2011). cH2Ax: biomarker of damage or functional participant in DNArepair ‘‘all that glitters is not gold!’’. Photochem. Photobiol. 87, 1230-1239.

Cuevas-Ramos, G., Petit, C. R., Marcq, I., Boury, M., Oswald, E. and Nougayrede,

J. P. (2010). Escherichia coli induces DNA damage in vivo and triggers genomicinstability in mammalian cells. Proc. Natl. Acad. Sci. USA 107, 11537-11542.

Deva, R., Sczepanski, M., Saha, S., Zafiriou, M. P., Kerstin, N., Jakobsson, P. J.,

Inoue, Y. and Nigam, S. (2010). C. albicans activates cyclooxygenase but not itsproduct prostaglandin E2 in HPV 16-stabilized cells. Eur. J. Obstet. Gynecol. Reprod.

Biol. 152, 205-209.

Erixon, K. and Ahnstrom, G. (1979). Single-strand breaks in DNA during repair ofUV-induced damage in normal human and xeroderma pigmentosum cells as

determined by alkaline DNA unwinding and hydroxylapatite chromatography: effectsof hydroxyurea, 5-fluorodeoxyuridine and 1-beta-D-arabinofuranosylcytosine on thekinetics of repair. Mutat. Res. 59, 257-271.

Fichorova, R. N., Desai, P. J., Gibson, F. C., 3rd and Genco, C. A. (2001). Distinctproinflammatory host responses to Neisseria gonorrhoeae infection in immortalizedhuman cervical and vaginal epithelial cells. Infect. Immun. 69, 5840-5848.

Follows, S. A., Murlidharan, J., Massari, P., Wetzler, L. M. and Genco, C. A.

(2009). Neisseria gonorrhoeae infection protects human endocervical epithelial cellsfrom apoptosis via expression of host antiapoptotic proteins. Infect. Immun. 77, 3602-3610.

Friedberg, E. C. (1995). Out of the shadows and into the light: the emergence of DNArepair. Trends Biochem. Sci. 20, 381.

Gartel, A. L. and Tyner, A. L. (2002). The role of the cyclin-dependent kinase inhibitorp21 in apoptosis. Mol. Cancer Ther. 1, 639-649.

Howie, H. L., Shiflett, S. L. and So, M. (2008). Extracellular signal-regulated kinaseactivation by Neisseria gonorrhoeae downregulates epithelial cell proapoptoticproteins Bad and Bim. Infect. Immun. 76, 2715-2721.

Hu, L. and Kopecko, D. J. (1999). Campylobacter jejuni 81-176 associates withmicrotubules and dynein during invasion of human intestinal cells. Infect. Immun. 67,4171-4182.

Jacinto, A., Martinez-Arias, A. and Martin, P. (2001). Mechanisms of epithelialfusion and repair. Nat. Cell Biol. 3, E117-E123.

Jones, A., Jonsson, A. B. and Aro, H. (2007). Neisseria gonorrhoeae infection causes aG1 arrest in human epithelial cells. FASEB J. 21, 345-355.

Kallstrom, H., Liszewski, M. K., Atkinson, J. P. and Jonsson, A. B. (1997).Membrane cofactor protein (MCP or CD46) is a cellular pilus receptor for pathogenicNeisseria. Mol. Microbiol. 25, 639-647.

Kallstrom, H., Islam, M. S., Berggren, P. O. and Jonsson, A. B. (1998). Cell signalingby the type IV pili of pathogenic Neisseria. J. Biol. Chem. 273, 21777-21782.

Kallstrom, H., Hansson-Palo, P. and Jonsson, A. B. (2000). Cholera toxin andextracellular Ca2+ induce adherence of non-piliated Neisseria: evidence for animportant role of G-proteins and Rho in the bacteria-cell interaction. Cell. Microbiol.

2, 341-351.

Kellogg, D. S., Jr, Cohen, I. R., Norins, L. C., Schroeter, A. L. and Reising,

G. (1968). Neisseria gonorrhoeae. II. Colonial variation and pathogenicity during 35months in vitro. J. Bacteriol. 96, 596-605.

Koomey, M. (2001). Implications of molecular contacts and signaling initiated byNeisseria gonorrhoeae. Curr. Opin. Microbiol. 4, 53-57.

Kundu, J. K. and Surh, Y. J. (2008). Inflammation: gearing the journey to cancer.Mutat. Res. 659, 15-30.

Lara-Tejero, M. and Galan, J. E. (2000). A bacterial toxin that controls cell cycleprogression as a deoxyribonuclease I-like protein. Science 290, 354-357.

Lindahl, T. and Barnes, D. E. (2000). Repair of endogenous DNA damage. Cold

Spring Harb. Symp. Quant. Biol. 65, 127-134.

Lofmark, S., de Klerk, N. and Aro, H. (2011). Neisseria gonorrhoeae infection inducesaltered amphiregulin processing and release. PLoS ONE 6, e16369.

Merz, A. J., Rifenbery, D. B., Arvidson, C. G. and So, M. (1996). Traversal of apolarized epithelium by pathogenic Neisseriae: facilitation by type IV pili andmaintenance of epithelial barrier function. Mol. Med. 2, 745-754.

Merz, A. J., Enns, C. A. and So, M. (1999). Type IV pili of pathogenic Neisseriae elicitcortical plaque formation in epithelial cells. Mol. Microbiol. 32, 1316-1332.

Michaud, D. S. (2007). Chronic inflammation and bladder cancer. Urol. Oncol. 25, 260-268.

Muenzner, P., Bachmann, V., Zimmermann, W., Hentschel, J. and Hauck, C. R.

(2010). Human-restricted bacterial pathogens block shedding of epithelial cells bystimulating integrin activation. Science 329, 1197-1201.

Naumann, M., Wessler, S., Bartsch, C., Wieland, B. and Meyer, T. F. (1997).Neisseria gonorrhoeae epithelial cell interaction leads to the activation of thetranscription factors nuclear factor kappaB and activator protein 1 and the inductionof inflammatory cytokines. J. Exp. Med. 186, 247-258.

Nias, A. H. W. (1998). An Introduction to Radiobiology. p. 69. London, UK: John Wiley& Sons Ltd.

Nikjoo, H., O’Neill, P., Terrissol, M. and Goodhead, D. T. (1994). Modelling ofradiation-induced DNA damage: the early physical and chemical event. Int. J. Radiat.

Biol. 66, 453-457.

Nougayrede, J. P., Taieb, F., De Rycke, J. and Oswald, E. (2005). Cyclomodulins:bacterial effectors that modulate the eukaryotic cell cycle. Trends Microbiol. 13, 103-110.

Oswald, E., Nougayrede, J. P., Taieb, F. and Sugai, M. (2005). Bacterial toxins thatmodulate host cell-cycle progression. Curr. Opin. Microbiol. 8, 83-91.

Popp, A., Billker, O. and Rudel, T. (2001). Signal transduction pathways induced byvirulence factors of Neisseria gonorrhoeae. Int. J. Med. Microbiol. 291, 307-314.

Rodrıguez-Vilarrupla, A., Dıaz, C., Canela, N., Rahn, H. P., Bachs, O. and Agell,

N. (2002). Identification of the nuclear localization signal of p21(cip1) andconsequences of its mutation on cell proliferation. FEBS Lett. 531, 319-323.

Russo, T., Zambrano, N., Esposito, F., Ammendola, R., Cimino, F., Fiscella,

M., Jackman, J., O’Connor, P. M., Anderson, C. W. and Appella, E. (1995). Ap53-independent pathway for activation of WAF1/CIP1 expression followingoxidative stress. J. Biol. Chem. 270, 29386-29391.

Scanlon, M., Shaw, A. P., Zhou, C. J., Visvesvara, G. S. and Leitch, G. J. (2000).Infection by microsporidia disrupts the host cell cycle. J. Eukaryot. Microbiol. 47,525-531.

Journal of Cell Science 126 (1)346

Journ

alof

Cell

Scie

nce

Sherr, C. J. and Roberts, J. M. (1995). Inhibitors of mammalian G1 cyclin-dependentkinases. Genes Dev. 9, 1149-1163.

Sutcliffe, S., Giovannucci, E., De Marzo, A. M., Leitzmann, M. F., Willett, W. C.and Platz, E. A. (2006a). Gonorrhea, syphilis, clinical prostatitis, and the risk ofprostate cancer. Cancer Epidemiol. Biomarkers Prev. 15, 2160-2166.

Sutcliffe, S., Zenilman, J. M., Ghanem, K. G., Jadack, R. A., Sokoll, L. J., Elliott,D. J., Nelson, W. G., De Marzo, A. M., Cole, S. R., Isaacs, W. B. et al. (2006b).Sexually transmitted infections and prostatic inflammation/cell damage as measuredby serum prostate specific antigen concentration. J. Urol. 175, 1937-1942.

Swanson, J. (1973). Studies on gonococcus infection. IV. Pili: their role in attachment ofgonococci to tissue culture cells. J. Exp. Med. 137, 571-589.

Swanson, J., Robbins, K., Barrera, O. and Koomey, J. M. (1987). Gene conversionvariations generate structurally distinct pilin polypeptides in Neisseria gonorrhoeae.J. Exp. Med. 165, 1016-1025.

Toller, I. M., Neelsen, K. J., Steger, M., Hartung, M. L., Hottiger, M. O., Stucki, M.,Kalali, B., Gerhard, M., Sartori, A. A., Lopes, M. et al. (2011). Carcinogenicbacterial pathogen Helicobacter pylori triggers DNA double-strand breaks and a DNAdamage response in its host cells. Proc. Natl. Acad. Sci. USA 108, 14944-14949.

Vermeulen, K., Van Bockstaele, D. R. and Berneman, Z. N. (2003). The cell cycle: a reviewof regulation, deregulation and therapeutic targets in cancer. Cell Prolif. 36, 131-149.

Ward, I. M., Minn, K., Jorda, K. G. and Chen, J. (2003). Accumulation of checkpointprotein 53BP1 at DNA breaks involves its binding to phosphorylated histone H2AX.J. Biol. Chem. 278, 19579-19582.

Wei, J., Nagy, T. A., Vilgelm, A., Zaika, E., Ogden, S. R., Romero-Gallo,J., Piazuelo, M. B., Correa, P., Washington, M. K., El-Rifai, W. et al. (2010).Regulation of p53 tumor suppressor by Helicobacter pylori in gastric epithelial cells.Gastroenterology 139, 1333-1343.

Gonococcal infection causes DNA damage 347