Application of Small Interfering RNA

Application of Small Interfering RNA for Inhibition ofLipopolysaccharide-induced Osteoclast Formation andCytokine StimulationFarshid S. Fahid, DDS, Jin Jiang, DDS, PhD, Qiang Zhu, DDS, PhD, Cuiping Zhang,Elizabeth Filbert, DMD, Kamran E. Safavi, DMD, MEd, and Larz S. W. Spångberg, DDS, PhD

Abstract

RNA interference (RNAi) is a unique and powerful toolused for the study of gene function by suppressing itsexpression. Nuclear factor of activated T cells (NFATc1)is the most strongly induced transcription factor medi-ated by receptor activator for nuclear factor kappa Bligand stimulation and has shown to be a key regulatorof osteoclastogenesis. To determine the application ofsmall interfering RNA (siRNA) for inhibition of lipopoly-saccharide (LPS)-induced cytokine stimulation and os-teoclast formation, murine monocyte, RAW 264.7 cellsas well as differentiated osteoclasts were transfectedwith NFATc1-specific siRNA and then stimulated with100 ng/mL LPS. By using real-time polymerase chainreaction analysis and enzyme-linked immunosorbentassay, we confirmed that monocytes whose NFATc1protein expression was silenced by using RNAi pro-duced lower levels of inflammatory cytokines, fewernumbers evolved into mature osteoclasts, and oste-oclasts expressed lower levels of osteoclast-specificgene markers such as tartrate-resistant acid phospha-tase and cathepsin K. These results suggested thatRNAi could be used to modulate the effects of LPSstimulation. (J Endod 2008;34:563–569)

Key Words

Lipopolysaccaride, NFATc1, osteoclast, RNA interfer-ence

The importance of the role played by bacteria in the pathogenesis of pulpal andperiapical disease has been established by numerous studies (1–3). Pulpal infec-

tions initially produce an inflammatory response within the pulp that often leads tocomplete pulpal necrosis and subsequently in the periapical region, which results inlesion formation with local bone destruction (4). The bacteria from infected root canalshave been extensively studied and shown to be predominantly gram-negative and strictlyanaerobic (5, 6).

Bacterial lipopolysaccharide (LPS) is an endotoxin and a major component of theouter membrane of gram-negative bacteria. It is also one of the most potent microbialinitiators of inflammation and endodontic pathogenesis (7, 8). A positive correlationhas been reported between the levels of LPS in root canals and the presence of peri-apical lesions (9, 10). The level of LPS in the root canals and periapical tissues of ratshas also been shown to increase from 1 to 70 days after exposure of the pulp to oralmicroorganisms (11).

Osteoclasts, which are bone-resorbing multinucleated giant cells, originate fromhematopoietic progenitors of the moncyte/macrophage lineage. Once committed,mononuclear cells fuse with each other to form multinucleated osteoclasts. They thenpolarize and adhere to bone matrix via a “sealing zone,” induce actin ring formation,acidify the bone surface, and release osteolytic enzymes. Osteoclastic bone resorptionis a predominant feature in periapical lesion formation. In normal bone homeostasis,bone is maintained through a modeling and remodeling process throughout life. Thisprocess involves bone resorption via osteoclasts and bone formation via osteoblasts.Receptor activator for nuclear factor kappa B ligand (RANKL), a novel member of thetumor necrosis factor (TNF) superfamily that is expressed by stromal cells, binds to itsfunctional receptor, RANK, on osteoclasts to induce osteoclastogenesis. RANKL/RANKsignaling plays an essential role in osteoclast differentiation, activation, and survival, asconfirmed by the osteopetrotic phenotypes of RANKL and RANK knockout mice(12–14).

Previous studies have demonstrated that LPS can directly stimulate osteoblasts toexpress RANKL, resulting in osteoclast formation (15). Recent findings have also indi-cated that LPS might be directly involved in osteoclast differentiation through a sharedmechanism, partially independent from the RANK/RANKL interaction (16, 17). Thetranscription factor, nuclear factor of activated T cells (NFATc1), is the most stronglyinduced transcription factor mediated by RANKL stimulation. It has also been shown tobe a key regulator of osteoclastogenesis and plays a critical role in the terminal differ-entiation of osteoclasts (18). At the final stage of osteoclast differentiation, NFATc1works to stimulate osteoclast-specific genes such as tartrate-resistant acid phosphatase(TRAP), calcitonin receptor, and cathepsin K (19). Thus, it appears that the NFATc1pathway is a crucial component of osteoclast differentiation, and inhibition of thispathway might provide a potential therapeutic approach for the treatment of bonediseases.

RNA interference (RNAi) is emerging as a new and potent method of gene silencingthat selectively targets and shuts off the post-transcriptional expression of mRNA (20).RNAi is induced when the host cell encounters long double-stranded RNA (dsRNA) fromviruses or exogenously administered synthetic siRNA, which are commercially availablefor target genes of interest. Once in the cytoplasm, they undergo processing by an

From the Division of Endodontology, School of DentalMedicine, University of Connecticut Health Center, Farming-ton, Connecticut.

Address requests for reprints to Dr Jin Jiang, Division ofEndodontology, MC 1715, University of Connecticut HealthCenter, 263 Farmington Ave, Farmington, CT 06030-1715.E-mail address:[email protected]/$0 - see front matter

Copyright © 2008 by the American Association ofEndodontists.doi:10.1016/j.joen.2008.01.024

Basic Research—Biology

JOE — Volume 34, Number 5, May 2008 Application of siRNA for Inhibition of LPS Effects 563

enzyme called Dicer. This results in the formation of a dsRNA moleculebetween 21–23 nucleotides in length, called small interfering RNA(siRNA), which are incorporated into an RNA-induced silencing com-plex (RISC). RISC then directs degradation of RNA containing a homol-ogous sequence, effectively silencing target gene expression as shown inFig. 1 (21). Nature publishing group provides an animation on themechanism of RNAi (http://www.nature.com/focus/rnai/animations/index.html).

In an effort to determine the application of siRNAs for the inhibi-tion of LPS-induced osteoclast formation and cytokine stimulation, theobjective of this study was to suppress NFATc1 expression in monocytesand osteoclast cells by using the RNAi technique. We hypothesized thatknocking down the expression of NFATc1 in these cells with effectivesiRNA will modulate the effects of LPS stimulation.

Materials and MethodsPreparation of Osteoclasts

Osteoclast-like cells (OCL) were differentiated from RAW 264.7cells, a mouse hematopoietic cell line (American Type Culture Collec-tion, Rockville, MD). RAW 264.7 cells were plated at a density of 40,000cells/well in 24-well plates in alpha modified Eagle medium ( MEM)with 10% fetal bovine serum (FBS) (Invitrogen, Carlsbad, CA). The cellswere incubated at 37°C in a humidified atmosphere containing 5% CO2

and stimulated cells with 50 ng/mL RANKL for the first 24 hours and 100ng/mL LPS (Escherichia coli O26:B6; Sigma, St Louis, MO) for anadditional 72 hours as previously described (17).

siRNA Transfection

NFATc1 was silenced by using NFATc1 SMARTPOOL siRNA reagent(Dharmacon, Chicago, IL).

For assessment of cytokine expression, RAW264.7 cells wereplated on 24-well plates with Dulbecco modified Eagle medium(DMEM) and 10% fetal calf serum (FCS) at density 40,000/well for 24hours. Cells were then transfected by using 100 nmol/L control or

experimental siRNA and combined with 2 �L lipofectamine 2000 (In-vitrogen) in Opti-MEM supplemented medium. Six hours after transfec-tion, the medium was replaced by DMEM with FBS. The cells wereincubated at 37°C in a CO2 incubator for additional 4 days. At last 24hours of culture, cells were stimulated with 100 ng/mL LPS. The mediawere collected and stored in 70°C for enzyme-linked immunosorbentassay (ELISA) analysis.

For osteoclast examination, RANKL-stimulated RAW 264.7 cells onglass coverslips in 24-well plates for first 24 hours were transfected byusing 100 nmol/L control or experimental siRNA combined with 2 �Llipofectamine 2000 (Invitrogen) in Opti-MEM supplemented medium.Six hours after transfection, the medium was replaced by MEM withFBS and 100 ng/mL LPS. The cells were incubated at 37°C in a CO2

incubator for additional 4 days, after which they were fixed in 2% para-formaldehyde. The nuclei were then counterstained with a DAPI stain(Sigma), which binds the nucleotides, to visualize the relative locationof the siRNA. Cells with uptake of the fluorescent-labeled siRNA wereidentified by fluorescent microscopy. There was no significant differ-ence in vitality between cells transfected with NFATc1 and control siRNA.

ELISA

Concentrations of interleukin-6 (IL-6) and TNF- in culture su-pernatants were determined by ELISA in triplicate with commercialELISA Duo systems (R&D Systems, Minneapolis, MN), according to therespective manufacturer’s instructions. For each sample and assay, themean of the triplicate measurement was calculated.

Immunocytochemistry

RAW 264.7–derived osteoclast-like cells were fixed in 2% para-formaldehyde in phosphate-buffered saline (PBS) on ice for 20 min-utes. Osteoclasts were detergent-permeabilized with 0.2% Triton X-100in PBS for 10 minutes, washed, and blocked in PBS with 2% bovineserum albumin for 2 hours. The cells were stained with antibodiesrecognizing NFATc1 (Santa Cruz Biotechnology, Santa Cruz, CA) at a

Synthetic siRNA designed to

target specific gene

Within cells, siRNA unwinds and

is corporated into RISC to target

specific mRNA

This mRNA is cleaved and

degraded, resulting in the

loss of protein expression

of targeted gene.

siRNA introduced into mammalian cells

RISCmRNA

Single strand siRNA

Target mRNA

Destroy mRNA

Synthetic siRNA

Figure 1. Mechanism of siRNA-induced gene silencing.


564 Fahid et al. JOE — Volume 34, Number 5, May 2008

dilution of 1:100 in PBS for 2 hours. The slides were washed in PBS andincubated for 40 minutes with donkey anti-rabbit Cy3 at 1:200 dilution.Slides were washed with PBS, dried, and mounted in Vectashield withDAPI (Vector Laboratories, Burlingame, CA) to stain DNA and reducefluorescence fading. Preparations were examined with fluorescencemicroscope (Nikon, Tokyo, Japan). As a negative control, the prepara-tions were treated with nonimmunized immunoglobulin G, or the pri-mary antibody was omitted. Neither of these control methods producedmarked staining.

Tartrate-Resistant Acid Phosphatase Assay

At 96 hours of culture, cells were fixed with 2% paraformaldehyde,washed with PBS, and treated for 20 minutes with 0.2% Triton X-100solution to permeabilize cell membranes. Cytochemical staining of tar-trate-resistant acid phosphatase (TRAP)–positive cells was performedas described previously (17). TRAP-positive cells appeared dark red.Only TRAP-positive cells with more than 3 nuclei were counted. The

values are expressed as means ! standard deviations of triplicatecultures.

RNA Extraction, Quantification, and Reverse Transcription

Total RNA was extracted by using TRIZOL reagent (Invitrogen) andphenol/choloroform according to manufacturer’s instructions. RNAwas dissolved in Tris– ethylenediaminetetraacetic acid (EDTA), pH 7.4,and concentration of RNA was determined by measuring the spectro-photometric absorbance at 260 nm. RNA was treated with DNAse I(Invitrogen) for 15 minutes followed by DNase I inactivation with 25mmol/L EDTA at 65°C to remove genomic DNA contamination. Reversetranscription was carried out in a 20-�L volume containing about 3�gof RNA, 1�L of 50 ng/�L random hexamers and 1�L annealing buffer,10�L 2X first-strand reaction mix, and 2�L superscript III/RNase OUTenzyme mix (Invitrogen) at 25°C for 10 minutes and then at 50°C for 50minutes.

A

B

C

D

E

F

Figure 2. Monocytes (A, B, and C) and osteoclasts (D, E, and F) transfected with control fluorescent-labeled siRNA and visualized by fluorescent microscopy. (A andD) Image of cells transfected with red fluorescent-labeled siRNA showing the distribution of siRNA. (B and E) Image of nuclei staining in the same cells. (C) Overlayimage of (A) and (B). (F) Overlay image of (D) and (E). Size bar is 100 �m.



Real-Time Polymerase Chain Reaction

Taqman real-time polymerase chain reaction (RT-PCR) wasperformed from 1 L of cDNA by using TaqMan Universal PCR MasterMix (Applied Biosystems, Foster City, CA) with 100-nmol/L primersand a 50-nmol/L probe. The Taqman RT-PCR was performed on aTaqman ABI 7500 sequence Detection System (Applied Biosys-tems). Unlabeled specific primers and the TaqMan MGB probes(6-FAM dye-labeled) for detecting the mouse TRAP gene, cathepsinK (CSK) gene, IL-6, and TNF- were used. A Taqman eukaryotic 18Sendogenous control kit was used for housekeeping gene control.Cycling conditions were as follows. After an initial hold of 2 minutesat 50°C and 10 minutes at 95°C, the samples were cycled 40 times at95°C for 15 seconds and 60°C for 1 minute. Each sample wasassayed in triplicate.

The comparative Ct method was applied to determine comparativeexpression levels between samples relative to control gene expression.To examine regulation by NFATc1 siRNA, the amplification thresholdcycle value (Ct) from the NFATc1 siRNA treated samples were sub-tracted from the control-treated sample cycle values ("Ct� Ct control Ct treated). The ratio was obtained by calculating the values obtainedfor gene of interest and the housekeeping gene 18S rRNA. The foldchange of test gene was determined as 2("Ct gene " Ct 18S).

Statistical Analysis

Data are expressed as mean values! standard deviations. Statis-tical significance of differences was determined by one-way analysis ofvariance and followed by post hoc test (Fisher protected least significantdifference. Differences were considered statistically significant atP # .05.

ResultsEvaluation of siRNA Transfection in Monocytes

Traditionally, it is difficult to manipulate gene expression in mono-cytes and especially osteoclasts because they are terminally differenti-ated cells. Monocyte cells were transfected with a fluorescently labeledcontrol siRNA by using a commercially available lipofectamine kit andvisualized by fluorescence microscopy 72 hours after transfection (Fig.2A). The nuclei were counterstained with a DAPI stain, which binds thenucleotides to visualize the relative location of the siRNA (Fig. 2B).

When the 2 images above are superimposed, the red dots aroundthe nuclei were noted, which indicates that the siRNA are delivered intothe cytoplasm with high efficency (Fig. 2C).

Evaluation of siRNA Transfection in Osteoclasts

Similarly, osteoclast-like cells that were differentiated by usingRANKL and LPS and tranfected with fluorescent-labeled siRNA (Fig. 2D)and the nuclei were stained with DAPI (Fig. 2E). When the 2 imagesabove are superimposed, it shows that osteoclasts are multinucleated(as indicated by the blue dots), and the fluorescent-labeled siRNA arehighly concentrated in the peripheral cytoplasm around the nucleus(Fig. 2F). When the slides were stained with TRAP, which is a purplestain and marker for osteoclast identification, TRAP� cells that arephenotypically identical to untransfectedosteoclasts were observed(data not shown).

Silencing NFATc1 Expression in Osteoclasts With siRNA

To examine the efficacy of siRNA in silencing NFATc1 protein ex-pression, RANKL-induced osteoclast cells were transfected with controlor NFATc1 siRNA. In the control group, osteoclast cells were transfectedwith control siRNA and immunocytologically stained with anti-NFATc1

A

C

B

D

Figure 3. Targeted inhibition of NFATc1 using gene-specific siRNA. Osteoclasts transfected with control (A) and NFATc1 (B) siRNA were immunostained for NFATc1and counterstained with DAPI (C and D). Size bar is 50 �m.



antibody (Fig. 3A). This method locates NFATc1 protein expressioninside the cell. Intense nuclear accumulation of NFATc1 protein in os-teoclasts was observed as expected, because NFATc1 is a transcriptionfactor and therefore should be located in the nucleus (Fig. 3A). In theexperimental group, osteoclasts were transfected with NFATc1-specificsiRNA and similarly stained with anti-NFATc1 antibody. A significantreduction of NFATc1 protein expression in the nucleus as comparedwith the control was noted (Fig. 3B).

When the nuclei were counterstained with DAPI and the imageswere superimposed, the nuclei were still present in both groups (Fig. 3C

and D), even though there was reduced NFATc1 protein expression inthe experimental group (Fig. 3D).

This indicates that partial silencing of the expression of NFATc1 inosteoclasts can be achieved by using NFATc1-specific siRNA. Similarresults were observed in monocytes transfected with NFATc1 siRNA(data not shown).

Biologic Effects of Silencing NFATc1 Expression in Monocytes

Because monocytes are the predominant producers of cytokinesin response to LPS stimulation, we then examined what effects silencingNFATc1 expression would have on TNF- and IL-6 production. Mono-cytes were transfected with either NFATc1-specific or control siRNA, andthe cells were stimulated with LPS for the last 24 hours of culture.Medium was collected after 24 hours and analyzed for cytokine pro-duction by using ELISA. Results are shown in Fig. 4. Analysis of variancerevealed that cells in the NFATc1-specific siRNA group showed a statis-tically significant resistance to LPS stimulation as indicated by the down-regulated TNF- and IL-6 levels compared with that in the control siRNAgroup (P# .05) (Fig. 4).

Biologic Effects of Silencing NFATc1 Expression on Osteoclast

Formation

Monocytes were stimulated with RANKL for 24 hours and thentransfected with either NFATc1-specific or control siRNA. Cells werethen stimulated with LPS for an additional 72 hours. TRAP� cells withmore than 3 nuclei were counted. Results are shown in Fig. 5. Analysisof variance revealed a statistically significant fewer number of oste-oclasts formed in cultures treated with NFATc1-specific siRNA as com-pared with that in the control siRNA group (P# .05) (Fig. 5).

Monocyte Cytokine Production

0

5

10

15

20

25

30

Control siRNA NFATc1 siRNA

Treatment Groups

IL-6

TNF-α

*

*

Cytokines (ng/ml)

Figure 4. Effect of inhibition of NFATc1 on cytokine release in monocytes on LPSstimulation. *Statistical significance relative to control, P# .05.

Number of Osteoclasts Formed

0

50

100

150

200

250

300

350

400

450

Control siRNA NFATc1 siRNA

Treatment Groups

TRAP+ Multinucleated

Cells/Well

Cultures transfected with control siRNA Cultures transfected with NFATc1 siRNA

*

Figure 5. Effect of inhibition of NFATc1 on osteoclast formation on LPS stimulation. *Statistical significance relative to control, P# .05. Size bar is 100 �m.



Biologic Effects of Silencing NFATc1 Expression on Osteoclast-

Specific Gene Expression

Osteoclast-like cells were differentiated by using RANKL for 24hours and then tranfected with either NFATc1-specific or control siRNA.Cells were then stimulated with LPS for an additional 72 hours. RNA wasextracted and reverse transcribed into cDNA, and RT-PCR was per-formed to determine comparative mRNA expression levels of CSK, TRAP,IL-6, and TNF- relative to control gene expression. The relative ex-pression fold change determined from the application curve of RT-PCRrevealed a statistically significant decrease of CSK, TRAP, IL-6, andTNF- mRNA levels in osteoclasts treated with NFATc1-specific siRNA ascompared with that in the control siRNA group (P# .05) (Fig. 6).

DiscussionRNA interference is a conserved biologic response by which dsRNA

induces the sequence-specific degradation of complementary mRNA,thereby silencing target gene expression. It has rapidly become themethod of choice for studies of gene function and gene silencing exper-iments. We chose to suppress the expression of NFATc1, which is atranscription factor involved in the mechanism of cytokine productionand osteoclast formation. NFATc1 is the most strongly induced tran-scription factor gene mediated by RANKL stimulation whose presencehas been shown to be required during the final stage of osteoclastogen-esis (18, 19).

It has been previously demonstrated that antisense sequence tar-geting of TLR4, which is the known receptor for LPS, inhibited TLR4expression and reduced TNF- release when RAW 264.7 cells werestimulated by LPS (22). Our ability to successfully transfect and silenceNFATc1 gene expression with specific siRNA in monocytes resulted in asignificant reduction of LPS-induced TNF- and IL-6 production andfewer numbers of mature osteoclasts formed. We also showed thatsuccessful transfection of osteoclasts with NFATc1-specific siRNA re-sulted in a significant reduction of osteoclast-specific gene expression.Additional studies with RNAi in animal models need to be conducted toobserve the therapeutic relevance in treatment of endodontic disease.

There are 2 reasons that can account for the fact that 100% silenc-ing of the gene products was not achieved. (1) This technique onlyblocks the expression of newly transcribed mRNA, and the observedlevels might be due to residual proteins expressed by the cells beforesiRNA transfection. (2) Background from mRNA or protein present incells that were not successfully transfected will make the knockdownappear less effective than it actually is (23). Nonetheless, this demon-strates the potential that RNAi might have as a novel therapeutic strategyin treating endodontic disease.

NFATc1 siRNA transfected cells show 50% fewer cells differenti-ated into multinucleated TRAP-positive cells (Fig. 5) by morphologicexamination and express about 20% less TRAP mRNA by RT-PCR assay(Fig. 6), compared with control siRNA transfected cells. The possibleexplanation for the discrepancy between morphologic examination andPCR assay could be that (1) the 2 approaches have different sensitivity,and (2) NFATc1 might influence the fusion of multinucleated cells.

In this study, RAW264.7 cells were pretreated with RANKL for first24 hours and then transfected with control or NFATc1 siRNA. Cells werethen stimulated with LPS to examine the osteoclast formation. From ourprevious study (17), LPS alone cannot induce RAW 264.7 preoste-oclasts to differentiate into mature osteoclast. Initial pretreatment withRANKL is essential for osteoclast formation. A recent study by Sundaramet al (24) used a similar approach by knocking down NFATc1 expres-sion with siRNA in RAW 264.7 cells to examine the role of NFATc1 inmodulating the matrix metalloproteinase-9 (MMP-9) gene expression.MMP-9 activity was significantly decreased in NFATc1 siRNA transfectedcells (24).

Many researchers have used this technology to elucidate the rolesof individual genes in regulating cell growth, differentiation, and sur-vival in a broad range of cell lines (25, 26). Other groups have deployedRNAi for the identification of potential drug targets (27). However, thegreatest promise for RNAi might be in the field of clinical medicine andits potential for therapy against a broad range of diseases. There isgrowing enthusiasm regarding the potential therapeutic applications ofRNAi for cancer. Multiple groups have successfully demonstratedsiRNA-induced silencing of various cancer-associated genes in vitro,without affecting the wild-type copy (28, 29). Preliminary studies havealso shown the effectiveness of siRNA to specifically target human im-munodeficiency virus regulatory proteins such as TAT and REV, whereasothers have targeted host cell receptors such as CXCR4 and CCR5 (30).

However, similar to other forms of gene-based therapies, there areseveral problems associated with the development of siRNA therapeu-tics. The primary obstacle is the in vivo delivery of these small moleculesto the desired cell type, tissue, or organ. Because of their relatively largemolecular mass and high negative charge, RNAs have difficulty crossingthe cell membrane on their own. Two different approaches have beendeveloped to address this concern: (1) direct introduction of chemi-cally modified synthetic siRNAs enhanced for improved pharmacoki-netic properties or (2) the use of plasmid or viral vectors containingDNA templates to express siRNA within cells.

Even though the original studies of siRNA silencing suggested highspecificity, several mechanisms have been described with both syntheticand vector-based siRNA expression that can lead to unintended effectson gene expression and other unexpected “side effects,” all of whichneed to be carefully considered when developing RNA-based therapies(31). One potential complication is that siRNA has the ability to triggerthe innate immune system. Induction of an interferon response couldpotentially cause a global and nonspecific suppression of protein trans-lation, particularly in highly sensitive reporter cell lines at high concen-trations of siRNAs (32). However, interferon response is typically in-duced when the dsRNA molecule is greater than 30 base pairs, which islonger than the 21-23 nt in length siRNA used in RNA interference.

Perhaps a more significant problem associated with siRNA is un-anticipated “off-target” effects resulting from mRNA cleavage or trans-lational repression of genes bearing partially complementary sequencesto either strand of the duplex siRNA. It was originally believed that siRNArequires almost complete homology throughout the length of its se-quence with the intended mRNA target for effective RNAi to occur. How-ever, it now appears that as few as 11 contiguous complementary basepairs of siRNA might be enough to evoke the off-target effect of RNAi-mediated silencing (31). Although short stretches of homology are of-

Osteoclast mRNA Expression

0

0.2

0.4

0.6

0.8

1

1.2

Control N FATc1

siRNA

Treatment Groups

Relative Expression

(Fold Change)

CSK

TRAP

IL-6

TNF

*

*

*

*

Figure 6. Effect of inhibition of NFATc1 on osteoclast gene expression on LPSstimulation. *Statistical significance relative to control, P# .05.



ten inevitable, care must be taken to avoid longer stretches, which havemore considerable effects on gene expression.

Although it is clear that more progress needs to be made to im-prove RNAi delivery systems and to evaluate off-target effects and otherpotential sources of toxicity, it is encouraging to note that there aremore than 30 pharmaceutical and biotechnology companies that havestated interest in or currently have an RNAi-based drug developmentprogram in progress, and many have published preliminary data ob-tained from in vivo and mammalian model systems validating theirprojects (33, 34). It should be noted that Fomiversen (Vitravene) man-ufactured by ISIS Pharmaceuticals (Carlsbad, CA), which is a drug forthe treatment of cytomegalovirus retinitis, is the first Food and DrugAdministration–approved siRNA-based drug available on the market atthis time.

Not enough might be known about the potential negative effects ofprolonged or repetitive use of RNAi on normal cellular metabolismwhen used for treatment of chronic diseases. It is possible that toxicitiesmight not show up for months or perhaps years. Clearly, such issuesrequire additional long-term studies in therapeutically relevant animalmodels of RNA interference. Nonetheless, considering the immenseinterest and the rapid pace by which RNAi research is advancing, it isforeseeable that this relatively new scientific discovery will have a dra-matic impact on the development of an innovative and new class ofdrugs whose therapeutic potential seems enormous.

ConclusionRNA interference is a unique and powerful tool that can be used for

the study gene function by suppressing its expression. It is also a fast andinexpensive method to selectively silence a gene product in complexbiologic systems whose clinical potential for treatment of various dis-eases and disorders has been demonstrated. By using RNAi, with spe-cific NFATc1 siRNA we were able to successfully do the following:

(1) Deliver siRNA into the cytoplasm of monocytes and osteoclastswith high efficiency.

(2) Demonstrate a significant reduction in the expression ofNFATc1 in cells that were transfected with specific siRNA.

(3) Demonstrate a significant reduction of TNF- and IL-6 produc-tion in transfected monocytes in response to LPS stimulation.

(4) Demonstrate a significant reduction in the number of matureosteoclasts formed in response to LPS stimulation.

(5) Demonstrate a significant reduction in osteoclast-specific geneexpression in response to LPS stimulation.

Although much work still remains in improving the delivery, spec-ificity, and effectiveness of siRNAs, RNAi-based therapies have emergedas highly promising prospects with applications for a wide spectrum ofdiseases.

AcknowledgmentsThis study was supported by an AAE Foundation research grant

to Dr Fahid and Dr Jiang and NIDCR grant (DE015696) to Dr JinJiang.

References1. Kakehashi S, Stanley HR, Fitzgerald RJ. The effects of surgical exposures of dental

pulps in germfree and conventional laboratory rats. Oral Surg Oral Med Oral Pathol1965;20:340–9.

2. Moller AJ, Fabricius L, Dahlen G, Ohman AE, Heyden G. Influence on periapicaltissues of indigenous oral bacteria and necrotic pulp tissue in monkeys. Scand J DentRes 1981;89:475–84.

3. Paterson RC. Bacterial contamination and the exposed pulp. Br Dent J1976;140:231–6.

4. Bergenholtz G. Pathogenic mechanisms in pulpal disease. J Endod 1990;16:98–101.5. Farber PA, Seltzer S. Endodontic microbiology: I—etiology. J Endod

1988;14:363–71.6. Fabricius L, Dahlen G, Ohman AE, Moller AJ. Predominant indigenous oral bacteria

isolated from infected root canals after varied times of closure. Scand J Dent Res1982;90:134–44.

7. Raetz CR. Biochemistry of endotoxins. Annu Rev Biochem 1990;59:129–70.8. Cohen J. The immunopathogenesis of sepsis. Nature 2002;420:885–91.9. Dahlen G, Bergenholtz G. Endotoxic activity in teeth with necrotic pulps. J Dent Res

1980;59:1033–40.10. Schonfeld SE, Greening AB, Glick DH, Frank AL, Simon JH, Herles SM. Endotoxic

activity in periapical lesions. Oral Surg Oral Med Oral Pathol 1982;53:82–7.11. Yamasaki M, Nakane A, Kumazawa M, Hashioka K, Horiba N, Nakamura H. Endotoxin

and gram-negative bacteria in the rat periapical lesions. J Endod 1992;18:501–4.12. Kapur RP, Yao Z, Iida MH, et al. Malignant autosomal recessive osteopetrosis caused

by spontaneous mutation of murine Rank. J Bone Miner Res 2004;19:1689–97.13. Odgren PR, Kim N, Mackay C, Mason-Savas A, Choi Y, Marks SC Jr. The role of RANKL

(TRANCE/TNFSF11), a tumor necrosis factor family member, in skeletal develop-ment: effects of gene knockout and transgenic rescue. Connect Tissue Res 2003;44(Suppl 1): 264–71.

14. Pettit AR, Ji H, von Stechow D, et al. TRANCE/RANKL knockout mice are protectedfrom bone erosion in a serum transfer model of arthritis. Am J Pathol2001;159:1689–99.

15. Kikuchi T, Matsuguchi T, Tsuboi N, et al. Gene expression of osteoclast differentiationfactor is induced by lipopolysaccharide in mouse osteoblasts via Toll-like receptors.J Immunol 2001;166:3574-–9.

16. Suda K, Woo JT, Takami M, Sexton PM, Nagai K. Lipopolysaccharide supports survivaland fusion of osteoclasts independent of TNF-?, IL-1 and RANKL. J Cell Physiol2002;190:101.

17. Jiang J, Li H, Fahid FS, et al. Quantitative analysis of osteoclast-specific gene markersstimulated by lipopolysaccharide. J Endod 2006;32:742–6.

18. Takayanagi H, Kim S, Koga T, et al. Induction and activation of the transcription factorNFATc1 (NFAT2) integrate RANKL signaling for terminal differentiation of osteoclasts.Dev Cell 2002;3:889–901.

19. Ikeda F, Nishimura R, Matsubara T, et al. Critical roles of c-Jun signaling in regulationof NFAT family and RANKL-regulated osteoclast differentiation. J Clin Invest2004;114:475–84.

20. Elbashir SM, Martinez J, Patkaniowska A, Lendeckel W, Tuschl T. Functional anatomyof siRNAs for mediating efficient RNAi in Drosophila melanogaster embryo lysate.EMBO J 2001;20:6877–88.

21. de Fougerolles A, Vornlocher HP, Maraganore J, Lieberman J. Interfering with dis-ease: a progreee report on siRNA-based therapeutics. Nat Rev Drug Discov2007;6:443–53.

22. Li J, Cao ZW, Zhang SC. Effect of antisense Toll-like receptor 4 expressing plasmidsmouse macrophages stimulated by endotoxin. J Fudan (Medical Sciences, Chinese)2004;31:251–3.

23. Zhou D, He QS, Wang C, Zhang J, Wong-Staal F. RNA interference and potentialapplications. Curr Top Med Chem 2006;6:901–11.

24. Sundaram K, Nishimura R, Senn J, Youssef RF, London SD, Reddy SV. RNAK ligandsignaling modulates the matrix metalloproteinase-9 gene expression during oste-oclast differentiation. Exp Cell Res 2007;313:168–78.

25. Klampfer L, Huang J, Sasazuki T, Shirasawa S, Augenlicht L. Oncogenic ras promotesbutyrate-induced apoptosis through inhibition of gelsolin expression. J Biol Chem2004;279:36680–8.

26. Yin P, Xu Q, Duan C. Paradoxical actions of endogenous and exogenous insulin-likegrowth factor binding protein (IGFBP)-5 revealed by RNA interference analysis. J BiolChem 2004;279:32660–6.

27. Nencioni A, Sandy P, Dillon C, Kissler S, Blume-Jensen P, Van Parijs L. RNA interfer-ence for the identification of disease-associated genes. Curr Opin Mol Ther2004;6:136–40.

28. Brummelkamp TR, Bernards R, Agami R. Stable suppression of tumorigenicity byvirus-mediated RNA interference. Cancer Cell 2002;2:243–7.

29. Butz K, Ristriani T, Hengstermann A, Denk C, Scheffner M, Hoppe-Seyler F. siRNA-targeting of the viral E6 oncogene efficiently kills human papillomavirus-positivecancer cells. Oncogene 2003;22:5938–45.

30. Jacque JM, Triques K, Stevenson M. Modulation of HIV-1 replication by RNA inter-ference. Nature 2002;418:435–8.

31. Jackson AL, Linsley PS. Noise amidst the silence: off-target effects of siRNAs? TrendsGenet 2004;20:521–4.

32. Sledz CA, Holko M, de Veer MJ, Silverman RH, Williams BRG. Activation of theinterferon system by short-interfering RNAs. Nat Cell Biol 2003;5:834–9.

33. Whelan J. First clinical data on RNAi. Drug Discov Today 2005;10:1014–5.34. Behlke MA. Progress towards in vivo use of siRNAs. Mol Ther 2006;13:644–70.



Antibiotic Resistance Gene Transfer betweenStreptococcus gordonii and Enterococcusfaecalis in Root Canals of Teeth Ex VivoChristine M. Sedgley, MDSc, MDS, PhD, Esther H. Lee, BS, Matthew J. Martin, BA, andSusan E. Flannagan, MA

Abstract

Multiple bacterial species coexisting in infected rootcanals might interact, but evidence for interspeciesgene transfer is lacking. This study tested the hypoth-esis that horizontal exchange of antibiotic resistancecan occur between different bacterial species in rootcanals. Transfer of the conjugative plasmid pAM81carrying erythromycin resistance between 2 endodonticinfection-associated species, Streptococcus gordoniiand Enterococcus faecalis, was investigated in an exvivo tooth model. Equal numbers of each species (onewith pAM81 and the other plasmid-free) were com-bined in prepared root canals of sterilized teeth andincubated at 37°C. At 24 and 72 hours, bidirectionalinterspecies antibiotic resistance gene transfer was ev-ident in microorganisms recovered from teeth; averagetransfer frequencies from S. gordonii to E. faecalis were10 3 transconjugants per donor and from E. faecalis toS. gordonii were 10 6 and 10 7 transconjugants perdonor at 24 and 72 hours, respectively. Microbial ac-cumulations were observed on root canal walls withscanning electron microscopy. Horizontal genetic ex-change in endodontic infections might facilitate adop-tion of an optimal genetic profile for survival. (J Endod2008;34:570–574)

Key Words

Antibiotic resistance, conjugative plasmid, Enterococ-cus faecalis, ex vivo, filter mating, gene transfer, mi-crobial accumulations, root canal, scanning electronmicroscopy, Streptococcus gordonii

Root canal infections are generally composed of a diverse microflora (1) that canexist as biofilm communities on root canal walls (2, 3). Polymicrobial biofilm

communities are well-suited for horizontal gene transfer (4), providing pathogens withthe means to adapt rapidly, for example, by the acquisition of genes encoding antibioticresistance (5). Conjugation is a particularly efficient means of horizontal gene transferoccurring naturally in bacteria, whereby DNA is transferred between cells that are inphysical contact, and can involve the crossing of species barriers. Plasmids, some ofwhich are conjugative, are extrachromosomal autonomously replicating elements im-portant for adaptation and survival by the provision of functions not encoded by thechromosome. Conjugative plasmids (“self-transmissible” plasmids) encode the essen-tial functions needed for their own intercellular transmission by conjugation; noncon-jugative plasmids do not encode these functions. Plasmids are found in bacteria (5),archaea (6), and yeasts (7) and are of particular clinical importance because they canbe involved in the dissemination of antibiotic resistance.

Multiple antibiotic resistance in clinical isolates from root canal infections hasbeen reported (8, 9). Persistent root canal infections have been significantly associatedwith the recovery of Streptococcus spp (10, 11), most commonly S. gordonii (10), andEnterococcus spp (12–15). The latter might include antibiotic resistant isolates car-rying up to 4 plasmids per strain, including potential conjugative plasmids (9). It hasbeen speculated that bacterial species in infected root canals could communicate,perhaps resulting in plasmid transfer (16), but evidence for this in root canals islacking.

Specific interactions between different species recovered from root canal infec-tions can vary. For example, endodontic E. faecalis isolates coaggregated with Fuso-bacterium nucleatum but not with Peptostreptococcus anaerobius, Prevotella ora-lis, and Streptococcus anginosus (17), a “5-strain” combination associated withapical periodontitis in monkeys (18). Similarly, Porphyromonas gingivalis co-invadeddentinal tubules with S. gordonii but not with Streptococcus mutans (19). However,beyond the concurrent recovery of both Streptococcus and Enterococcus species frominfected root canals (10), there are no reports on interactions between these 2 gram-positive species in the root canal environment. In this study, the hypothesis was testedthat horizontal exchange of antibiotic resistance can occur between different bacterialspecies in root canals. Bidirectional transfer of an erythromycin resistance determinanton the conjugative plasmid pAM81 between 2 endodontic infection–associated species,S. gordonii and E. faecalis, was investigated using an ex vivo root canal infection modelthat has been previously described (20).

Materials and MethodsMicroorganisms

Bacterial strains used were S. gordonii Challis-Sm (21) (resistant to streptomycin[Sm]), E. faecalis JH2-2 (22) (resistant to rifampin [Rif] and fusidic acid [Fus]),E. faecalis JH2-2/pAM81 (23, 24) (resistant to Rif, Fus, and erythromycin [Em]), andS. gordonii Challis-Sm/pAM81 (resistant to Sm and Em). Conjugative plasmid pAM81 isapproximately 27 kb in size and encodes Em resistance (24). S. gordonii Challis-Sm/pAM81 was created for this study by conjugation between E. faecalis JH2-2/pAM81 andS. gordonii Challis-Sm with overnight filter mating methods as previously described (25).

From the Department of Cariology, Restorative Sciencesand Endodontics, The University of Michigan, School of Den-tistry, Ann Arbor, Michigan.

Address requests for reprints to Dr Christine Sedgley,Department of Cariology, Restorative Sciences and Endodon-tics, The University of Michigan, School of Dentistry, 1011 NUniversity Ave, Ann Arbor, MI 48109-1078. E-mail address:[email protected]/$0 - see front matter

Copyright © 2008 by the American Association ofEndodontists.doi:10.1016/j.joen.2008.02.014


570 Sedgley et al. JOE — Volume 34, Number 5, May 2008

Application of Small Interfering RNA

Documents

Transcript of Application of Small Interfering RNA