| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Experimental Therapeutics, Preclinical Pharmacology |
Expression, Achieved by RNA Interference, on the Activity of the New Generation Antifolate Pemetrexed in HeLa Cells
Departments of Medicine and Molecular Pharmacology, Albert Einstein College of Medicine and Albert Einstein Cancer Research Center, Bronx, New York
 |
ABSTRACT |
|---|
 Top ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
. The role of FR- in PMX pharmacological activity is uncertain. Whereas high-level expression may enhance the activity of this agent, it is not clear what role constitutive levels of this transporter contribute to PMX activity. In this study, constitutive levels of FR- expression were abolished by small interfering RNA-induced silencing in HeLa cells and RFC-null HeLa R5 cells as confirmed by Northern blotting, immunohistochemistry, and cell surface binding. PMX growth inhibition was unchanged in HeLa and R5 cells in the absence of FR- expression. Loss of FR- expression did not decrease net accumulation of PMX in either wild-type or RFC-null HeLa cells. Likewise, folate pools in wild-type HeLa cells were not decreased by FR- gene silencing and were negligibly affected in the RFC-null R5 subline grown with 5-formyltetrahydrofolate. FR- surface binding in HeLa cells was shown to be greater than that in a variety of other human solid tumor cell lines. Hence, constitutively expressed FR- in HeLa cells does not contribute to PMX activity in the presence or absence of RFC function. This is likely the case in many human solid tumor cell lines. |
INTRODUCTION |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
PMX is an excellent substrate for several of the known folate transporters. It has an affinity for reduced folate carrier (RFC) about twice that of methotrexate (MTX) in human and murine leukemia cells. It has an affinity for folate receptor (FR)- that is comparable with that of folic acid, its preferred substrate, and at least 2 orders of magnitude greater than that for MTX (8)
. However, the relative contributions of these two pathways to the delivery of PMX to cells and the role that RFC and FR- play as determinants of cytotoxicity are unclear.
Studies that have explored the role of FR- in PMX activity used cells with high endogenous receptor expression or with high expression achieved by transfection of FR- cDNA in which RFC expression was constitutive or decreased (4
, 8
, 9)
. There is no information on the contribution of FR- at levels of expression found in most human solid tumors (10, 11, 12)
. To explore this further, studies were undertaken with HeLa cells that express both transporters to determine the impact of the selective loss of FR- expression achieved by small interfering RNA (siRNA) gene silencing on PMX activity. In addition, FR- siRNA gene silencing was assessed in a HeLa cell variant with a genomic deletion of RFC (13)
to explore the role of receptor-mediated transport under these conditions. The levels of FR- as assessed by membrane folic acid binding were also quantitated in a variety of other human solid tumor cell lines to determine the extent to which observations in HeLa cells might have broad relevance to human tumors.
|
MATERIALS AND METHODS |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Cell Culture Conditions and Growth Inhibition Studies.
HeLa cells were purchased from American Type Culture Collection (Manassas, VA). R5 cells are HeLa cells with a genomic deletion of the RFC gene and were previously developed in this laboratory (13)
. Other human solid tumor cell lines were obtained from the National Cancer Institute Developmental Therapeutics Program (Bethesda, MD), with the exception of hepatoma cells, which were obtained from the Liver Center of the Albert Einstein College of Medicine (Bronx, NY). Cell lines were initially maintained in RPMI 1640 (Hyclone, Logan, UT) supplemented with 10% fetal bovine serum (Gemini Bio-Products, Woodland, CA), 2 mmol/L glutamine, 20 µmol/L 2-mercaptoethanol, 100 units/mL penicillin, and 100 µg/mL streptomycin at 37°C in a humidified atmosphere of 5% CO2. Transfected HeLa and R5 cells were maintained in hygromycin (350 µg/mL). At least 1 week before all experiments, cells were adapted to folate-free RPMI 1640 supplemented with 25 nmol/L 5-CHO-THF (and hygromycin for siRNA-transfected HeLa and R5 cells) in addition to the supplements noted above. Folate binding studies were also conducted in cell lines growing in regular RPMI 1640 containing 
2.3 µmol/L folic acid as the folate source. Cell cultures were monitored regularly with a Mycoplasma detection kit (American Type Culture Collection) and shown to be free of this organism. For assessment of PMX growth inhibition, cells were transferred to 96-well plates (500 cells per well) and exposed to a spectrum of 11 different concentrations of PMX (ranging from 2 nmol/L to 2 µmol/L) for 6 days in the absence of hygromycin. Cell growth was quantified by sulforhodamine B staining (14)
.
Construction of Folate Receptor Small Interfering RNA.
Two cell lines were selected for FR- gene silencing, wild-type HeLa cells and the HeLa subline R5, which has a genomic deletion of the RFC (13)
. The Silencer Express (Human U6) kit (Ambion, Austin, TX) was used initially to produce siRNA expression cassettes (SECs) to screen for the most effective siRNA sequence. Four target sequences were selected from the FR- cDNA sequence (GenBank accession number NM_000802), based on guidelines listed in the kit protocol. These were 19 nucleotides starting at positions 130, 253, 433, and 595. SECs, constructed as per the manufacturer’s protocol for the target sequences, are polymerase chain reaction (PCR) products consisting of promoter and terminator sequences flanking a hairpin siRNA template. The four SECs were transiently transfected into HeLa cells, and the construct that resulted in maximum reduction in folate binding capacity was cloned into the pSEC hygro vector (Ambion) that also encodes the hygromycin resistance gene. This was then transfected into wild-type HeLa and RFC-null R5 cells followed by culture in RPMI 1640 containing 350 µg/mL hygromycin to obtain stable clones. Cells were also transfected with the negative control pSilencer 2.1-V6 hygro vector (Ambion) that expresses a siRNA with limited homology to any known sequences in the human, mouse, and rat genomes followed by selection in the same medium. Ten clones were picked from the FR- siRNA plate, and the four clones that had a maximum reduction in folate binding capacity were then used for additional studies. Four clones from the negative control plate were also randomly picked for comparison.
Transient and Stable Transfection.
Four SECs were initially screened by transient transfection in HeLa cells. Cells were seeded in 6-well plates to reach 40% to 50% confluence by the next day. SECs were transfected into cells in duplicate using LipofectAMINE PLUS reagent. After an additional 2 days of growth, folate binding capacity was assessed. For stable transfection, the vector containing SEC-130 was transfected into HeLa or R5 cells at 50% confluence in 100-mm plates using the same reagent. Forty eight hours after transfection, cells were passaged into fresh 100-mm plates containing hygromycin at 350 µg/mL.
Northern Blot Analysis.
Total RNA was isolated from HeLa or R5 cells using the TRIzol reagent method (Invitrogen, Carlsbad, CA). Total RNA (30 µg) was resolved by electrophoresis on a 1% denaturing agarose gel containing formaldehyde. RNA was then blotted onto Nytran N-membranes (Schleicher & Schuell, Keene, NH) and fixed with a Stratalinker UV cross-linker (Stratagene, La Jolla, CA). The membrane was probed with human FR- cDNA and then stripped and reprobed with ß-actin cDNA.
Immunofluorescence Staining.
HeLa or R5 cells were seeded on autoclaved glass slides and grown in folate-free medium containing 25 nmol/L 5-CHO-THF and 350 µg/mL hygromycin for 3 days. The slides were then washed twice in PBS, and the cells were fixed by incubation with 2% paraformaldehyde in PBS for 30 minutes. After three more washings to eliminate paraformaldehyde, cells were incubated in the blocking solution containing 1% bovine serum albumin (in PBS) for 1 hour. Excess bovine serum albumin was then wiped off, the primary anti–folate-binding protein monoclonal antibody (Mov18/ZEL; Alexis Biochemicals, San Diego, CA) was added at a concentration of 2 µg/mL, and the slides were kept overnight at 4°C. The next morning, the slides were washed three times with PBS, and the secondary fluorescein isothiocyanate (FITC)-linked sheep antimouse antibody (Amersham Biosciences, Buckinghamshire, United Kingdom) was added at a 1:25 dilution with further incubation at room temperature for 1 hour. Prolong antifade solution (Molecular Probes, Eugene, OR) was then used under coverslips, and the slides were examined by fluorescence microscopy.
Folate Binding Capacity.
Small interfering RNA-transfected HeLa and R5 cells were seeded in 12-well plates, grown to near confluence, and washed with ice-cold acid buffer [10 mmol/L NaAC, 150 mmol/L NaCl (pH 3.5)], followed by a wash with ice-cold HBS [20 mmol/L HEPES, 140 mmol/L NaCl, 5 mmol/L KCl, 2 mmol/L MgCl2, 5 mmol/L glucose (pH 7.4)]. Cells were then exposed to 5 nmol/L [3H]folic acid for 15 minutes in ice-cold HBS and washed three times with [3H]folic acid-free buffer. The other human solid tumor cell lines were seeded in triplicate in 12-well plates and grown to near confluence. Folate binding capacity in 2 wells was assessed with 5 nmol/L [3H]folic acid, alone, as described above. In the 3rd well, 500 nmol/L nonlabeled folic acid was added so that specific binding could be calculated from the difference between [3H]folic acid bound in the presence and absence of nonlabeled folic acid. [3H]Folic acid bound to the cell surface was released with acid buffer (0.5 mL) and measured on a liquid scintillation spectrometer. The adherent cells were subsequently lysed with 0.2 mol/L NaOH (0.5 mL), and 10 µL of the lysate were used for protein determination (see below). Folate binding capacity is expressed in nmol/g or pmol/g protein. The FR- siRNA clones had folate binding capacities close to background, indicating that essentially all of the [3H]folic acid was specifically bound. Hence, the addition of nonlabeled folic acid was not required.
Pemetrexed Accumulation.
HeLa and R5 cells were seeded in 6-well plates in folate-free RPMI 1640 supplemented with 25 nmol/L 5-CHO-THF. [3H]PMX (at a concentration of 50 nmol/L) and 200 µmol/L glycine, 100 µmol/L adenosine, and 10 µmol/L thymidine were added to the growth medium. After 3 days, the cells reached confluence and were passaged into fresh 6-well plates containing the same constituents. After 3 additional days (6 days, total), the cells were washed three times with ice-cold HBS. Adherent cells were digested in 0.2 mol/L NaOH (0.5 mL) at 65°C for 40 minutes. Lysate (0.4 mL) was transferred to scintillation vials, fluor was added, and radioactivity was determined. Lysate (10 µL) was also processed for protein determination using the BCA protein kit (Pierce, Rockford, IL). Cellular uptake is expressed in units of pmol/mg protein.
Intracellular Folate Pool Measurement.
Cells were seeded in 6-well plates in folate-free RPMI 1640. [3H]5-CHO-THF was added to achieve a concentration of 25 nmol/L. Cells were grown for a total of 7 days (with one passage at the end of 3 days) and washed, and intracellular tritium was determined as described above.
Statistical Tests.
The Mann-Whitney nonparametric test was used to evaluate differences in the siRNA-transfected versus control cells. All values were ranked from low to high, and the mean rank from each group was compared. The resulting P values are depicted in the figures.
|
RESULTS |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Gene Silencing. cDNA and SEC constructs transiently transfected into HeLa cells. The SEC found to elicit the most effective gene silencing, as measured by the reduction in folate binding capacity, was the construct beginning at position 130 as depicted in Fig. 1
. This construct was then cloned into the pSEC hygro vector (Ambion), a plasmid that also encodes the hygromycin resistance gene, to obtain stable transfectants. Wild-type HeLa and RFC-null R5 cells were transfected with this plasmid and selected in RPMI 1640 containing hygromycin. Cells were also transfected with a negative control pSilencer nonsense siRNA vector and selected in the same medium. The four most optimal FR- siRNA clones (as assessed by the reduction in folate binding capacity) were used for additional studies. Four clones from the negative control plate were also randomly picked for comparison.
|
, FR- mRNA was not detected in the siRNA clones in wild-type HeLa or RFC-null R5 cell lines. All 16 clones were then probed with the murine monoclonal antibody to human FR- and screened for the presence of fluorescence after labeling with FITC-tagged antimouse sheep secondary antibody. Fluorescence was completely absent in the siRNA clones, as indicated in Fig. 2B
. Finally, membrane folate binding was assessed (Fig. 2C)
; there was a near complete absence of folate binding in the siRNA clones, indicating that virtually all membrane-bound folic acid was associated with FR-. It is of interest that there was a 2-fold clonal variability in folic acid binding in HeLa cells. This was not noted in the R5 group; however, too few clones were available to document the validity of this difference. The average folate binding capacities of vector control-transfected HeLa and R5 clones (768 ± 129 and 606 ± 21 pmol/g protein) were comparable with the folate binding capacities reported previously in nontransfected wild-type HeLa and R5 cells respectively growing in 5-CHO-THF (15)
.
|
Loss on Pemetrexed Cytotoxicity. expression on PMX cytotoxicity, growth inhibition was assessed over 6 days in the FR- siRNA and nonsense siRNA transfected HeLa and RFC-null R5 cells. There was clonal variability in sensitivity to PMX in wild-type HeLa cells, but not in R5 cells, as indicated in Table 1
. The four clones from each group were averaged, and the resulting cytotoxicity curves were analyzed. As demonstrated in Fig. 3
, HeLa cells transfected with FR- siRNA were as sensitive to PMX as cells transfected with the nonsense siRNA. In the absence of RFC, silencing FR- did not alter (P = 0.114) the sensitivity of R5 cells to PMX. The R5 cells were about 2-fold more sensitive to PMX as compared with wild-type HeLa cells, consistent with previous studies from this laboratory (15)
. This is presumably secondary to the loss of RFC-mediated transport of 5-CHO-THF and consequent depletion of cellular folate levels when cells are grown with this folate source in the medium (see Discussion). The IC50 values of wild-type HeLa and R5 cells growing in 25 nmol/L 5-CHO-THF (40 and 20 nmol/L, respectively; ref. 15
) are similar to the average of the HeLa and R5 vector control-transfected clones studied here (38.5 ± 5 and 21 ± 1.3 nmol/L, respectively).
|
|
on Net Accumulation of Pemetrexed and Formyltetrahydrofolate.. Net accumulation of 50 nmol/L [3H]PMX (the approximate IC50 concentration) was assessed over 6 days (the total duration of the growth inhibition studies). Net accumulation of 25 nmol/L [3H]5-CHO-THF was assessed under similar conditions.
As shown in Fig. 4
, PMX accumulation was not decreased in the siRNA-transfected HeLa and R5 cells. There was a negligible decrease (10%; P = 0.057) in net folate accumulation in the R5 siRNA clones but not in wild-type HeLa siRNA clones grown with [3H]5-CHO-THF (Fig. 5)
. Hence, FR-, at the constitutive expression levels in these cells, does not play a role in the net accumulation of PMX and 5-CHO-THF in the presence or absence of RFC function.
|
|
expression in HeLa cells is representative of human solid tumor cell lines in general, and because the relative importance of FR- should correlate with its levels of expression, quantitative determination of surface expression was made in a panel of human solid tumor cell lines using folate binding capacity as a marker for FR- protein present on the cell membrane. As indicated in Table 2
, the folate binding capacity of HeLa and R5 cells was higher than that of all of the 11 cell lines surveyed, whether cells were grown in folic acid or 5-CHO-THF, although there was a general increase among all of the cell lines under the latter conditions. The highest levels were present in the ovarian carcinoma and CaCO2 cell lines; levels of membrane FR- in most of the other cell lines were 1 order of magnitude below what was detected in HeLa cells.
|
|
DISCUSSION |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
, comparable with that of folic acid, a greater affinity for RFC than MTX, and the marked enhancement of activity that occurs when cellular folate pools are contracted (7
, 16)
.
FR- is widely expressed in human solid tumors; however, there is little information on the relationship among FR- message, protein, surface binding, and function or the extent to which constitutive levels of FR- expression contribute to the pharmacological activity of this agent. For instance, FR- message is expressed in most mesothelioma surgical specimens, an observation that led to the suggestion that this is one basis for the activity of PMX in this disease (17)
. A subsequent study demonstrated that MTX transport is mediated almost exclusively by RFC in mesothelioma cell lines despite the expression of FR- (18)
; however, no information was provided on transport of PMX, which has a 2-order higher affinity for FR- than MTX (8)
. Transport mediated by FR- is a slow process relative to RFC because it involves binding of substrate to receptor at the cell surface, formation of an endocytotic vesicle, invagination, and vesicle acidification, followed by release into the cytoplasm (19)
. In fact, the rate of cycling of FR- is about one hundredth that of RFC (20
, 21)
. Hence, it is only when the receptor is highly overexpressed that delivery by this route can be comparable with or exceed that of RFC (21)
.
Previous studies have evaluated the relationship between FR- expression and PMX cytotoxicity (4
, 8
, 9)
. In L1210 murine leukemia cells grown under folate-deplete conditions to achieve high expression of FR-, blockade of the receptor with folic acid resulted in 4-fold resistance to PMX with intact RFC and 10-fold resistance in the absence of the carrier (8)
. In folate-depleted conditions, ZR-75-1 human breast carcinoma cells became 4- to 5-fold sensitive to PMX when transfected to high levels of FR- in the presence of RFC, but they became 2- to 3-fold resistant to PMX in the absence of the carrier (4)
. The latter may have been due to a FR-–mediated expansion of cellular folate pools. This is consistent with the observation that cells with increased FR expression have decreased sensitivity to trimetrexate under low folate conditions (22)
. Trimetrexate inhibition of dihydrofolate reductase is highly sensitive to the level of cellular folate pools, and this antifolate is not transported by RFC or FR- (16)
. Likewise, A431 epidermoid vulval cells overexpress FR- (with constitutive levels of RFC) and are about 14-fold more sensitive to PMX when grown in 1 nmol/L 5-CHO-THF compared with wild-type cells, whereas they are only 3-fold more sensitive when grown in the more physiologic 20 nmol/L 5-CHO-THF (9)
.
The current study demonstrates for the first time that loss of constitutive FR- expression in HeLa cells, achieved by stable transfection of FR- siRNA, does not result in a decrease in PMX activity. Consistent with this are the observations that accumulation of PMX and 5-CHO-THF were essentially unchanged in the FR-–silenced cells. Hence, the demonstration of FR- message, protein, or even surface binding in tumors, alone, provides no information about function and, in the case of PMX, no clue as to impact on pharmacological activity. As observed in wild-type cells, the loss of FR- did not alter PMX accumulation or activity in the HeLa-derived, MTX-resistant R5 cell line, in which there is a genomic deletion of RFC, although, as discussed below, there was RFC-independent residual transport in the latter cells (13)
.
As reported previously (15)
, when grown with 5-CHO-THF as the folate growth source, RFC-null R5 cells are collaterally sensitive to PMX as compared with wild-type HeLa cells (Fig. 3
; Table 1
). This is due to the contraction of folate cofactors in all of the R5 clones (both siRNA and control) resulting from diminished 5-CHO-THF transport and accumulation in the absence of RFC, leading to augmented polyglutamylation of PMX and possibly diminished competition between antifolate and folate substrates at the target enzymes (7
, 15)
. In addition, PMX transport persists at a modestly reduced rate (45% decrease) at neutral pH in HeLa R5 cells that lack RFC activity and at an unchanged rate at low pH (15)
. Hence, PMX has alternative routes of entry into HeLa cells in the absence of both RFC and FR-, and the findings in this study underscore the importance of this residual transport in maintaining adequate delivery of PMX in their absence.
Studies from this laboratory provide some insights into the characteristics of this RFC-independent secondary transport route in HeLa cells. It has a low pH optimum, but there appears to be residual activity at neutral pH; it is RFC independent; it has the characteristics of a facilitative carrier; and it is widely prevalent in human solid tumor cell lines but especially prominent in HeLa cells (13 , 23) . It has a particularly high affinity for PMX (Kt = 45 nmol/L at pH 5.5) relative to MTX (Kt = 1 µmol/L at pH 5.5; ref. 23 ). The evidence that supports the notion that the transport activities at neutral and low pH in RFC-null cells are related is based on the following observations: (a) R5 cell sublines obtained after MTX selective pressure lose both low and neutral pH activities. (b) These activities reappear concurrently when the drug is removed.1 (c) The RFC-independent activity at neutral pH has a similar spectrum of relative affinities as at low pH, but at much lower affinity levels. For instance, at pH 7.4, the Kt values for PMX and MTX are 12 and 90 µmol/L, respectively (23)
Of interest was the clonal variability (Table 1
; Figs. 2
, 4
, and 5
) in the various vector control-transfected and FR- siRNA-transfected cells. This was most prominent in parental HeLa cells as compared with R5 cells that originated from a single clone. This emphasizes the vagaries of comparing single clones after transfection experiments. To quantitate and minimize complications due to this phenomenon, several clones were evaluated in each experimental group.
It is difficult to compare, quantitatively, FR- expression in HeLa cells in this study with measurements made by other laboratories in other cell lines (9, 10, 11, 12
, 24
, 25)
because (a) quantitative assessment of message by Northern blot or reverse transcription-PCR provides no information on the level of membrane protein or function. (b) Measurements have been made on whole cells or isolated membranes and expressed on the basis of cell numbers or protein. (c) Studies have been conducted on cells grown with usual levels of folic acid or under folate-depleted conditions (8)
; the latter is known to increase FR- expression (26
, 27)
, as also observed in this study. The current studies used the same technique in HeLa cells and 11 other human solid tumor cell lines to assess and compare FR- surface binding. In all cases, the level of expression was less than that in HeLa cells. This included two ovarian carcinoma and CaCO2 cell lines, which have been reported to express high levels of FR- message (11
, 12 , 28)
. The other tumor cell lines had FR- levels one tenth that of HeLa cells. Hence, the lack of importance of FR-–mediated PMX transport in HeLa cells appears to be relevant, more generally, to many other human solid tumor cell lines. It is, of course, uncertain as to whether these observations are relevant to fresh human tumor tissues, for which, due to technical reasons, there is no information available on specific folate binding or function.
|
FOOTNOTES |
|---|
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
Requests for reprints: I. David Goldman, Departments of Medicine and Molecular Pharmacology, Albert Einstein College of Medicine, 1300 Morris Park Avenue, Bronx, NY 10461.
1 Zhao R, Chattopadhyay S, Hanscom M, Goldman ID, unpublished observations. 
Received 6/24/04; revised 9/ 1/04; accepted 9/10/04.
|
REFERENCES |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  S. Chattopadhyay, R. G. Moran, and I. D. Goldman Pemetrexed: biochemical and cellular pharmacology, mechanisms, and clinical applications Mol. Cancer Ther., February 1, 2007; 6(2): 404 - 417. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Chattopadhyay, R. Zhao, S. A. Krupenko, N. Krupenko, and I. D. Goldman The inverse relationship between reduced folate carrier function and Pemetrexed activity in a human colon cancer cell line. Mol. Cancer Ther., February 1, 2006; 5(2): 438 - 449. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Cancer Research | Clinical Cancer Research |
| Cancer Epidemiology Biomarkers & Prevention | Molecular Cancer Therapeutics |
| Molecular Cancer Research | Cancer Prevention Research |
| Cancer Prevention Journals Portal | Cancer Reviews Online |
| Annual Meeting Education Book | Meeting Abstracts Online |
