Degranulation of mast cells in response to Ag or the calcium mobilizing agent, thapsigargin, is dependent on emptying of intracellular stores of Ca2+ and the ensuing influx of external Ca2+, also referred to as store-operated calcium entry. However, it is unlikely that the calcium release-activated calcium channel is the sole mechanism for the entry of Ca2+ because Sr2+ and other divalent cations also permeate and support degranulation in stimulated mast cells. In this study we show that influx of Ca2+ and Sr2+ as well as degranulation are dependent on the presence of the canonical transient receptor potential (TRPC) channel protein TRPC5, in addition to STIM1 and Orai1, as demonstrated by knock down of each of these proteins by inhibitory RNAs in a rat mast cell (RBL-2H3) line. Overexpression of STIM1 and Orai1, which are known to be essential components of calcium release-activated calcium channel, allows entry of Ca2+ but not Sr2+, whereas overexpression of STIM1 and TRPC5 allows entry of both Ca2+ and Sr2+. These and other observations suggest that the Sr2+-permeable TRPC5 associates with STIM1 and Orai1 in a stoichiometric manner to enhance entry of Ca2+ to generate a signal for degranulation.
An increase in cytosolic Ca2+ levels in mast cells provides an essential signal for exocytotic release of granules whether cells are stimulated via the IgE receptor (FcεRI) or thapsigargin (1, 2). The increase in cytosolic Ca2+ is associated with release from intracellular stores and influx of external Ca2+ (3, 4) by a mechanism, originally referred to as capacitative calcium entry but is now often referred to as store-operated calcium entry (SOCE),5 in which depletion of intracellular stores of Ca2+ leads to entry of Ca2+ (5, 6). Ca2+ entry was initially thought to occur through a calcium release-activated calcium (CRAC) current (ICRAC) that was first identified in mast cells by patch-clamp techniques (7, 8). ICRAC is a Ca2+-selective current of low conductance that requires external Ca2+ to achieve maximal conductance. Replacement of Ca2+ with Sr2+ or Ba2+ results in a decline in ICRAC activity in T lymphocytes (9, 10). Nevertheless, it appears unlikely that ICRAC is the sole mechanism for conductance of Ca2+ as early studies of calcium-dependent exocytosis clearly demonstrated that stimulated mast cells become highly permeable to a variety of divalent cations including Sr2+, Ba2+, and Mn2+ and that such ions can support degranulation in the absence of Ca2+ (11, 12, 13).
Members of the subfamily of transient receptor potential (TRP) channels, especially the canonical TRP (TRPC) channels, were initially considered as candidates for mediating SOCE (14, 15). The TRPC channels are activated as a consequence of inositol 1,4,5-trisphosphate generated through phospholipase C (PLC). Some TRPC channel proteins, such as TRPC1 and TRPC4, are activated by store depletion via inositol 1,4,5-trisphosphate or thapsigargin as demonstrated in cells deficient in TRPC1 or TRPC4 (16) or by overexpression (14, 15, 17, 18, 19) or knock down (18, 20, 21, 22, 23, 24, 25) of these proteins. TRPC5 may be regulated by a variety of signals, including Ca2+ store depletion (26), although there are contrary reports (27, 28). On the basis of structure and function, TRPC1, TRPC4, and TRPC5 appear to form one category of TRPC channel proteins and TRPC3, TRPC6, and TRPC7 form another (29). The latter category appears to be activated primarily by diacylglycerides (i.e., receptor-operated) rather than store depletion (30, 31). However, TRPC channels can conduct divalent cations such Sr2+ and Ba2+ in addition to Ca2+ and none of the TRPC channel proteins appear to have the exact electrophysiological features of CRAC (32).
The molecular identity of the ICRAC putative channel CRAC has been clarified recently with the identification of two proteins, STIM1, a Ca2+ sensor, and Orai1, a channel protein, which are essential for CRAC channel activity. STIM1 was first identified as a component of CRAC channel in Drosophila S2 cells and Jurkat T cells by use of inhibitory RNAs (33). Of the two known mammalian homologs (STIM1 and STIM2), STIM1 appears to be the primary sensor of Ca2+ in intracellular stores in mammalian cells (33, 34, 35, 36, 37). STIM1 is strategically located in the endoplasmic reticulum and knock down of STIM1, but not of STIM2, by small inhibitory RNA (siRNA) abolishes SOCE and CRAC following Ca2+-store depletion. The second component of the CRAC channel was identified as a mammalian homolog of Drosophila Orai (38) also referred to as CRACM (39). A key discovery was the presence of an inactivating mutation of Orai1 in a patient with severe immunodeficiency (SCID) that was associated with low CRAC channel activity in T cells (38, 40). Subsequent studies showed that coexpression of Orai1 and STIM1 dramatically enhanced CRAC and SOCE (41, 42, 43). A paradoxical and unexplained finding was that overexpression of Orai1 by itself suppresses Ca2+ entry. In addition, the electrophysiological features of coexpressed Orai1 and STIM1 did not match those originally described for CRAC, which left the possibility that additional molecules might be involved (32). Recent work now indicates that TRPC channels may associate with STIM1 and Orai1 and thus alter channel properties as well as enhance Ca2+ entry (44, 45, 46). For example, transfection of cells with exogenously tagged proteins indicated that STIM1 can interact with TRPC1, TRPC4, and TRPC5 to regulate their activities (46). STIM1 also colocalizes with both TRPC1 and Orai1 and thus regulates SOCE (44). Interestingly, TRPC3 and TRPC6, which normally operate as receptor-activated and store-independent channels, appear to operate in a store-dependent manner when coexpressed with Orai1 (45).
We investigated the potential role of endogenous TRPC channels in supporting entry of Ca2+ and Sr2+ as well as degranulation in a rat mast cell line RBL-2H3 by overexpression or knock down of TRPC channel proteins, Orai1, and STIM1. As reported in this study, we found that, among the various TRPC channel proteins expressed in these cells, TRPC5 permitted influx of Sr2+, optimal influx of Ca2+, and degranulation. However, TRPC5 function was dependent on Orai1 and STIM to suggest that TRPC5 acted in conjunction with these two proteins.
Materials and Methods
d-glucosaminide from Sigma-Aldrich; thapsigargin, 1-oleoyl-2-acetyl-sn-glycerol, U73122, and RHC80267 from Calbiochem; Abs against TRPC1, TRPC3, and TRPC5 from Alomone Laboratories and against TRPC2 from Chemicon International; TRPC7 Ab from Bethyl Laboratories; Abs against Orai1 from ProSci; Abs against STIM1 from BD Biosciences; and fura 2-AM ester from Molecular Probes. Orai1 and STIM1 plasmids as well as HuSH 29mer short hairpin RNA (shRNA) constructs against Orai1, STIM1, and various TRPC channel proteins were obtained from OriGene Technologies. Although all available shRNA constructs from OriGene Technologies were tested, representative data are shown with the following constructs (product number): TI548839, TI100018, TI502171, TI304029, and TI336191 to knock down TRPC1, TRPC5, TRPC7, Orai1, and STIM1, respectively. All other chemicals were molecular biology grade from several sources.p-nitrophenyl N-acetyl-β-
The siRNA targeted against rat TRPC3 was generated with the Silencer Express kit, pSEC-neo (Ambion), and the oligonucleotide (the target sequence was 541–561 of rat TRPC3, AB022331) sense 5′-GTGCTACACAAACACACCTACTACGAAGTCTCGGTGTTTCGCCTTTCCACAAG-3′ and antisense 5′-CGGCGAAGCTTTTTCCAAAAAAGACTTCGTAGTAGGTGTGCTACACAAACACA-3′, according to the manufacturer’s instruction. A negative control siRNA was generated by using the oligonucleotides that were provided in the Silencer Express kit.
Detection of mRNA for TRPCs by RT-PCR
RBL-2H3 cells (5 × 106 cells) were used for extraction of RNA as described elsewhere (47). The Advantage RT-for-PCR kit (Clontech Laboratories) was used to transcribe 1 μg of RNA to cDNA, of which 1 μg was used for PCR with gene-specific primers for rat TRPC family members (48
Transient transfection, stimulation of cells, and immunoblotting
RBL-2H3 and the muscarinic m1 receptor-bearing RBL-2H3(m1) cells (49, 50) were cultured in complete growth medium as previously described (51). Cells were transiently transfected by electroporation (Cell Line Nucleofector kit L, Amaxa transfection system) with the plasmids or inhibitory RNAs (1 μg/106 cells) along with the expression vector that encodes enhanced yellow fluorescent protein (eYFP) (pd2EYFP-N1; Clontech Laboratories) in the ratio of 5:1. Cells were examined 24 or 48 h after transfection with the inhibitory RNAs or plasmids, respectively. Transfection efficiency ranged from 50% to 80%. Where necessary, cells were incubated overnight with IgE (50 ng/ml) to achieve 100% occupancy of FcεRI. Cells were stimulated with 100 ng/ml Ag (dinitrophenylated human serum albumin) or 1 μM thapsigargin to achieve maximal responses. Immunoblotting of whole cell lysates were performed as described elsewhere (51).
Measurement of degranulation
Cells in 24-well plates (2 × 105 cells/0.4 ml/well) were washed and the medium replaced with a PIPES-buffered medium (25 mM PIPES (pH 7.2), 159 mM NaCl, 5 mM KCl, 0.4 mM MgCl2, 1 mM CaCl2, 5.6 mM glucose, and 0.1% fatty acid-free fraction V from bovine serum) before stimulation. Where indicated, Ca2+ was omitted from the medium or 3 mM Sr2+ was substituted for Ca2+. Degranulation was determined by measurement of the release of the granule marker, β-hexosaminidase, by use of a colorimetric assay in which release of p-nitrophenol from p-nitrophenyl N-acetyl-β-d-glucosaminide was measured (1). Values were expressed as the percentage of intracellular β-hexosaminidase that was released into the medium.
Imaging of intracellular calcium in single cells
After transfection, cells were grown on coverslips in complete growth medium for 48 h. Medium was then replaced with PIPES-buffered medium (see previous section) and loaded with fura 2-AM (2 μM) for 25 min at 25°C. Cells were washed, and the dye was allowed to de-esterify for a minimum of 15 min at 25°C. The coverslips were placed in Ca2+-free PIPES-buffered medium and 1 mM CaCl2 or 3 mM SrCl2 was added as shown in the experiments. Cytosolic Ca2+ was measured in individual transfected and nontransfected cells in the InCyt dual wavelength fluorescence imaging system (Intracellular Imaging). eYFP served as the transfection marker in cells and was detected at an excitation wavelength of 485 nm. Nontransfected cells (not expressing eYFP) were identified from the same field and served as control cells. After cell identification, fluorescence emission at 505 nm was monitored with alternating excitation at wavelengths 340 and 380 nm. Relative intracellular content of divalent cation (Ca2+ or Sr2+) is depicted as the average ratio of fluorescence (340/380 nm) in 10–20 single transfected and nontransfected cells as previously described (52). It should be noted that Sr2+ has much lower affinity for fura 2 than Ca2+ (estimated Kd ∼ 0.23 μM for Ca2+ and 2.6 μM or greater for Sr2+) (53, 54, 55), although the fluorescence ratio is similar for both ions at half maximal saturation of fura 2 (55). Therefore, the relative increases in the fluorescence ratios for Sr2+ and Ca2+ understate the actual concentration of cytosolic Sr2+ compared with that of Ca2+ in the data shown in this study. All measurements are representative of three or more independent experiments.
Sr2+ entry is dependent on store depletion
Previous findings that Sr2+ can substitute for Ca2+ to support degranulation in stimulated RBL-2H3 cells (13) raise the possibility that Ca2+-conducting channels also conduct Sr2+. To examine whether entry of Sr2+, like Ca2+, is dependent on depletion of intracellular stores of Ca2+, we compared the effects of thapsigargin to those of Ag and the muscarinic agonist, carbachol, in RBL-2H3(m1) cells, a stably transfected cell line made to express the G protein-coupled muscarinic m1 receptor (49, 50). In contrast to thapsigargin, which minimally stimulates PLC in RBL-2H3 cells (56), Ag and carbachol activate PLCγ (57) and PLCβ (58), respectively. Both Ag and carbachol induced a transient Ca2+ release from intracellular Ca2+ stores in the absence of external Ca2+ followed by Sr2+ or Ca2+ entry when either of these ions were added externally (Fig. 1⇓, A and B). Depletion of the intracellular Ca2+ stores by thapsigargin, which does so by blocking uptake of cytosolic Ca2+ into these stores (6), also induced Sr2+ and Ca2+ entry (Fig. 1⇓C). These results suggested that entry of Sr2+ and Ca2+ was dependent on depletion of Ca2+ stores regardless of the mechanism of depletion.
Another indication that entry of Sr2+, as well as Ca2+, can occur independently of PLC was that the PLC inhibitor, U73122, inhibited release of intracellular Ca2+ and reduced entry of Sr2+ and Ca2+ when cells were stimulated by either Ag or carbachol (Fig. 1⇑, D and E) but it failed to suppress entry of these ions when Ca2+ stores were depleted with thapsigargin (Fig. 1⇑F). However, the entry of both Sr2+ and Ca2+ in response to all three stimulants was blocked by La3+ (Fig. 1⇑, G–I) at a concentration (3 μM) that is known to block CRAC activity in mast cells (7).
Identification of TRPC5 as a channel for store-dependent Sr2+ and Ca2+ entry
Messenger RNAs for several TRPC channel proteins were identified in RBL-2H3 cells by RT-PCR. These included TRPC1, TRPC2, TRPC3, TRPC5, and TRPC7 (Fig. 2⇓A). Of these, TRPC1, TRPC3, and TRPC5 could be detected at the protein level (Fig. 2⇓B), whereas TRPC2, a pseudo-gene in some but not all species (59, 60), was not detectable in RBL-2H3 cells. Abs against TRPC7 lacked sufficient specificity for this purpose.
Knock down of TRPC1, TRPC3, TRPC5 (Fig. 2⇑B), and TRPC7 with shRNAs indicated that only knock down of TRPC5 substantially impaired entry of Sr2+ and Ca2+ in thapsigargin-stimulated cells (Fig. 3⇓, A–E). Entry of these cations was also unimpaired by prior transfection of cells with empty vector (Fig. 3⇓A) or with shRNA against TRPC2 (data not shown). A small decrease in entry of cations was observed with knockdown of TRPC1 although the expression of this protein was decreased to a similar extent as was TRPC5 (Fig. 2⇑B). Other control experiments showed that shRNA against TRPC5 had no effect on expression of TRPC1 (data not shown).
Although entry of Sr2+ and Ca2+ is dependent on store depletion in RBL-2H3 cells (Fig. 1⇑) and TRPC5 is reported to be activated by store-depletion (26), overexpressed TRPC5 can be activated in a PLC/diacylglycerol-dependent manner (61). However, we found that the diacylglycerol cell-permeant analog, 1-oleoyl-2-acetyl-sn-glycerol, or the use of the lipase inhibitor RHC80267 to increase diacylglycerol levels (30), failed to stimulate activate entry of Ca2+ or Sr2+ (data not shown). Together, these results suggested that of the TRPC family members, TRPC5 is a likely component of a SOCE channel in RBL-2H3 cells, although a minor role for TRPC1 cannot be excluded.
Endogenous STIM1 and Orai1 also regulate store-operated Sr2+ and Ca2+ entry
STIM1, the sensor of free Ca2+ in calcium stores in the endoplasmic reticulum, was originally believed to regulate SOCE through its interaction with Orai1 to activate CRAC (32). Recent studies now suggest that STIM1 can also interact with TRPC proteins individually or in combination with Orai1.
We found that endogenous STIM1 as well as Orai1 were essential for TRPC5 function because knock down or either STIM1 or Orai1 by use of shRNA significantly suppressed thapsigargin-induced Sr2+ as well as Ca2+ entry (Fig. 4⇓). Therefore, both TRPC5 and Orai1 are required for Sr2+ and Ca2+ entry. Presumably, STIM1 serves as an essential sensor for the state of depletion of calcium stores.
Effects of overexpressed Orai1 or TRPC5 with STIM1 on store-operated Sr2+ and Ca2+ entry
To verify that Orai1, TRPC5, and STIM1 acted in combination to facilitate entry of either Ca2+ or Sr2+, we examined the effects of overexpression of these molecules in RBL-2H3 cells. Of note, overexpression of STIM1 along with Orai1 or TRPC5 permitted constitutive entry of these cations in nonstimulated cells. This finding is illustrated in Figs. 5⇓ and 6⇓, which show typical traces for fura 2-loaded cells before and after depletion of intracellular Ca2+ stores with thapsigargin.
Overexpression of STIM1 by itself resulted in little or no constitutive entry of Sr2+ but it did result in modest influx of Ca2+ in the absence of stimulation (Fig. 5⇑, A and B). Otherwise, the overexpression of STIM1 did not affect the stimulated entry of either Sr2+ or Ca2+ after depletion of Ca2+ stores with thapsigargin (Fig. 5⇑C). In contrast to STIM1, overexpression of Orai1 did not permit constitutive entry of Sr2+ or Ca2+ and suppressed entry of these ions in thapsigargin-stimulated cells (Fig. 5⇑, D–F). However, coexpression of STIM1 with Orai1 permitted constitutive entry Ca2+, but not Sr2+ (Fig. 5⇑, G and H), and fully restored entry of Ca2+ after depletion of Ca2+ stores with thapsigargin, whereas Sr2+ entry remained blocked (Fig. 5⇑I). The suppression of thapsigargin-induced Ca2+ entry following overexpression of Orai1 and restoration of Ca2+ entry on coexpression with STIM1 has been noted by other investigators in different cell lines (41, 42, 43). Our results show, in addition, that the coexpression of STIM1 and Orai1 allows substantial constitutive entry of Ca2+ (Fig. 5⇑H), but excludes entry of Sr2+ under all conditions (Fig. 5⇑I).
In contrast to Orai1, overexpression of TRPC5 resulted in some constitutive entry of Sr2+ and Ca2+ (Fig. 6⇑A) and suppressed entry of both ions in thapsigargin-treated cells but to a lesser extent than Orai1 (Fig. 6⇑B). In combination with STIM1, the constitutive entry of Sr2+ and Ca2+ was substantially enhanced (Fig. 6⇑C) and entry of both ions was fully restored in thapsigargin-treated cells (Fig. 6⇑D).
These data demonstrate that transient coexpression of Orai1 and STIM1 allows influx of only Ca2+ (Figs. 5⇑ and 6⇑), whereas coexpression of TRPC5 and STIM1 allows influx of Sr2+ as well as Ca2+. The studies with shRNAs (Figs. 3⇑ and 4⇑) had indicated that endogenous TRPC5 as well as STIM1 and Orai1 are required for optimal entry Sr2+ and Ca2+ following depletion of Ca2+ stores with thapsigargin to suggest that all three proteins interact with each other.
TRPC5, STIM1, and Orai1 are required for cation influx and degranulation in Ag-stimulated cells
As with thapsigargin-stimulated cells, influx of Sr2+ and Ca2+ was significantly suppressed when cells were stimulated with Ag following knock down of TRPC5, STIM1, or Orai1 (Fig. 7⇓, B–D) as compared with empty vector transfected cells (Fig. 7⇓A). Ag-stimulated release of intracellular Ca2+ was unaffected. Knock down of each of these proteins also significantly impaired degranulation in Ag-stimulated cells (Fig. 7⇓E). As in previous studies (13), Sr2+ could substitute for Ca2+ in promoting degranulation, partially so at 1 mM and fully so at 3 mM Sr2+ (Fig. 7⇓F). If TRPC5 is the carrier of Sr2+ as the preceding data suggests, TRPC5 would appear to be essential for degranulation as well as SOCE.
The original studies of Ca2+-dependent exocytosis (11, 12) and our early studies (2, 13) showed that Sr2+ and other divalent cations are taken-up by stimulated mast cells and can replace Ca2+ in promoting degranulation. These observations preclude CRAC as the sole mechanism for conveyance of Ca2+ into the cell because of the high Ca2+ selectivity of CRAC (62, 63) and the diminution of CRAC activity on replacement of Ca2+ with other divalent cations (9, 10).
It has been suggested that the channel properties of Orai1 may be modified by other proteins because the properties of CRAC, especially its low conductivity and specificity for Ca2+, are not entirely recapitulated by overexpressed Orai1 and STIM1 (32). We suggest that TRPC5 may be one such protein that allows entry of Sr2+ and Ca2+ into RBL-2H3 cells. Although influx of these two ions is readily distinguishable (Figs. 5⇑ and 6⇑), entry of both are dependent on TRPC5 (Fig. 3⇑D), STIM1 (Fig. 4⇑A), and Orai1 (Fig. 4⇑B). A plausible scenario is that entry is dependent on formation of a ternary complex of TRPC5/Orai1/STIM1 in which STIM1 acts as a Ca2+-sensor and Sr2+-permeable TRPC5 in conjunction with Orai1 enhances entry of Ca2+. This scenario is analogous to the proposals based on a ternary complex of TRPC5/Orai1/STIM1 (44), the interaction of Orai1 with TRPC3 or TRPC6 (45), and the interaction of STIM1 with various TRPC channel proteins (46). These interactions have been demonstrated by coimmunoprecipitation of expressed tagged proteins (44, 45, 46) or, in one case, endogenous TRPC1 with STIM1 (44). Our attempts to demonstrate similar associations for endogenous proteins in RBL-2H3 cells were unsuccessful because of the limitations of the available Abs. As noted earlier, overexpression of Orai1 suppresses Ca2+ entry (41, 42, 43). This and our finding that suppression of Sr2+ or Ca2+ influx by overexpression of Orai1 (Fig. 5⇑D) or TRPC5 (Fig. 6⇑B) could indicate that channel competency requires stoichiometric interaction of Orai1 or TRPC5 with STIM1. One possibility is that overexpressed Orai1 may preferentially oligomerize with itself (64, 65) rather than with TRPC5, and vice versa, with loss of conductance of Sr2+ and Ca2+ through a TRPC5-Orai1 complex. A caveat is that overexpression may lead to protein interactions that do not normally occur. The constitutive entry of cations noted in this study with overexpressed STIM1 in combination with Orai1 or TRPC5 might indicate such interactions.
A possible indication that Orai1 and TRPC5 are both necessary for cation influx in RBL-2H3 cells is that overexpression of either protein with STIM1 did not further enhance thapsigargin-induced Ca2+ influx (Figs. 5⇑ and 6⇑). This finding is in contrast to studies in human embryonic kidney HEK293 cells (41, 42, 43) where such influx was enhanced by overexpression of Orai1 and STIM1. However, thapsigargin fails to stimulate influx of Sr2+ in HEK293 cells (66) and it is unlikely that TRPC proteins interact with Orai1 for Ca2+ entry in these cells. If, however, both Orai1 and TRPC5 are components of CRAC, as appears to be the case in RBL-2H3 cells, then overexpression of one of these two CRAC proteins would not further enhance entry of Ca2+ or Sr2+ because ion conductance would still be dependent on endogenous levels of the other CRAC protein.
Although TRPC5 as well as Orai1 and STIM1 are essential for influx of Ca2+ and degranulation in RBL-2H3 cells (Fig. 7⇑), we cannot exclude a minor contribution of TRPC1, although this protein is thought not to be an endogenous component of CRAC in RBL-2H3 cells (44). Also, given the phenotypic diversity of mast cells (67, 68) and the possible variations in the combinations of Orai and TRPC family members, it is possible that such combinations may vary among different subtypes of mast cells.
We thank Sachini Bandara (Laboratory of Allergic Diseases, National Institute of Allergy and Infectious Diseases) for skillful assistance in the performance of RT-PCR.
The authors have no financial conflict of interest.
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.
↵1 This work was supported by the Intramural Programs of the National Heart, Lung, and Blood Institute and the National Institute of Allergy and Infectious Diseases.
↵2 Address correspondence and reprint requests to Dr. Michael A. Beaven or Dr. Hong-Tao Ma, Laboratory of Molecular Immunology, Room 8N109, Building 10, National Heart, Lung, and Blood Institute, National Institutes of Health, Bethesda MD 20892-1760. E-mail addresses: or
↵3 Current address: Division of Hematology, Center for Biologics Evaluation and Research, Food and Drug Administration, 8800 Rockville Pike, Bethesda, MD 20892.
↵4 Current address: Department of Dermatology, Hiroshima University Faculty of Medicine, 1-2-3, Kasumi, Minami-ku 734-8551, Hiroshima, Japan.
↵5 Abbreviations used in this paper: SOCE, store-operated calcium entry; TRP, transient receptor potential; TRPC, canonical TRP; CRAC, calcium release-activated calcium; ICRAC, CRAC current; eYFP, enhanced yellow fluorescent protein; PLC, phospholipase C; siRNA, small inhibitory RNA; shRNA, short hairpin RNA.
- Received December 4, 2007.
- Accepted December 4, 2007.
- Copyright © 2008 by The American Association of Immunologists