NGF Responsiveness Research Articles

Electrically controlled release of the nerve growth factor from a collagen-carbon nanotube composite for supporting neuronal growth.

We report the use of a nerve growth factor-loaded collagen-carbon nanotube (NGF-Col-CNT) composite as a substrate for in vitro growth of PC 12 cells. The feasibility of such novel composites as drug carriers was assessed using an in vitro drug-release test (ELISA immunoassay) in the presence of an electrical field. Subsequent studies included four-point probes, cyclic voltammograms, and in vitro PC 12 growth of fibers. The facile incorporation of NGF within fibrous collagen structures significantly mediated PC 12 (nerve) fiber process formation during the application of electrical voltage. Indeed, electrical stimulation (500 mV) using non-polarized Ag/AgCl electrodes to electro-active composites facilitated the sustained release of NGF. Given that PC 12 cells require NGF in the media to produce fiber outgrowth, this assay was clearly informative when the differentiation and branching of PC 12 cells into neurons was significantly enhanced. NGF responses were also evident from other developmental events including protrusions of neuronal filopodia and the formation of microspikes. Col-CNT composites, when electrically stimulated, could serve not only as neuronal scaffolds, but also as an electrically controllable drug release system to improve the performance and preferential timing of growth factor release in various biomedical applications.

Quantitative Automated Microscopy (QuAM) Elucidates Growth Factor Specific Signalling in Pain Sensitization

BackgroundDorsal root ganglia (DRG)-neurons are commonly characterized immunocytochemically. Cells are mostly grouped by the experimenter's eye as "marker-positive" and "marker-negative" according to their immunofluorescence intensity. Classification criteria remain largely undefined. Overcoming this shortfall, we established a quantitative automated microscopy (QuAM) for a defined and multiparametric analysis of adherent heterogeneous primary neurons on a single cell base.The growth factors NGF, GDNF and EGF activate the MAP-kinase Erk1/2 via receptor tyrosine kinase signalling. NGF and GDNF are established factors in regeneration and sensitization of nociceptive neurons. If also the tissue regenerating growth factor, EGF, influences nociceptors is so far unknown. We asked, if EGF can act on nociceptors, and if QuAM can elucidate differences between NGF, GDNF and EGF induced Erk1/2 activation kinetics. Finally, we evaluated, if the investigation of one signalling component allows prediction of the behavioral response to a reagent not tested on nociceptors such as EGF.ResultsWe established a software-based neuron identification, described quantitatively DRG-neuron heterogeneity and correlated measured sample sizes and corresponding assay sensitivity. Analysing more than 70,000 individual neurons we defined neuronal subgroups based on differential Erk1/2 activation status in sensory neurons. Baseline activity levels varied strongly already in untreated neurons. NGF and GDNF subgroup responsiveness correlated with their subgroup specificity on IB4(+)- and IB4(-)-neurons, respectively. We confirmed expression of EGF-receptors in all sensory neurons. EGF treatment induced STAT3 translocation into the nucleus. Nevertheless, we could not detect any EGF induced Erk1/2 phosphorylation. Accordingly, intradermal injection of EGF resulted in a fundamentally different outcome than NGF/GDNF. EGF did not induce mechanical hyperalgesia, but blocked PGE2-induced sensitization.ConclusionsQuAM is a suitable if not necessary tool to analyze activation of endogenous signalling in heterogeneous cultures. NGF, GDNF and EGF stimulation of DRG-neurons shows differential Erk1/2 activation responses and a corresponding differential behavioral phenotype. Thus, in addition to expression-markers also signalling-activity can be taken for functional subgroup differentiation and as predictor of behavioral outcome. The anti-nociceptive function of EGF is an intriguing result in the context of tissue damage but also for understanding pain resulting from EGF-receptor block during cancer therapy.

Time encoding by ERKs

![Graphic][1] An EGF response (top) is transient but an NGF response (bottom) is sustained. KURODA/MACMILLANThere are many more biological processes than there are signaling pathways. How, then, do multiple signals converge on one pathway and yet elicit diverse responses? Satoru

Neurotrophin Signaling through the p75 Receptor Is Deficient intraf6-/- Mice

Activation of the neurotrophin receptor p75 has been shown to elicit opposing cellular signals. Depending on the context of the cell, p75 will either promote survival or induce apoptosis after neurotrophin stimulation. p75-induced apoptosis occurs through activation of c-Jun N-terminal kinase (JNK), whereas the survival signal is mediated by nuclear factor kappaB (NFkappaB). The receptor proximal signals that produce these responses are unknown, although several molecules have been identified that associate with the intracellular domain of p75. One such interactor, TRAF6, a member of the tumor necrosis factor receptor-associated factor family, has been implicated in p75 signaling. To assess the role of TRAF6 in p75 signaling, we analyzed mice with this gene deleted. In Schwann cells isolated from traf6+/+ animals, NGF elicited an 80% increase in transcription of an NFkappaB reporter; however, in traf6-/- cells, the NGF response was abrogated. Similarly, NGF activation of JNK was not apparent in Schwann cells from mice lacking traf6. Deficiencies in p75 signaling in traf6-/- animals resulted in a loss of p75-mediated apoptosis. In sympathetic neurons cultured from traf6+/+ superior cervical ganglia (SCGs), there was an increase in JNK activation and apoptosis after BDNF binding to p75; however, traf6-/- neurons did not respond. In vivo during naturally occurring cell death, there was a 55.6% reduction in TUNEL (terminal deoxynucleotidyl transferase-mediated biotinylated UTP nick end labeling)-positive cells in the SCG of postnatal day 4 traf6-/- animals relative to traf6+/+ littermates. These results indicate that TRAF6 plays an essential role in mediating p75 signal transduction and induction of apoptosis.

Effect of Oligodendroglial Microdomain Components on NGF Signaling

Recent results indicate that the integrity of the myelin sheath is compromised with age [1], and that myelin destruction could be a precipitating event in age-related disorders [2]. Furthermore, oligodendrocytes (OL), which furnish the CNS myelin, do not support remyelination in older animals to the same extent as in younger animals. An increased process formation reflects an OL response to NGF, however, differentially depending on age of the donor [3]. We were interested to know as to whether NGF signaling is modulated by microdomain components such as caveolin and cholesterol, which might undergo age-related changes. Immunocytochemistry of cultured pig OL revealed the co-expression of caveolin-1 and the 140 kDa NGF receptor TrkA. Caveolin-containing microdomains were isolated via previously published buoyant density centrifugation methods (±Triton X-100) and by using MACS technology. Western blotting showed a co-labeling of the caveolar protein, flotillin-1, in addition to TrkA, which is enriched in the Triton X-100 insoluble fraction, p75 NTR, and p21 Ras. Preliminary results indicate that oligodendroglial caveolin and cholesterol are up-regulated from 8 DIV to 16 DIV, a period of extensive oligodendroglial process regeneration. Cells exposed to PEG-cholesterol increased their process formation; PEG-cholesterol plus NGF accelerated the NGF response; under both conditions, an in-gel-kinase assay demonstrated an increased MAPK activity, a step of the downstream TrkA signaling. On the other hand, exposure to cyclodextrin (1.2 mM), which disrupts caveolar microdomains by removing cholesterol from the plasma membrane, resulted in a less effective NGF response. The results at present indicate that NGF signaling is modulated by cholesterol; the effect of caveolin-1 siRNA is under investigation.

Altered immediate early gene expression in injured diabetic nerve: implications in regeneration.

To study the role that immediate early gene responses may play in impaired nerve fiber regeneration in diabetes, diabetic male BB/Wor rats were subjected to sciatic nerve crush at 6 wk of diabetes. Sciatic nerve mRNA expression of IGF-I, IGF-1-receptor, NGF, and p75 (low affinity NGF receptor), as well as protein expression of C-FOS, were examined at various time points following crush injury and compared with age- and sex-matched nondiabetic BB/Wor rats. Diabetic rats showed a delay in the early peak expression of IGF-1, C-FOS, NGF, and p75. The earliest immediate gene responses were those of IGF-I and IGF-1-receptor, which peaked at 0.5 h post-crush in control rats. In diabetic rats, IGF-1 peaked at 24 h whereas IGF-1-receptor mRNA revealed no early peak. The early NGF mRNA expression showed a maximum response at 6 h and of p75 at 4 days post-crush in control rats, whereas in diabetic rats they occurred at 2 days and 6 days, respectively. C-FOS protein expression showed a maximum at 6 h in control rats and in diabetic animals an attenuated peak was present at 2 days. These data provide the first evidence that immediate early gene responses are delayed in diabetes following sciatic nerve crush injury. The delayed IGF-1 expression may affect C-FOS induction and may be responsible for the delay in the NGF response in diabetic rats. The delayed immediate early gene responses precede the previously described perturbed macrophage recruitment and delayed Wallerian degeneration in this type I model and provide a possible explanation for impaired nerve regeneration in diabetes.

NGF-mediated alteration of NF-κB binding activity after partial immunolesions to rat cholinergic basal forebrain neurons

There are age-associated cognitive and cholinergic deficits in the neurotrophin-dependent cholinergic basal forebrain neurons (CBFNs). There are also increases in the activity of the transcription factor NF-κB in the aged rodent brain that may reflect chronic enhancement of stress response signaling. We used partial immunolesions (PIL) to CBFN to examine the role of endogenous NGF on choline acetyltransferase (ChAT) activity and NGF-mediated NF-κB alteration after cholinergic deafferentation. We injected 192 IgG-saporin, an immunotoxin selectively taken up by neurotrophin receptor p75NTR-bearing neurons, into lateral ventricles, followed by infusions of anti-NGF to assess NF-κB, ChAT and NGF responses to PIL after anti-NGF infusion. Treatment with anti-NGF decreased ChAT activity by 17–34% in the cortex, hippocampus, and olfactory bulb and PIL decreased ChAT activity by 47–73%. Changes in AChE activity levels paralleled those observed for ChAT after PIL. NGF protein levels in the olfactory bulb, but not the cortex or hippocampus, increased significantly after PIL treatment. Infusion of anti-NGF abolished the PIL-induced eight-fold NGF increase in CNS. NF-κB binding activity to the IgG-κB and ChAT specific NF-κB consensus sequences, increased in the cortex but not hippocampus after PIL followed by anti-NGF infusion. It is likely that immunolesion-induced changes in ambient NGF levels may perturb NF-κB activity.

Overexpression of atypical PKC in PC12 cells enhances NGF-responsiveness and survival through an NF-kappaB dependent pathway.

Removal of atypical PKC blocks NGF-induced differentiation of PC12 cells.1 We now examine the consequences that overexpression of atypical PKCs had upon NGF responses. PC12 cells were stably transfected with either PKC-iota or PKC-zeta. Overexpression of atypical PKCs markedly enhanced NGF- induced neurite outgrowth as well as enhanced NGF-stimulated JNK kinase. Cotransfection of HA-JNK1 along with increasing concentrations of PKC-iota, resulted in dose-dependent phosphorylation of GST c-Jun (1 - 79). NGF treatment of PC12 cells resulted in activation of NF-kappaB. In comparison, overexpression of atypical PKC-iota was by itself sufficient to activate NF-kappaB and shift the kinetics of NGF-induced kappaB activity. Furthermore, transfection of full-length antisense PKC-iota blocked basal and NGF-stimulated NF-kappaB. Differentiated and undifferentiated PC12 cells overexpressing atypical PKC-iota were protected from serum deprivation-induced cell death. Collectively, these findings demonstrate that atypical PKC-iota lies in a pathway that regulates NF-kappaB and contributes to both neurotrophin-mediated differentiation and survival signaling.

The transcriptional corepressor NAB2 inhibits NGF-induced differentiation of PC12 cells.

The PC12 pheochromocytoma cell line responds to NGF by undergoing growth arrest and proceeding to differentiate toward a neuronal phenotype. Among the early genetic events triggered by NGF in PC12 cells are the rapid activation of the zinc finger transcription factor Egr1/NGFI-A, and a slightly delayed induction of NAB2, a corepressor that inhibits Egr1 transcriptional activity. We found that stably transfected PC12 cells expressing high levels of NAB2 do not differentiate, but rather continue to proliferate in response to NGF. Inhibition of PC12 differentiation by NAB2 overexpression was confirmed using two additional experimental approaches, transient transfection, and adenoviral infection. Early events in the NGF signaling cascade, such as activation of MAP kinase and induction of immediate-early genes, were unaltered in the NAB2-overexpressing PC12 cell lines. However, induction of delayed NGF response genes such as TGF-beta1 and MMP-3 was inhibited. Furthermore, NAB2 overexpression led to downregulation of p21(WAF1), a molecule previously shown to play a pivotal role in the ability of PC12 cells to undergo growth arrest and commit to differentiation in response to NGF. Cotransfection with p21(WAF1) restored the ability of NAB2-overexpressing PC12 cells to differentiate in response to NGF.

Time course of changes in cholinergic and neurotrophin-related markers after infusion of colchicine into the basal forebrain

After bilateral infusions of colchicine or vehicle in the rat nucleus basalis magnocellularis, the time course of changes in several cholinergic and neurotrophin-related markers were assessed. Animals were sacrificed at 3, 7, 14, 28, 35 and 84 days post-lesion, and both the NBM and cortical areas were assessed. Sections were stained immunohistochemically for choline acetyltransferase (ChAT) or p140 trk ( trk) or histochemically for acetylcholinesterase (AChE). ChAT activity and neurotrophin protein levels were assessed regionally. The number of ChAT immunoreactive NBM neuronal profiles decreased beginning 3 days post-lesion and reach maximal loss by 28 days post-lesion, with no recovery. Examination of trk-IR around the NBM revealed a time-dependent decrease in trk-IR of magnocellular neurons and an increase in trk-IR of astrocytes at 14 and 28 days post-lesion. The density of AChE-stained cortical fibers was maximally decreased 3 days post-lesion followed by an increase in fiber staining across the remaining time points. Cortical ChAT activity showed the largest decrease at 7 days followed by recovery 84 days after colchicine infusion. There was an increase in NGF in the parietal cortex after colchicine infusion but no change in BDNF level. These patterns of changes in the cholinergic and neurotrophin-related markers suggest an association between NGF and lesion-induced compensatory responses in the basal forebrain cholinergic system.

Role of the bZIP transcription factor IREBF1 in the NGF induction of stromelysin-1 (transin) gene expression in PC12 cells.

Stromelysin-1 (ST-1) is one of the most nerve growth factor-(NGF) responsive gene products expressed in PC12 cells. In previous work, we identified a novel NGF-responsive element in the proximal promoter region of the ST-1 gene that participates in this induction, and showed that it bound a protein present in the nuclei of PC12 cells. Here, we identify a transcription factor that specifically recognizes this regulatory element-the interferon-response element binding factor-1 (IREBF1), a member of the basic leucine zipper gene family. We show that IREBF1 is constitutively expressed in PC12 cells and that overexpression of IREBF1 augments NGF-responsive ST-1 gene regulation, but does not affect basal levels of expression. On the other hand, expression of a mutated form of this transcription factor lacking the DNA binding domain attenuated NGF responsiveness, without affecting basal levels of expression. These data suggest that IREBF1 is part of the NGF-responsive transcriptional machinery necessary for the expression of ST-1 in PC12 cells.

Retinoic Acid-Mediated Increase in TrkA Expression Is Sufficient to Elicit NGF-Dependent Survival of Sympathetic Neurons

Sympathetic neurons depend on the classical neurotrophin NGF for survival by the time they innervate their targets, but the mechanisms controlling the onset of NGF responsiveness in developing neuroblasts have not been defined. Immature chick sympathetic neurons are unresponsive to NGF, but express low mRNA levels of the high-affinity NGF receptor trkA. Treatment with retinoic acid (RA) leads to increased levels of both trkA mRNA and protein, a response mediated through retinoic acid receptor alpha (RARα). Ectopic expression of trkA in these cells results in the ability to survive with NGF, suggesting that RA-induced trkA expression is sufficient to elicit NGF-dependent survival. Our data establish a mechanism controlling NGF responsiveness and implicate a function for RA at defined late stages of neuron development.

TrkA expression in the CNS: evidence for the existence of several novel NGF-responsive CNS neurons

NGF acts as a neurotrophic factor by binding and activating its receptor on certain neuronal populations in the CNS and PNS. TrkA is a receptor for NGF. Recent findings in vitro indicate that this NGF-activated receptor tyrosine kinase transduces the NGF signal. To further define NGF actions in the CNS, we examined trkA expression in the adult rat brain. We found that trkA mRNA and immunoreactivity (IR) coincided in specific, defined neuronal populations in the forebrain and brainstem. In addition to cholinergic neurons in the basal forebrain and neostriatum, trkA expression was found in noncholinergic neurons in (1) the paraventricular anterior and reuniens thalamic nuclei, (2) the rostral and intermediate subnuclei of the interpeduncular nucleus (IPN), (3) scattered neurons in the ventrolateral and paramedian medulla, (4) the prepositus hypoglossal nucleus, and (5) the area postrema. NGF responsiveness was demonstrated for each of these populations. In contrast to trkA, p75NGFR was found only in a minority of NGF-responsive populations. Our data provide further evidence that expression of trkA marks NGF-responsive CNS neurons and suggests novel roles for NGF in the brain.

Neutrophins and neurite outgrowth in the peripheral nervous system

In this review we summarize evidence for the involvement of neutrophins in the regulation of neurite outgrowth in the peripheral nervous system. Recent studies have shown that many neurons at early developmental stages respond to NT3 by increased survival and morphological differentiation in vitro. In addition, NT3 mRNA can be detected at this time in embryonic tissues trespassed by outgrowing neurites. Later during development, the NT3 response may be lost and replaced by a NGF response. The data show that an early NGF-independent, NT3-sensitive and a late NGF-responsive period can be distinguished for different developing peripheral neurons in vitro and raise the possibility that neurite outgrowth may be influenced by the different neutrophins, acting at distinct developmental periods in vivo.

Viral Transduction of trkA into Cultured Nodose and Spinal Motor Neurons Conveys NGF Responsiveness

The trkA gene encodes the tyrosine kinase receptor that transduces the NGF signal. We have used a defective herpes simplex virus-1 (HSV-1) vector containing a full-length rat trkA sequence (pHSVtrkA) to express NGF receptors in primary cultures of neonatal nodose neurons and embryonic spinal motor neurons which are normally unresponsive to NGF. Transduction of rat neonatal nodose ganglion neurons and spinal motor neurons with pHSVtrkA resulted in expression of functional NGF receptors. Stimulation of these receptors by NGF resulted in the survival and the morphologic and biochemical differentiation of these neurons. These observations indicate that nodose and spinal motor neurons can express the downstream signal transduction apparatus necessary to mediate the survival and differentiating effects mediated by ligand binding to trkA and that transduction with pHSVtrkA can effectively convert NGF nonresponsive neurons into responsive ones.

Trk receptors use redundant signal transduction pathways involving SHC and PLC-γ1 to mediate NGF responses

In response to NGF, the Trk receptor tyrosine kinase forms a complex with SHC, a protein that couples receptor tyrosine kinases to p21ras. Complex formation between Trk and SHC, SHC tyrosine phosphorylation, and association of SHC with Grb2 were mediated by autophosphorylation at Y490 in Trk [sequence: see text]. To determine the role of SHC and other Trk substrates in NGF signaling, Trk receptors with mutations in Y490 and Y785 (the PLC-gamma 1 association site) were introduced into PC12nnr5 cells. NGF treatment of PC12nnr5 cells expressing Trk with mutations in either substrate-binding site resulted in normal neurite outgrowth and Erk1 activity and tyrosine phosphorylation. However, PC12nnr5 cells expressing Trk with mutations at both sites failed to stably extend neurites and efficiently induce Erk1 activity and tyrosine phosphorylation in response to NGF. We postulate that Trk receptors can activate Erk1 by either SHC- or PLC-gamma 1-dependent signaling pathways. These results suggest a model whereby Trk receptors utilize at least partially redundant signal transduction pathways to mediate NGF responses.

Effects of nerve growth factor on catalase and glutathione peroxidase in a hydrogen peroxide-resistant pheochromocytoma subclone

Stepwise selection in increasing H 2O 2 concentrations was used to obtain a PC12 cell variant designated HPR. This variant was stably resistant to H 2O 2 as compared with the parental PC12 cell line. HPR cells responded to nerve factor (NGF) by further enhancing H 2O 2 resistance. This variant was subcloned by limiting dilution to obtain the line referred to as HPR-C, which was stably resistant to H 2O 2 toxicity and retained NGF responses, including morphologic changes and further reduction of H 2O 2 toxicity. When compared with the parental PC12 line, the HPR-C subclone did not have higher levels of catalase or glutathione peroxidase (GSH Px) activity or mRNA expression (as assessed by PCR analysis of cDNA reverse transcribed from total cellular RNA). HPR-C cells retained the ability to respond to NGF treatment by increasing catalase and GSH Px activity and expression. These data suggest that the protective effects of conditioning lesions, unlike those of neurotrophins, are in part independent of changes in the activity or expression of antioxidant enzymes.

Ectopic trkA expression mediates a NGF survival response in NGF-independent sensory neurons but not in parasympathetic neurons.

We have investigated the role of trkA, the tyrosine kinase NGF receptor, in mediating the survival response of embryonic neurons to NGF. Embryonic trigeminal mesencephalic (TMN) neurons, which normally survive in the presence of brain-derived neurotrophic factor (BDNF) but not NGF, become NGF-responsive when microinjected with an expression vector containing trkA cDNA. In contrast, microinjection of ciliary neurotrophic factor (CNTF)-dependent embryonic ciliary neurons with the same construct does not result in the acquisition of NGF responsiveness by these neurons despite de novo expression of trkA mRNA and protein. The failure of trkA to result in an NGF-promoted survival response in ciliary neurons is not due to absence of the low-affinity NGF receptor, p75, in these neurons. Quantitative RT/PCR and immunocytochemistry showed that TMN and ciliary neurons both express p75 mRNA and protein. These findings not only provide the first direct experimental demonstration of trkA mediating a physiological response in an appropriate cell type, namely NGF-promoted survival of embryonic neurons, but indicate that not all neurons are able to respond to a trkA-mediated signal transduction event.

Neurotrophin signalling in the nervous system

Neurotrophins appear to have important trophic roles in thedeveloping and mature nervous system. The discovery that members of the trk gene family are signalling receptors for the neurotrophins suggests that studies of trk gene expression will be informative with respect to the locus, timing and biological significance of neurotrophin actions. NGF is a target-derived neurotrophin. Localization of trkA expression is predictive of NGF responsiveness in neurons in both the PNS and CNS. The onset of trk expression in the developing PNS appears to allow for NGF signalling as soon as neurites contact NGF in the target. Studies on NGF and trkA raise the possibility that neurotrophins may normally act via internalization of their trk receptors with subsequent retrograde transport of the activated receptor from the neurite tip to the cell body. In the CNS and in PC12 cells, trkA gene expression is induced by NGF. Regulation of trkA by NGF may compensate for receptor internalization and degradation. The localization of expression for trkB and C is more widespread than trkA, and novel trophic relationships are suggested by the complicated, sometimes overlapping patterns for BDNF and NT-3 and their corresponding trk mRNAs. While for some neurons the actions of BDNF, NT-3 and NT-4/5 may be similar to NGF, in other cases there may be autocrine or paracrine modes of action. Understanding where and how activated trk's signal is key to elucidating the actions of NGF and the other neurotrophins.

Transfection with trk restores "slow" NGF binding, efficient NGF uptake, and multiple NGF responses to NGF-nonresponsive PC12 cell mutants

NGF binds to and activates the protein tyrosine kinase gp 140prototrk. Expression of this receptor is required for at least some responses to NGF. Three outstanding issues are addressed in the present work. First, we determined whether expression of gp 140prototrk is required for all neuronal NGF responses. Second, we examined the role of gp 140prototrk in NGF binding and internalization. Third, we addressed the utility of NGF-nonresponsive PC12nnr5 cells for study of the NGF mechanism. In contrast to wild-type PC12 cells, PC12nnr5 cells do not express endogenous gp 140prototrk. We therefore asked whether they possess other defects that compromise NGF signaling pathways. To answer these questions, we transfected PC12nnr5 cells with a cDNA encoding full-length human gp 140prototrk and isolated cell lines permanently expressing the receptor. Introduction of trk rescued all of the many and varied NGF responses assessed, including enhanced protein tyrosine phosphorylation, induction of immediate-early and neural-specific genes, neurite outgrowth and regeneration, maintenance of survival in serum-free medium, and stimulation of AChE activity. In contrast to PC12nnr5 cells, the trk-transfected lines also bind and internalize NGF with wild-type PC12 cell characteristics. These findings indicate that gp 140prototrk is required for many, if not all, responses of neuronal cells to NGF and is necessary for proper NGF binding and internalization. Additionally, as no signaling defect other than the absence of trk expression was revealed in PC12nnr5 cells, this work supports the utility of this line for genetic dissection of the NGF mechanism of action.