Spinal protein phosphatase 1 constrains respiratory plasticity after sustained hypoxia

Abstract

Plasticity is an important aspect of the neural control of breathing. One well-studied form of respiratory plasticity is phrenic long-term facilitation (pLTF) induced by acute intermittent but not sustained hypoxia. Okadaic acid-sensitive protein phosphatases (PPs) differentially regulate phrenic nerve activity with intermittent vs. sustained hypoxia, at least partially accounting for pLTF pattern sensitivity. However, okadaic acid inhibits multiple serine/threonine phosphatases, and the relevant phosphatase (PP1, PP2A, PP5) for pLTF pattern sensitivity has not been identified. Here, we demonstrate that sustained hypoxia (25 min, 9–10.5% O₂) elicits phrenic motor facilitation in rats pretreated with bilateral intrapleural injections of small interfering RNAs (siRNAs; Accell-modified to preferentially transfect neurons, 3.33 μM, 3 days) targeting PP1 mRNA (48 ± 14% change from baseline, n = 6) but not PP2A (14 ± 9% baseline, n = 6) or nontargeting siRNAs (4 ± 10% baseline, n = 7). In time control rats (no hypoxia) treated with siRNAs (n = 6), no facilitation was evident (−9 ± 9% baseline). siRNAs had no effect on the hypoxic phrenic response. Immunohistochemistry revealed PP1 and PP2A protein in identified phrenic motoneurons. Although PP1 and PP2A siRNAs significantly decreased PP1 and PP2A mRNA in PC12 cell cultures, we were not able to verify “knockdown” in vivo after siRNA treatment. On the other hand, PP1 and PP2A siRNAs significantly decreased PP1 and PP2A mRNA in PC12 cell cultures, verifying the intended siRNA effects. In conclusion, PP1 (not PP2A) is the relevant okadaic acid-sensitive phosphatase constraining phrenic motor facilitation after sustained hypoxia and likely contributing to pLTF pattern sensitivity.

NEW & NOTEWORTHY This study demonstrates that the relevant okadaic acid-sensitive Ser/Thr protein phosphatase (PP) constraining facilitation after sustained hypoxia is PP1 and not PP2A. It suggests that PP1 may be critical in the pattern sensitivity of hypoxia-induced phrenic motor plasticity.

INTRODUCTION

Pattern sensitivity is a hallmark of many forms of plasticity, including respiratory motor plasticity. In some cases, pattern sensitivity is conferred through the balance between protein kinases (which phosphorylate) vs. protein phosphatases (PPs; which dephosphorylate) to control target phosphorylation states (4, 26, 34). The role of the kinase/phosphatase balance in the pattern sensitivity of hypoxia-induced respiratory motor plasticity is not well understood (11). However, okadaic acid serine/threonine PPs constrain facilitation with sustained hypoxia (SH; 25-min hypoxia) but have no effect on acute intermittent hypoxia-induced phrenic motor facilitation (2, 41). Thus PPs play a key role in regulating pattern-sensitive, hypoxia-induced respiratory motor plasticity. However, we do not yet know the specific phosphatases constraining facilitation.

Of three serine/threonine PP families, our focus concerns the phosphoprotein phosphatases (PP1, PP2A, PP2B, PP4, PP5, PP6, and PP7). Since okadaic acid preferentially inhibits PP1, PP2A, PP4, and PP5 (20), we focused primarily on the more abundant okadaic-acid sensitive PPs: PP1 and PP2A (35, 40). While PP4 is ubiquitous in eukaryotes, its primary roles are associated with development (32) and injury responses (43). Thus it was considered unlikely to play a major role in adult respiratory motor plasticity. Although PP5 is also of interest, the pool of small interfering RNAs (siRNAs) used in this study to target PP5 had off-target effects; therefore, we limited our analysis of PP5 (see data in Fig. 5). This study provides new mechanistic insights concerning the regulation of hypoxia-induced respiratory motor plasticity, including emergent properties such as pattern sensitivity.

Fig. 5.

No facilitation or changes in immunofluorescence were evident after small interfering (si)RNA targeting protein phosphatase (PP) 5. However, gene expression data (Fig. 4) indicate siPP5 off-target effects, as it reduced PP1, PP2A, and PP5 gene expression in PC12 cells. As a result, phrenic (Phr) nerve recordings and immunohistochemistry data were not included in other analyses. siNT, nontargeting siRNA sequence; TC, time control.

METHODS

All experiments conformed to policies laid out by the National Institutes of Health in the Guide for the Care and Use of Laboratory Animals and were approved by the Institutional Animal Care and Use Committee at the School of Veterinary Medicine, University of Wisconsin-Madison. Experiments were performed on 3- to 4-mo-old Harlan male Sprague-Dawley rats (colony 211a). Rats were housed under standard conditions, with food and water ad libitum and a 12-h light-dark cycle.

RNA Interference with siRNAs

As described previously (6, 10), Accell-modified siRNAs (Dharmacon, Thermo Scientific) consisting of pools of four 21-duplexes were used since they preferentially transfect neurons. Four siRNAs targeting PP1 (siPP1; cat. no. E-100–263–00–0010), PP2A (siPP2A; cat. no. E-097852-00-0010), PP5 (siPP5; cat. no. E-088823-00-0010), and nontargeting (siNT; cat. no. D-001910-10) were aliquoted and stored at −20°C in Dharmacon siRNA buffer until needed. Each siRNA was prepared as previously described (6, 10) to a final siRNA concentration of 3.33 µM (20 µl siRNA, 6 µl 5× siRNA buffer, 3.2 µl Oligofectamine Transfection Reagent, and 0.8 µl of RNAse free H₂O) and mixed for 20 min before intrapleural injection to facilitate siRNA complexing with the transfection reagent.

Intrapleural Injections

Cholera toxin-B (CtB) subunit and siRNAs were injected into the intrapleural space as described previously (7, 8, 14, 16, 21, 28, 31). Briefly, anesthetized rats (4% isoflurane induction, 2% maintenance with 100% O₂) received bilateral intrapleural CtB (25 µg in 12.5 µl/side; Calbiochem, Billerica, MA) injections on day 1 (Fig. 1). On the subsequent 3 days (days 2–4), anesthetized rats received bilateral intrapleural siRNA injections (30 µl/side) (Fig. 1).

Fig. 1.

Experimental timeline to investigate the effects of small interfering (si)RNAs targeting protein phosphatases (PPs) after sustained hypoxia (Hx). Bilateral intrapleural injections of cholera toxin-b subunit were given on day 1, followed by 3 subsequent days of injections (days 2–4) of a siRNA [siPP1, siPP2A, siPP5, or small interfering nontargeting (siNT)]. On day 5, rats were used for electrophysiology analyses and subjected to sustained Hx (25 min, 10.5% O₂) before being harvested for immunohistochemical (IHC) protein analyses. CtB, cholera toxin-B.

Experimental Groups

Rats were randomly assigned to one of five groups to investigate the effects of PPs in restraining phrenic motor facilitation after sustained hypoxia (SH). These groups were as follows: 1) siPP1 + SH (n = 6); 2) siPP2A + SH (n = 6); 3) siPP5 + SH (n = 6); 4) siNT + SH (n = 7); and 5) time control (TC; no hypoxia) (1 siPP1, 2 siPP2A, 2 siPP5, 1 siNT) (n = 6).

Electrophysiological Experiments

The protocol for electrophysiological experiments has been described in detail previously (1, 3, 21, 22). Briefly, on day 5, rats were anesthetized with isoflurane, tracheotomized, and pump ventilated (Small Animal Ventilator 683; Harvard Apparatus, Holliston, MA) for the duration of the surgery. After surgery was completed, rats were slowly converted to urethane anesthesia (1.8 g/kg iv; Sigma-Aldrich). During the 1-h stabilization period after conversion to urethane, pancuronium bromide (1 mg iv) was given to paralyze the rats. Anesthetic level was assessed by monitoring blood pressure and phrenic nerve responses to toe pinch. Approximately 30 min into the stabilization period postanesthetic conversion, an intravenous infusion of 1.5–2 ml/h began with a solution consisting of Hetastarch (0.3%) and sodium bicarbonate (0.84%) in lactated Ringer solution. The infusion rate was adjusted to maintain blood volume, pressure, and acid-base balance throughout the experiment.

Surgical preparation.

Rats were vagotomized, and a catheter was inserted into the right femoral artery to measure blood pressure and take arterial blood samples. A rectal probe was used to monitor body temperature.

From a dorsal approach, the left phrenic nerve was isolated, cut distally, desheathed, and placed on a bipolar silver recording electrode submerged in mineral oil. Nerve activity was amplified (gain ×10 K), bandpass filtered (300 Hz to 20 kHz; A-M Systems, Carlsberg, WA), and integrated (absolute value, Powerlabs 830; AD Instruments, Colorado Springs, CO; time constant: 50 ms). The signal was digitized, recorded, and analyzed using Powerlabs 830 (version 7.2.2; AD Instruments).

Protocol.

Baseline nerve activity was established with $\text{F}{\mathrm{I}}_{{\text{O}}_2}$ ~0.56 ($\text{P}{\text{a}}_{{\text{O}}_2}$ >300 mmHg) and CO₂ added to the inspired gas (balance nitrogen). The CO₂ apneic threshold was determined by slowly lowering inspired CO₂ until phrenic nerve activity ceased. Inspired CO₂ was slowly increased until phrenic nerve activity resumed (recruitment threshold). End-tidal CO₂ was set 2 mmHg above the recruitment threshold to establish baseline nerve activity. End-tidal CO₂ was monitored and maintained throughout an experiment using a flow-through capnograph (Respironics, Andover, MA).

After phrenic nerve activity stabilized, an arterial blood sample was taken to establish baseline conditions; these conditions were maintained for the duration of the experiment. Blood samples (62.5 µl in heparinized plastic catheter) were analyzed for Po₂, Pco₂, pH, and base excess using a blood gas analyzer (ABL 800; Radiometer, Copenhagen, Denmark). After baseline conditions were established, a sustained hypoxia protocol began consisting of one continuous hypoxic episode (25-min duration, 9–10.5% O₂, Fig. 1) as described previously (41). Blood samples were taken during the hypoxic episode and 15, 30, and 60 min post-SH. Data were included in analysis only if they complied with the following criteria: 1) $\text{P}{\text{a}}_{{\text{O}}_2}$ during baseline and post-SH was >180 mmHg; 2) $\text{P}{\text{a}}_{{\text{O}}_2}$ during hypoxic episode was between 35 and 45 mmHg; and 3) $\text{P}{\text{a}}_{{\text{O}}_2}$ remained within 1.5 mmHg of baseline throughout the post-SH period. Maximum hypoxia + hypercapnia was assessed at the end of each experiment. Rats that did not respond to maximal stimulation were removed from analysis. Phrenic nerve amplitude was evaluated for 30 bursts before each blood sample. Upon completion of the experiment, rats were transcardially perfused and used for immunohistochemical analyses.

Immunofluoresence

Following electrophysiology experiments on day 5 (Fig. 1), spinal tissue was harvested to examine of PP1, PP2A, and PP5 protein expression in the phrenic motor nucleus. Rats were perfused with 4% paraformaldehyde, immersion fixed overnight, and dehydrated. The cervical spinal cord (C3–C6) was then cut into 40-µm sections using a freezing microtome (SM 2000R; Leica). Triple-label immunohistochemistry was used to assess PP expression compared with nontargeting controls. For each immunohistochemical run, rats receiving an siRNA for each of the three PP ran concurrently with tissue from a siNT rat to facilitate comparisons. Primary antibodies targeting PP1 (rabbit polyclonal; 1:100; Abcam), PP2A (rabbit polyclonal; 1:250; Sigma-Aldrich), or PP5 (rabbit polyclonal, 1:100; Sigma-Aldrich) were incubated with anti-CtB (goat polyclonal; 1:5,000; Calbiochem) and anti-GFAP (mouse monoclonal; 1:500; Millipore) primary antibodies. Secondary antibodies (Invitrogen) used were donkey anti-rabbit (1:1,000; AlexaFluor 488), donkey anti-goat (1:1,000; AlexaFluor 633), and donkey anti-mouse (1:1,000; AlexaFluor 555) to fluorescently tag primary antibodies. Appropriate negative controls were run concurrently to confirm specific staining of all antibodies.

Image processing.

All immunofluorescent images (1,024 × 1,024 pixels) were acquired on a Nikon C1 laser scanning confocal microscope with lambda strobing using the Nikon EZ-C1 Gold (Version 3.80) confocal imaging software (2-µm step increments were used to acquire z stacks). All image pairs (siPP1, siPP2a, or siPP5 vs. siNT) were identically adjusted for contrast and brightness in the Nikon EZ-C1 FreeViewer Gold (Version 3.90) software.

PC12 Cell Culture, siRNA Transfections, and Quantitative RT-PCR

Rat pheochromocytoma PC12 cells were cultured in a 37°C incubator at 8% CO₂, in growth medium [DMEM containing supplemental 5% FBS (Hyclone, Defined), 5% HIHS (Hyclone), 1× l-glutamine (ThermoFisher), and 100 U/ml penicillin/streptomycin (Corning Cellgro)]. Cells were cultured until ~90% confluency, and routinely passaged or fed every 2–3 days.

The day before siRNA transfection, cells were seeded in a 96-well plate at a density of 15,000 cells/well in 200 μl of growth medium per well. The following day, the growth medium was removed from the cells and replaced with 100 μl of Accell Delivery Media (cat. no. 005000-500; GE Dharmacon, Lafayette, CO) containing a final concentration of 1 μM of the appropriate nontargeting control Accell siRNA SMART pool (cat. no. D-001910-10), or Accell siRNA SMARTpools against Pp1cb (cat. no. E-100-263-00-0010), Pp2ca (cat. no. E-097852-00-0010), or Pp5 (cat. no. E-088823-00-0010). The cells were incubated for 72 h before the Delivery Media were removed and the cells harvested for analysis of gene knockdown.

The cells were collected into TriReagent (Sigma), and gene knockdown was assessed by quantitative RT-PCR as we have previously described (21, 22). Briefly, total RNA (0.5 µg) was reverse transcribed to cDNA using Moloney murine leukemia virus MMLV reverse transcriptase, and quantitative PCR for PP1, PP2 and PP5 was performed with Sybr Green (Applied Biosystems, Foster City, CA) using the following primer sequences: Ppp1: forward, 5′-CCTGCTGGAGGTACGAGGAT and reverse, 5′-CACAAATCTTCAGTGGCGCT; Ppp2: forward, 5′-GCCTCCCCTCTTGTTGTGTT, and reverse, 5′-CTGCGGTGACTTGTGCAGTA; and Ppp5: forward, 5′-TCGTTGCTGTGTCCCTTACC and reverse, 5′-GGTGTCCCCACACACTGTAA.

Gene expression analyses were performed using the dCT method (27).

Data Analysis

Electrophysiology and gene expression analyses.

Peak amplitude of integrated phrenic nerve activity was averaged for 30 bursts at each time of interest. Changes in phrenic nerve burst amplitude were normalized to baseline values (i.e., percent change from baseline). Phrenic amplitude is reported for baseline, during the sustained hypoxia episode, and at the 60 min post-SH time points only.

Statistical comparisons for hypoxic responses were made from data at minute 2 during the hypoxia episode using a one-way ANOVA (Kruskal-Wallis). Nonparametric analyses were used when data failed normality or equal variance.

Statistical comparisons for changes in phrenic burst amplitude after SH were made using a repeated measures two-way ANOVA with Fisher’s least significant difference post hoc test to identify individual differences (Sigma Stat version 11; Systat Software, San Jose, CA). Differences were considered significant if P < 0.05. All values are expressed as means ± SD. Analyses included time control rats treated with siPP5 (n = 2) since results were the same when included or excluded. All other siPP5 data were removed from statistical analyses (see data in Fig. 5).

Statistical comparisons for changes in gene expression were made on dCT values using a repeated measures one-way ANOVA with a Bonferroni post hoc test. All values are expressed as fold change means ± 1 SD relative to siNT.

RESULTS

siRNAs Targeting PP1 Reveal Phrenic Motor Facilitation

To identify which PP is important in constraining facilitation after SH, we used siRNAs targeting PP1 and PP2A, and a nontargeting siRNA control (siNT). As expected, no facilitation was observed post-SH in siNT treated rats (4 ± 10% change from baseline at 60 min post-SH; n = 7; Fig. 2, A and C), confirming neither SH nor siRNA treatment reveal facilitation. In contrast, SH-induced facilitation was revealed in siPP1-treated rats (48 ± 14% change from baseline, n = 6); this value was significantly increased vs. all other groups (Fig. 2, A and C; P ≤ 0.001). In siPP2A-treated rats, significant facilitation was not observed after SH (14 ± 9% change from baseline, n = 6) when compared with siNT rats, but was significantly increased vs. time controls (P = 0.029). In TC rats (n = 1 siPP1, n = 2 siPP2A, n = 2 siPP5, and n = 1 siNT) as expected, no facilitation was evident (−9 ± 9% change from baseline; Fig. 2, A and C). TC-treated rats were not significantly different from siNT rats (P = 0.228) but were significantly reduced vs. siPP1-treated rats (P < 0.001).

Fig. 2.

Facilitation after sustained hypoxia is revealed after small interfering (si)RNA targeting protein phosphatase (PP) 1, but not after nontargeting (NT), or PP2A siRNA. A: representative integrated phrenic (Phr) neurograms from anesthetized rats during sustained hypoxia or time control (no hypoxia). Dashed line indicates baseline phrenic amplitude in each trace. B: group data show treatment with siRNAs did not alter the magnitude of the short-term hypoxic responses (P = 0.278, Kruskal-Wallis one-way ANOVA). C: group data demonstrate a significant increase in phrenic nerve amplitude after siPP1 and sustained hypoxia compared with all other treatment groups. siPP2a had significantly greater phrenic nerve amplitude after sustained hypoxia compared only with the time control group (no hypoxia). ***P ≤ 0.001 significantly different from all other groups, #P < 0.05 significantly different from time control, two-way repeated measures ANOVA, Fisher least significant difference posttest.

Short-term hypoxic ventilatory responses were unaltered by siRNAs (Fig. 2B). SH evoked hypoxic phrenic responses with a 78 ± 10% increase from baseline for siNT, 74 ± 8% for siPP1, and 64 ± 3% for siPP2A; these values were not significantly different (P = 0.278).

siRNAs Had No Detectable Effect on PP Protein Expression

PPs are ubiquitously expressed in eukaryotes and are known to alter neural function; however, few have examined expression in the phrenic motor nucleus (41), and none have examined their expression in labeled phrenic motoneurons. We identified PP1 and PP2A protein expression throughout the phrenic motor pool in transverse sections from C3–C6 in rats treated with siNT, siPP1, and siPP2A (Fig. 3). Brightest staining was evident in cells colocalizing with CtB, indicating expression within phrenic motoneurons. PP1 staining was prevalent in the nucleus of CtB-labeled cells, with some cytoplasmic staining. For PP2A, staining was evident throughout CtB-labeled cells.

Fig. 3.

Protein phosphatase (PP) immunofluorescence was not different in the phrenic motor nucleus after small interfering (si)RNA targeting PP1 or 2A compared with rats treated with nontargeting siRNAs. Representative confocal images (×20) for rats given a nontargeting siRNA sequence (siNT, left) and those given targeted siRNA for each PP (middle) show no obvious differences in staining for each of the PPs. Representative overlay images (×20, right) show immunofluorescence for PP1 (A) and PP2A (B) colocalizing with cholera toxin B retrogradely labeled phrenic motoneurons (blue, right). Minimal colocalization was evident between PPs and microglia (CD11b, red, right).

PP Gene Expression Is Reduced by siRNAs in Cultured PC12 Cells

While immunofluorescence did not reveal evidence of siRNA target protein knockdown, cultured rat pheochromocytoma PC12 cells of neural cell origin were used to verify the efficacy and specificity of the specific siRNAs used in this study (Fig. 4). siPP1 significantly decreased PP1 gene expression in PC12 cells to 60% of that measured in siNT-treated cells (0.4 ± 0.03 fold; n = 5; P < 0.001); however, siPP1 did not significantly affect PP2A (0.8 ± 0.05 fold; n = 5) or PP5 (1.4 ± 0.3 fold; n = 5) mRNA. Similarly, siPP2A significantly decreased PP2A gene expression by 70% (to 0.3 ± 0.02 fold; n = 4; P < 0.001) but had no significant effect on PP1 (1.0 ± 0.1 fold; n = 4) or PP5 (0.9 ± 0.2 fold; n = 4) mRNA. Unfortunately, siPP5 significantly reduced PP1 (0.4 ± 0.04 fold; n = 4; P < 0.001), PP2A (0.3 ± 0.03 fold; n = 4; P < 0.001), and PP5 (0.3 ± 0.03 fold; n = 4; P = 0.001) mRNA, demonstrating a lack of specificity in the siRNA pool used here. Based on these results, we no longer considered the siPP5 a viable tool to investigate the impact of PP5 knockdown; consequently, siPP5 data (physiology, immunofluorescence) were not considered in our overall analyses but are presented in Fig. 5.

Fig. 4.

Small interfering (si)RNA treatment knocks down mRNA expression in PC12 cells. Average fold change (±SD) in gene expression relative to nontargeting siRNA (siNT, dotted line). Treatment with siRNA targeting protein phosphatase (PP) 1 knocks down PP1 gene expression (white bars, circles) and siRNA targeting PP2A knocks down PP2A gene expression (gray bars, squares). Treatment with siRNA targeting PP5 knocks down PP5 (black bars, triangles), PP2A, and PP1 gene expression. ***P ≤ 0.001 from siNT.

DISCUSSION

To further our understanding concerning mechanistic underpinnings of pattern sensitivity in hypoxia-induced respiratory motor plasticity, we investigated the differential involvement of PP1 vs. PP2A in constraining phrenic motor facilitation after SH. Okadaic acid-sensitive PPs are important for pLTF pattern sensitivity (41), but previous work did not identify the specific PP(s) involved. With the development of siRNAs to knock down targeted PPs, we demonstrate that siRNAs targeting PP1 reveal phrenic motor facilitation after SH, whereas siRNAs targeting PP2A mRNA have no effect. This is the first report concerning the roles of specific PPs, and we identify PP1 as a relevant okadaic acid-sensitive Ser/Thr PP constraining facilitation, contrary to previously stated hypotheses from our group (11, 41).

Significant advances have been made in understanding the mechanisms of respiratory motor plasticity (15), yet we have not yet elaborated an understanding of pLTF pattern sensitivity (10–12). Indeed, the cellular basis of pattern sensitivity in any form of neuroplasticity has seldom been investigated (12).

One major factor in pLTF pattern sensitivity is cross-talk inhibition between competing pathways to phrenic motor facilitation (12). Although a role for PP activity has been demonstrated (41), the details of its involvement are not yet understood. This study provides support for the key role of PP1 vs. PP2A activity, consistent with at least some other forms of pattern-sensitive central nervous system plasticity (5, 42).

Understanding factors constraining plasticity is important in our quest to understand this important aspect of neural function and, further, as we attempt to harness plasticity as a therapeutic modality to improve motor function in severe clinical disorders compromising breathing and other motor behaviors. Interest in exploiting hypoxia-induced spinal plasticity is growing since it has been shown to elicit functional recovery of breathing (37) and walking in people with chronic, incomplete spinal cord injuries (17, 29, 38). Thus enhancing our understanding of the mechanisms promoting and inhibiting plasticity is of significant clinical and biological interest. Controlling mechanisms constraining plasticity, such as PP activity and/or inflammation (18, 19, 21, 22), is vital to optimize therapeutic protocols and ensure their success in restoring function by promoting plasticity.

siRNA knockdown of specific molecular targets has become more common, and is being used clinically. siRNAs treatments show minimal toxicity in animal models and have been used to treat cancer, HIV, and neurodegenerative diseases (9, 25, 33). siRNAs have been used experimentally in rodents to demonstrate the involvement of PKCθ (10) and TrkB receptors (6) in respiratory motor plasticity and to investigate functional recovery after cervical spinal injury (30). While concerns have been raised that siRNAs can induce an immunologic response (24), they are also able to enhance plasticity, as demonstrated here. The ability to reveal plasticity shown here argues against significant immunological response from intrapleural siRNA injections since respiratory plasticity is exceptionally sensitive to even modest inflammation (18, 19, 21–23, 39).

One challenge in using siRNAs is verification of selective target knockdown. While we were unable to verify knockdown using immunofluorescence (perhaps due to limitations in antibody selectivity of PP subtype), we verified knockdown using cultured PC12 cells. Accell-modified siRNAs effectively target mRNA and protein in cortical neuron cultures (13, 36). We cannot rule out differential siRNA effects in vivo vs. in vitro, but we used siRNAs with intrapleural injections to target other mRNAs in phrenic motor neurons (6, 10) and successfully verified target knockdown in vivo. Furthermore, the differential physiological responses to siPP1 vs. siPP2A are consistent with selective and effective knock-down of the target molecule. Since the relative knockdown in vitro was similar for siPP1 and siPP2A, these differential physiological effects are unlikely due to differences in the relative potency of the selected siRNA pools.

Collectively, the opposing physiological effects of siPP1 and siPP2A in vivo and similar knock down in vitro support the hypothesis that PP1 is a relevant okadaic acid-sensitive Ser/Thr PP constraining facilitation after sustained hypoxia and may be critical in the pattern sensitivity of hypoxia-induced phrenic motor plasticity.

GRANTS

This study was supported by National Heart, Lung, and Blood Institute Grant HL-111598, Francis Family Foundation (to A. G. Huxtable), and American Physiological Society Undergraduate Fellowship (to T. J. Paterson).

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the authors.

AUTHOR CONTRIBUTIONS

A.G.H., J.J.W., and G.S.M. conceived and designed research; A.G.H., T.J.P., and J.N.O. performed experiments; A.G.H., T.J.P., J.N.O., and J.J.W. analyzed data; A.G.H., T.J.P., J.J.W., and G.S.M. interpreted results of experiments; A.G.H. and J.J.W. prepared figures; A.G.H. and T.J.P. drafted manuscript; A.G.H., T.J.P., J.J.W., and G.S.M. edited and revised manuscript; A.G.H., J.J.W., and G.S.M. approved final version of manuscript.