C‐type natriuretic peptide induces inotropic and lusitropic effects in human 3D‐engineered cardiac tissue: Implications for the regulation of cardiac function in humans

Abstract The role of C‐type natriuretic peptide (CNP) in the regulation of cardiac function in humans remains to be established as previous investigations have been confined to animal model systems. Here, we used well‐characterized engineered cardiac tissues (ECTs) generated from human stem cell‐derived cardiomyocytes and fibroblasts to study the acute effects of CNP on contractility. Application of CNP elicited a positive inotropic response as evidenced by increases in maximum twitch amplitude, maximum contraction slope and maximum calcium amplitude. This inotropic response was accompanied by a positive lusitropic response as demonstrated by reductions in time from peak contraction to 90% of relaxation and time from peak calcium transient to 90% of decay that paralleled increases in maximum contraction decay slope and maximum calcium decay slope. To establish translatability, CNP‐induced changes in contractility were also assessed in rat ex vivo (isolated heart) and in vivo models. Here, the effects on force kinetics observed in ECTs mirrored those observed in both the ex vivo and in vivo model systems, whereas the increase in maximal force generation with CNP application was only detected in ECTs. In conclusion, CNP induces a positive inotropic and lusitropic response in ECTs, thus supporting an important role for CNP in the regulation of human cardiac function. The high degree of translatability between ECTs, ex vivo and in vivo models further supports a regulatory role for CNP and expands the current understanding of the translational value of human ECTs.


INTRODUCTION
C-type natriuretic peptide (CNP) is released by endothelial cells, fibroblasts and cardiomyocytes in the heart where it elicits autocrine and paracrine signalling via its interaction with natriuretic peptide receptor (NPR) 2 and NPR3 (Cerra & Pellegrino, 2007;Moyes & Hobbs, 2019;Nyberg et al., 2023).NPR2 is a particulate guanylyl cyclase that catalyses the synthesis of cyclic guanosine monophosphate (cGMP) whereas NPR3 clears CNP via a receptor-mediated internalization and degradation process, but a signalling function (G i protein-linked receptor) has also been reported (Moyes & Hobbs, 2019).Following ligation of NPR2, the resulting increase in intracellular cGMP and subsequent increase in protein kinase G (PKG) activity leads to phosphorylation of phospholamban (regulator of sarco/endoplasmic reticulum Ca 2+ -ATPase 2 (SERCA2) controlling the influx of calcium to the SR), cardiac troponin I (regulatory protein that controls calciummediated interaction between actin and myosin) and titin (controlling cardiomyocyte stiffness) (Moyes & Hobbs, 2019;Sangaralingham et al., 2023).
Findings in murine in vitro, ex vivo and in vivo systems have revealed a direct positive lusitropic effect of CNP (Beaulieu et al., 1997;Manfra et al., 2022;Moltzau et al., 2013Moltzau et al., , 2014;;Qvigstad et al., 2010), whereas some controversy exist regarding its effect on force generation as both increases (Beaulieu et al., 1997;Hirose et al., 1998) and decreases (Manfra et al., 2022;Moltzau et al., 2013Moltzau et al., , 2014;;Qvigstad et al., 2010) in inotropy have been reported.Importantly, due to distinct species differences in cardiac structure and function between rodents and humans (Kusunose et al., 2012;Milani-Nejad & Janssen, 2014;Popovic et al., 2005), caution should be taken when translating mechanistic findings in rodent models into humans.Hence, since insight into the role of CNP for the regulation of cardiac function has been confined to observations from animal models, mechanistic studies in human systems are warranted to provide clarity on the human relevance of these findings.
Human 3D cardiac microphysiological systems, including engineered cardiac tissues (ECTs) using human induced pluripotent stem cell (hiPSC)-derived cardiomyocytes, are gaining significant interest given the increasing need for complex in vitro model systems to improve translatability and consequently also replace animal testing.By utilizing the Biowire II Platform, 3D ECTs can be generated from hiPSC-derived cardiomyocytes and cardiac fibroblasts.These ECTs display a phenotype that mimics adult human myocardium including a lack of spontaneous beating, the presence of a positive force-frequency response from 1 to 4 Hz and prominent post-rest potentiation.Furthermore, canonical responses to pharmacological agents that affect contractility in humans via a variety of mechanisms are also evident in this model system (Feric et al., 2019).Accordingly, this model is suitable for assessing human-relevant effects of CNP on contractility.
In the present study, we leveraged ECTs to study the acute effects of CNP on contractility in a well-characterized human 3D system.To provide insight into the translatability between animal findings and

Highlights
• What is the central question of this study?
What are the acute responses to C-type natriuretic peptide (CNP) in human-engineered cardiac tissues (ECTs) on cardiac function and how well do they translate to matched concentrations in animal ex vivo and in vivo models?
• What is the main finding and its importance?
Acute stimulation of ECTs with CNP induced positive lusitropic and inotropic effects on cardiac contractility, which closely reflected the changes observed in rat ex vivo and in vivo cardiac models.
These findings support an important role for CNP in the regulation of human cardiac function and highlight the translational value of ECTs.
those obtained in the present in vitro model system, CNP-induced changes in contractility were assessed in rat ex vivo and in vivo models in which circulating concentrations of CNP were matched to those applied in ECTs.

Ethics approval
The experiments were conducted in part in Denmark under a license from the Danish Ministry of Justice (License No.

Synthesis of CNP
CNP22 was synthesized by solid-phase peptide synthesis using Reversed-phase high-performance liquid chromatography (RP-HPLC) using a gradient of acetonitrile and water with 0.1% trifluoroacetid acid.The peptide was cyclized by the addition of 1.0 equivalent of 4-aldrithiol.The cyclized peptide was purified using RP-HPLC as described above and lyophilized to yield the peptide as a colourless powder.

Study protocol: in vitro experiments
3D ECTs were generated with the Biowire platform as previously described (Zhao et al., 2019).In short, ECTs were formed in polystyrene microwells containing parallel poly (octamethylene maleate (anhydride) citrate) (POMaC) wires.For each tissue, 100,000 iPSCderived cardiomyocytes (iCell® Cardiomyocytes 2 , Fujifilm Cellular Dynamics International, Madison, WI, USA) and 10,000 human ventricular cardiac fibroblasts (Lonza, Allendale, NJ, USA) in a collagen/Matrigel/fibrin gel were seeded in each well.After 7 days of culture, the cardiomyocytes and fibroblasts formed 3D ECTs around the polymer wires (Figure 1).Custom chambers containing parallel carbon electrodes were used to provide electrical field stimulation using biphasic pulses of 2 ms duration, at twice the excitation threshold.
Stimulation was started at 1 Hz and increased by 0.1 Hz increments daily to a maximum of 6 Hz.The ECTs used in this study were matured for 7 weeks in the Biowire II platform.
For calcium and contractility measurements, ECTs were stained with 5 μM Fura-4F-AM ( tracking the deflection of POMaC wires as previously described (Zhao et al., 2019).Briefly, the polymer wires were illuminated using 480 nm excitation, and videos were recorded at 100 frames per second at 515 nm emission.The pixel displacement of the polymer wires was converted to force using custom software to determine the amplitude and kinetics of contractility (Figure 1).CNP was added to the recording chamber at increasing doses (10 and 100 nM) with 15 min incubation per dose, after which contractility and calcium measurements were acquired from each tissue.At the end of each experiment, 100 nM isoproterenol (ISO, Cayman Chemical, Ann Arbor, MI, USA) was added to each tissue as a positive control for tissue responsiveness, and only tissues displaying a canonical response were included in the data.
The isolated effects of ISO and dobutamine at three doses were also tested in separate experiments, as cardiac dynamics following the application of these compounds are well described in humans and preclinical models.

Study protocol: in vivo experiments
Male Sprague-Dawley rats, 7-8 weeks old (n = 13; Vital River Laboratory Animal Technology Co. Ltd, Beijing, China), were randomly assigned to groups upon arrival.Animals were allowed an acclimatization period for at least 2 weeks before entering the study.The rats had unlimited access to tap water and standard chow and all animals were housed in an enriched environment with standard bedding and nesting material, under a 12/12-h day-night cycle (lights on at 06.00 h) in a humidity (30−70%) and temperature (20−25 • C)-controlled facility.
The surgical procedures were performed under anaesthesia with isoflurane (4-5% for induction and 2% for maintenance) and only after complete absence of reflexes.Animals were placed on controlled heating pads, and body temperature, measured via a rectal probe, was maintained at 37

Ex vivo and in vivo
The maximal slope of the systolic pressure increment and the diastolic pressure decrement was calculated and presented as dP/dt max and dP/dt min .To provide further insight into diastolic function, the left ventricular relaxation time constant, tau, was also calculated.Left ventricular developed pressure (LVDP) was obtained by subtracting left ventricular end-systolic pressure (LVESP) with left ventricular end diastolic pressure (LVEDP), and rate pressure product (RPP) was calculated by multiplying heart rate (HR) by systolic blood pressure.

Biowire sample prep for RNAseq
Each Biowire tissue was placed directly in 100 μl of Trizol (Thermo Fisher Scientific) and flash frozen.RNA extraction, library preparation, sequencing and analysis were conducted at Azenta Life Sciences (South Plainfield, NJ, USA) as follows.

RNA extraction and quality control
Total RNA was extracted from fresh frozen tissue samples using the Qiagen RNeasy Plus Universal mini kit following manufacturer's instructions (Qiagen, Hilden, Germany).RNA samples were quantified using Qubit 2.0 fluorometer (Thermo Fisher Scientific) and RNA integrity was checked using Agilent TapeStation 4200 (Agilent Technologies, Santa Clara, CA, USA).

Library preparation
RNA sequencing libraries were prepared using the NEBNext Ultra RNA Library Prep for Illumina using manufacturer's instructions (NEB, Ipswich, MA, USA).Briefly, mRNAs were initially enriched with Oligod(T) beads.Enriched mRNAs were fragmented for 15 min at 94 • C.
First-and second-strand cDNA were subsequently synthesized.cDNA fragments were end-repaired and adenylated at 3′ ends, and universal adapters were ligated to cDNA fragments, followed by index addition and library enrichment by PCR.Sequencing libraries were validated on the Agilent TapeStation, and quantified using Qubit 2.0 fluorometer and by quantitative PCR (KAPA Biosystems, Wilmington, MA, USA).

Illumina sequencing
Sequencing libraries were clustered on a lane of a HiSeq flowcell, and after clustering the flowcell was loaded on the Illumina instrument (4000 or equivalent) according to the manufacturer's instructions.
The samples were sequenced using a 2 × 150 bp paired-end (PE) configuration.Image analysis and base calling were conducted by the Control software.Raw sequence data (.bcl files) generated by the sequencer were converted into fastq files and demultiplexed using Illumina's bcl2fastq 2.17 software.One mismatch was allowed for index sequence identification.

RNAseq data analysis
Sequence reads were trimmed to remove adapter sequences, and the trimmed reads were mapped to the reference genome available on ENSEMBL using the STAR aligner v.2.5.2b.BAM files were generated as a result of this step.Unique gene hit counts were calculated by using feature Counts from the Subread package v.1.5.2.Only unique reads that fell within exon regions were counted.After extraction of gene hit counts, the gene hit counts table was used for downstream analyses including a filter for calling likely expression.The expression filter was set to 10 cpm across replicate samples and genes were called as expressed if they were detected at ≥10 cpm in all replicates.

RNA sequencing confirmed NPR2 expression and tissue maturity
RNA sequencing was performed on ECTs after maturation to confirm tissue maturity and CNP binding receptor expression (NPR2, NPR3).

3.3
In vitro effects of CNP application in 3D-engineered cardiac tissues After 7 weeks of maturation, the ECTs reached an average force of ∼13 μN, and baseline parameters of contractility and calcium trans-ients were similar between groups, except for calcium ratio (340/380) time to decay at 90% (s), which was higher in the CNP group compared to the vehicle group (0.36 vs. 0.34 s, respectively, P < 0.0206; Table 1).
Application of CNP (10 and 100 nM) elicited a positive inotropic response (dose × group, P < 0.0001, Figures 4a,b and 5a).Specifically, CNP treatment increased maximum twitch amplitude (μN) as evidenced by a significant dose × group difference (P < 0.0001, Figure 4a).For individual comparisons, 10 nM CNP (P = 0.0549) did TA B L E 1 In vitro contractility and calcium parameters at baseline.was significantly increased compared to vehicle (Figure 4a).Similarly, CNP treatment induced an increase in the maximum contraction slope (μN/s) also evidenced by a significant dose × group difference (P < 0.0001, Figure 4b); however, numerical differences did not reach significance when performing individual comparisons between groups:
Acute CNP treatment also induced a positive lusitropic response (dose × group, P < 0.0001, Figures 4c,d and 5b,c).This was evidenced by reductions in time from peak contraction to 90% of relaxation (s) at both 10 nM CNP (P = 0.0003) and 100 nM CNP (P = 0.0003) (Figure 4c) and an increase in the maximum relaxation slope (μN/s) at both 10 nM CNP (P = 0.0427) and 100 nM CNP (P = 0.0427) (Figure 4d).These effects were paralleled by reductions in time from maximum calcium amplitude to 90% of decay (s) at both 10 nM CNP (P = 0.0106) and 100 nM CNP (P = 0.0013) and an increase in the maximum calcium decay slope (ratio/s) at 10 nM CNP (P < 0.0001) and 100 nM CNP (P < 0.0001) (Figure 5b,c).Higher concentrations of CNP (300 nM and 1 μM) were also tested acutely in ECTs.Here, no differences were observed at baseline in ECTs between vehicle and CNP groups.Higher concentrations of CNP resulted in a similar inotropic and lusitropic response (dose × group, P < 0.0001) as those observed with the lower concentrations of CNP.
Coronary perfusion was unaltered before and after ischaemia between groups (Figure 6g,h).

in vivo effects of CNP infusion on rat cardiac function
There were no differences in indices of cardiac function and central hemodynamics at baseline (t = 0, Table 2: individual data for all measurements included in Supplementary File 2).Intravenous constant infusion of CNP improved cardiac contractility as evidenced by an increase (time × group, P < 0.0001) in dP/dt max (Figure 7c).
Similarly, diastolic function was enhanced as demonstrated by an increase (time × group, P = 0.0005) in dP/dt min (Figure 7f) and lower (time × group, P < 0.0001) diastolic time constant Tau and LVEDP (compared to vehicle, Figure 7d,e).These cardiac effects of CNP infusion were observed in the absence of alterations in systolic blood pressure, diastolic blood pressure, mean arterial blood pressure, HR, LVESP, LVDP and RPP (Figure 7a,b,g-k).

DISCUSSION
By utilizing a human ECT model, we demonstrate for the first time that CNP induces an inotropic and lusitropic response in a 3D in vitro system composed of matured human iPSC-derived cardiomyocytes.The effects on kinetics mirrored those observed in rat ex vivo and in vivo model systems, whereas an increase in maximal force generation with CNP application was only observed in ECTs.
In the human heart, CNP is both produced and released to the circulation as evidenced by positive arteriovenous gradients for CNP and N-terminal pro C-type natriuretic peptide (NT-proCNP) across the working heart (Palmer et al., 2009).The cellular source of CNP is likely to include endothelial cells, cardiac fibroblasts and cardiomyocytes as CNP is expressed in these cell types in the human heart (Koenig et al., 2022).Once released from the cell, CNP may induce paracrine and autocrine signalling via NPR2 expressed by cardiomyocytes (Koenig et al., 2022;Subramanian et al., 2018) where the present findings in ECTs suggest that it will have a modulatory effect on cardiac contractility.This proposed translatability of acute effects of CNP from ECTs into humans is supported by findings of canonical responses to a variety of small molecule modulators of cardiac contractility and calcium kinetics at pharmacologically relevant concentrations in similar ECTs (Feric et al., 2019).
To further establish the physiological relevance of the observed here was the lack of significance with regards to change in dP/dt max in the ex vivo model; however, this is likely to reflect insufficient power to detect changes in this specific variable given the effect size of ∼20% and very low P-value of 0.063.This translatability between model systems, which to some extent may reflect a very high conservation within the CNP system (Potter et al., 2009), indicates that other findings on the cardiac effects of CNP signalling in preclinical models may also translate very well.
We did observe one discrepancy between the models wherein an increase in developed force (twitch amplitude) was only detected in ECTs while LVDP remains unaltered in the intact heart (excluding the effect observed following ischaemia).This lack of increase in inotropy could, however, reflect that the ex vivo and in vivo setting provides an integrative response to CNP stimulation in which afterload (arterial blood pressure) is an important physiological determinant of LVDP.
Hence, as blood pressure was not altered in these models, it may be speculated that a CNP-induced increase in developed pressure may be evident in situations where arterial blood pressure is increased.In support of this proposition, a positive correlation between coronary sinus plasma CNP concentrations and mean pulmonary artery pressure has been reported in humans (Palmer et al., 2009).
Regarding the finding of unaltered blood pressure dynamics in the in vivo model, it should be noted that blood pressure is the product of cardiac output and systemic vascular resistance.Hence, a potential augmentation of blood pressure as a result of an increase in cardiac output may have been counterbalanced by a CNP-induced reduction in systemic vascular resistance (peripheral vasodilatation).However, the finding that LVDP and a proxy for cardiac work (RPP) did not change with treatment supports that CNP is not acting as an inotropic agent that increases cardiac output.Hence, the unaltered blood pressure is likely to reflect that CNP alters cardiac kinetics without altering stroke volume and systemic resistance in these healthy animals.
Early studies using isolated right atria preparations from healthy dogs have reported positive inotropic (Beaulieu et al., 1997;Hirose et al., 1998) and lusitropic (Beaulieu et al., 1997) responses to CNP application.However, more recent studies in rodent HF models examining the effects of CNP on isolated cardiac muscle strips have demonstrated negative inotropic and positive lusitropic effects (Manfra et al., 2022;Moltzau et al., 2013Moltzau et al., , 2014;;Qvigstad et al., 2010).
Yet others have reported a positive acute inotropic response followed by a negative inotropic response in wild-type (WT) isolated mouse heart preparations (Pierkes et al., 2002;Wollert et al., 2003).Taken together, these observations could indicate distinct effects of CNP depending on the health status of the tissue (Table 3).The experiments in the present study were performed in healthy tissue preparations, and the findings are thus in line with previous findings on CNP in healthy or WT tissue preparations, although we were not able to detect a potential decline in developed force given the single sampling point.Importantly, future studies should assess whether CNP induces negative inotropic responses in ECTs displaying a disease-relevant phenotype.
In isolated cardiomyocytes, findings have been more varied when it comes to inotropic responses.In line with ex vivo findings, negative inotropic and positive lusitropic responses in isolated rat cardiomyocytes have been reported (Moltzau et al., 2013).In contrast, Wollert et al. (2003) showed increased cell shortening of isolated WT cardiomyocytes treated with 300 nM CNP along with decreased time to 90% relaxation.In another study, positive inotropic effects at 10 and 100 nM of CNP treatment in isolated mouse WT cardiomyocytes were reported (Szaroszyk et al., 2022).These results are in line with the observations in the present study, demonstrating positive inotropic and lusitropic effects at 10 and 100 nM of CNP in ECTs.
In a recent study, CNP was shown to have antiarrhythmic effects in isolated murine cardiomyocytes, Langendorff-perfused hearts and human iPSC-derived cardiomyocytes (Cachorro et al., 2023).Although CNP promoted a lusitropic response, no changes in inotropy were observed at 1 μM CNP in their preparations of isolated murine cardiomyocytes.This study also examined the effects of CNP on

TA B L E 3
Overview of in vitro, ex vivo and in vivo findings of CNP effects on cardiac function.suggesting both positive inotropic and lusitropic effects of CNP (Soeki et al., 2005).In a mouse model of cardiac hypertrophy and remodelling induced by angiotensin II, CNP infusion also increased fractional shortening and decreased LVED dimension compared to the angiotensin II vehicle group, but no changes were observed with CNP when comparing with vehicle in the saline groups (Izumiya et al., 2012).In pacing-induced HF in dogs, CNP augmented LV contraction, relaxation, diastolic filling and LV arterial coupling (Li et al., 2016).

Study
Taken collectively, observations in vivo suggest that CNP elicits both inotropic and lusitropic responses in preclinical models of HF.
One very consistent finding across model systems has been a CNP-induced enhancement of diastolic function in both health and cardiac disease.When combined with the present observation of a similar lusitropic response to CNP application in ECTs and the findings of reduced CNP mRNA levels in the failing heart (Ichiki et al., 2014;Moyes et al., 2020), it may be speculated that a pharmacological treatment with a long-acting CNP analogue could improve cardiac performance in patients suffering from impaired LV filling.Here, HF with preserved ejection fraction (HFpEF) would be one patient group in which such a pharmacotherapy could be effective, as one hallmark of HFpEF is impaired relaxation and/or increased viscoelastic chamber stiffness (Gaasch & Zile, 2004), leading to elevated filling pressures (Borlaug et al., 2016) that promote symptoms of dyspnoea, impair exercise capacity (Obokata et al., 2018), increase risk for HF hospitalization (Adamson et al., 2014), and decrease survival in HFpEF (Dorfs et al., 2014).
Some experimental considerations should be highlighted.Applied CNP concentrations ex vivo and in vivo were in the low nM range whereas CNP concentrations in human plasma have been reported to be in the low pM range (Palmer et al., 2009).Here, it should be emphasized that local concentrations of CNP in the contracting myocardium are expected to be significantly higher than what is observed in plasma since CNP is produced by cardiac tissue-resident cells (Koenig et al., 2022), and any CNP that reaches the systemic circulation is diluted by the large plasma volume where it is also rapidly cleared (half-life ∼2 min) by enzymatic degradation and NPR3mediated clearance (Potter et al., 2009).Interestingly, in ECTs, similar responses were observed with 10 and 100 nM CNP, which could indicate that a maximum effect was achieved at 10 nM.Thus, future studies should utilize lower concentrations to characterize a potential dose-response relationship in ECTs.Furthermore, to confirm the presence and functionality of pathways that modulate cardiac contractility in humans, ECTs were stimulated with isoproterenol and dobutamine and here canonical responses were observed.These confirmatory experiments were only performed in ECTs, as the cardiac responses to these pharmacological agents have been well-described for both perfused heart preparations and in vivo settings (e.g., Dobson et al., 1990;Osadchii et al., 2007;Romano et al., 1991).
In conclusion, the present study demonstrates that CNP induces standard fluorenylmethoxycarbonyl (Fmoc)-based amino acids and reagents.Amino acid couplings were performed using four equivalents of Fmoc-protected amino acid, N,N′-diisopropylcarbodiimide (DIC) and Oxyma Pure, respectively, in dimethylformamide (DMF) for 90 min.After coupling, the resin was capped by treatment with 1.2 M acetic anhydride and 0.6 M N,N-diisopropylethylamine (DIPEA) in DMF for 30 min.Fmoc groups were removed by treatment with 20% F I G U R E 1 Illustration of the engineered cardiac tissue model and measured variables.(a) A representative image of an ECT in the Biowire platform attached to the polymer wires.Twitch force is measured by recording a video of the contracting tissues under field stimulation and converting pixel displacement of the polymer wires into force measurements.(b) Measured parameters for contractility and calcium transients.CM, cardiomyocyte; ECT, engineered cardiac tissue.piperidine in DMF for 2 × 15 min.The peptide was cleaved from the resin by treatment with trifluoroacetic acid, triisopropylsilane, 1,4dithiothreitol and water (90:5:2.5:2.5) in addition to ammonium iodide at 15 mg/ml cleavage mixture.The crude peptide was purified by • C.An incision was then made in the left inguinal area along the natural angle of the hind leg, after which blunt dissection of the tissue was performed to expose the femoral artery.A pressure catheter (Scisense 1.2F solid-state pressure catheter, Transconic, Ithaca, NY, USA) was inserted into the left femoral artery using a 25-30-G needle to allow for arterial measurement of blood pressure.Then, an incision over the right carotid artery from mandible to sternum was performed to expose the carotid artery.The vagus nerve was gently dissected away and sterile 6-0 sutures were placed around the proximal and distal parts of the carotid artery after which a pressure catheter (Scisense 1.2F solid-state pressure catheter, Transconic) was inserted using a 25-30-G needle.The catheter was gently advanced into the left ventricle and then secured by the tightening of the sutures.After stabilization for 10-15 min, haemodynamic parameters were acquired (SP200 Pressure System, Transconic) and recorded (Biopac MP160, Biopac systems, Inc., Goleta, CA, USA) at baseline (−15 to 0 min) and during constant infusion of CNP (200 pmol/min/kg) via the tail vein for 30 min.The infusion rate of CNP was based on calculations (one compartment model) of the dose needed to obtain a concentration of ∼10 nmol/l at steady state.

F
I G U R E 2 Isoproterenol and dobutamine treatment induce positive inotropic and lusitropic contractility responses in 3D-engineered cardiac tissues.All data are expressed as a percentage of baseline.Values are means ± SD and represent n = 6, except (c) where n = 3. Significant difference between baseline and dobutamine/isoproterenol: ***P < 0.001, ****P < 0.0001.F I G U R E 3 Isoproterenol and dobutamine treatment induce positive inotropic and lusitropic calcium responses in 3D-engineered cardiac tissues.All data are expressed as a percentage of baseline.Values are means ± SD and represent n = 6.Significant difference between baseline and CNP: ****P < 0.0001.CNP, C-type natriuretic peptide.maximum decay slope was increased with both agonists (P < 0.0001), apart from 5 nM dobutamine (P = 0.2169) (Figure 3e,f).

Note:
Significant difference between vehicle and CNP at baseline: *P < 0.05.F I G U R E 4 CNP treatment induces positive inotropic and lusitropic contractility responses in 3D-engineered cardiac tissues.All data are expressed as a percentage of their own baseline.Values are means ± SD and represent n = 8 (vehicle) and n = 6 (CNP).Significant difference between vehicle and CNP: *P < 0.05, ***P < 0.001.CNP, C-type natriuretic peptide; ISO, isoproterenol.not reach statistical significance, whereas 100 nM CNP (P = 0.0489)
the first part of the Langendorff experiments, a suitable IR injury was established by testing IR periods of 15, 17.5, 20, 25 and 30 min.Ischaemic times of 20, 25 and 30 min resulted in large ~80% reductions in LVDP.Ischaemic time of 15 min resulted in no IR injury, whereas 17.5 min resulted in a ~45% reduction in LVDP.Prior to ischaemia in the Langendorff isolated heart model, no effect on LVDP was observed with CNP treatment (Figure 6a: individual data for all measurements

F
I G U R E 6 CNP treatment increases systolic and diastolic parameters in isolated hearts.Values are means ± SD and represent n = 14 (vehicle) and n = 15 (CNP).CNP was dissolved in Krebs-Henseleit buffer to a final concentration of 30 nM.Coronary perfusion was before and after 17.5 min of global ischaemia.Significant difference between vehicle and CNP: *P < 0.05, **P < 0.01, ***P < 0.001.CNP, C-type natriuretic peptide; dP/dt max and dP/dt min , maximal slope of the systolic pressure increment, and the diastolic pressure decrement; LVDP, left ventricular developed pressure.

F
I G U R E 7 CNP infusion increases indices of cardiac performance in vivo without altering systemic hemodynamics.All data are expressed as a percentage of baseline (t = 0).Values are means ± SD and represent n = 7 (vehicle) and n = 6 (CNP).CNP was infused at a constant rate of 200 pmol/min/kg (plasma concentration of ∼10 nmol/l at steady state) via the tail vein.Significant difference between vehicle and CNP: ***P < 0.001, ****P < 0.0001.CNP, C-type natriuretic peptide; DBP, diastolic blood pressure; dP/dt max and dP/dt min , maximal slope of the systolic pressure increment, and the diastolic pressure decrement; HR, heart rate; LVDP, left ventricular developed pressure; LVEDP, left ventricular end-diastolic pressure; LVESP, left ventricular end-systolic pressure; MABP, mean arterial blood pressure; RPP, rate-pressure product; SBP, systolic blood pressure.cardiac parameters using human cells in vitro; however, contractility parameters were not assessed for the 2D hiPSC-derived cardiomyocytes.Furthermore, calcium sparks detected in hiPSC-derived cardiomyocytes were performed in a traditional 2D setting, which is associated with less mature cardiomyocytes compared to those in 3D ECTs.To our knowledge, no studies have examined the effects of CNP in isolated human cardiomyocytes.However, the use of isolated human cardiomyocytes faces major challenges related to the limited availability of human donor hearts, the unstable isolation efficiency and quality, and rapid cell death.In addition, single-cell cardiomyocytes lack important mechanical properties such as stretch and loading present in cardiac tissues.In contrast, ECTs use human pluripotent stem-cell derived cardiomyocytes that are readily available and can be cultured for long periods.Furthermore, the ECTs used in current study, exhibited high expression of the CNP binding receptors along with several markers of tissue maturity, although further investigation is required to determine protein levels.Future studies are needed to compare human-isolated cardiomyocytes and ECTs to enhance our understanding of the translatability between these model systems.Effects of CNP in vivo have been studied in several animal models of HF.Infusion of CNP in myocardial infarction (MI)-induced rats increased LV dP/dt max and fractional shortening, whereas LVEDP and LV dP/dt min were decreased compared to the MI-vehicle group,

a
positive inotropic and lusitropic response in human ECTs, thus supporting an important role for CNP in the regulation of human cardiac function.The finding that the in vitro effects of ECTs were highly reflective of CNP-induced changes in rat cardiac dynamics ex vivo and in vivo provides additional support to this hypothesis and expands our current understanding of the translational value of ECTs.Future studies investigating the cardiac effects of manipulating CNP signalling in humans are warranted to substantiate the translational significance of the present findings in ECTs.AUTHOR CONTRIBUTIONS Julian C. Bachmann, Jeppe E. Kirchhoff, Giulia Borghetti and Michael Nyberg conceived and designed the research.Jeppe E. Kirchhoff, Julia E. Napolitano, Steve Sorota, William M. Gordon, Nicole Feric, Roozbeh Aschar-Sobbi, Juan Lv and Zhiyou Cao performed experiments and analysed data.Julian C. Bachmann, Jeppe E. Kirchhoff, Julia E. Napolitano, Steve Sorota, William M. Gordon, Nicole Feric, Roozbeh Aschar-Sobbi, Giulia Borghetti and Michael Nyberg interpreted the results of experiments.Julian C. Bachmann and Michael Nyberg drafted the manuscript.Julian C. Bachmann, Jeppe E. Kirchhoff, Julia E. Napolitano, Steve Sorota, William M. Gordon, Nicole Feric, Roozbeh Aschar-Sobbi, Ken Coppieters, Giulia Borghetti and Michael Nyberg edited and revised the manuscript.All authors have read and approved the final version of this manuscript and agree to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved.All persons designated as authors qualify for authorship, and all those who qualify for authorship are listed.
Analyses were performed separately for the period before and after ischaemia in the ex vivo experiments.All analysis were conducted in GraphPad Prism software (version 9.0.1,GraphPad Software, San Diego, CA, USA).Statistical significance was set at P < 0.05.Values are expressed as means ± SD.
A Mann-Whitney test was used to detect differences in baseline values.For the in vitro, ex vivo and in vivo experiments, a two-way ANOVA followed by Šídák's multiple comparison test was used, except for Figures2 and 3, where a one-way ANOVA with Dunnett's multiple comparisons test was used to test for differences from baseline.
In vivo haemodynamic parameters at baseline (time = 0).DBP, diastolic blood pressure; dP/dt max and dP/dt min , maximal slope of the systolic pressure increment, and the diastolic pressure decrement; HR, heart rate; LVDP, left ventricular developed pressure; LVEDP, left ventricular end-diastolic pressure; LVESP, left ventricular end-systolic pressure; MABP, mean arterial blood pressure; RPP, rate pressure product; SBP, systolic blood pressure.
effects of CNP in ECTs, CNP-induced changes in contractility were assessed in rat ex vivo and in vivo models in which circulating concentrations of CNP were matched to those applied in ECTs.In both the isolated heart preparation and in the intact animals, increases in dP/dt max and dP/dt min (as well as lower tau and LVEDP in vivo) were detected following CNP stimulation, thus mirroring the effects on contractility and calcium kinetics observed in ECTs.The only exception TA B L E 2 AngII, angiotensin II; AP, action potential; CM, cardiomyocyte; E/A, atrial filling wave; E, early filling wave; ECT, engineered cardiac tissue; FS, fractional shortening; GC-A, guanylyl cyclase-A; LV, left ventricular; LVEDD, left ventricular end-diastolic; LVEDP, left ventricular end-diastolic pressure; MI, myocardial infarction; PKG, protein kinase G; PWT, posterior wall thickness; SAN, sinoatrial node; TG, transgenic; TR80, time to 80% relaxation; WT, wildtype.