Pharmacological inhibition of focal segmental glomerulosclerosis-related, gain of function mutants of TRPC6 channels by semi-synthetic derivatives of larixol

Pharmacological inhibition of FSGS-related TRPC6 gain of

function mutants by semisynthetic larixol-derived

compounds

,1

2

Nicole Urban* , Sonja Neuser* , Anika Hentschel , Sebastian Köhling , Jörg Rademann ,

1

Michael Schaefer

1

:

Rudolf-Boehm-Institut für Pharmakologie und Toxikologie, Universität Leipzig, Leipzig,

Germany

2

:

Institut für Pharmazie, Freie Universität Berlin, Berlin, Germany

*: equal contribution

Running title: Inhibitors of FSGS-related TRPC6 channel mutants

Corresponding author: Michael Schaefer, Rudolf-Boehm-Institut für Pharmakologie und

Toxikologie, Härtelstr. 16-18, 04107 Leipzig, Germany, phone: +49 341 9724600, fax: +49

341 9724609, e-mail: michael.schaefer@medizin.uni-leipzig.de

Word count:

-

Abstract: 247

Text body without methods, figure legends and references: 5307

Discussion: 1946

References: 50

Figures: 11

Tables: 2

Supporting materials: 3 figures, 1 table

Abbreviations:

2+

[

Ca ], intracellular Ca concentration; FSGS, focal segmental glomerulosclerosis; TRPC,

i

transient receptor potential canonical; HBS, HEPES-buffered solution; HEK, human

embryonic kidney; HEK_h_T_R_P_C₆_-_Y_F_P_,HEK293 cells stably expressing human TRPC6 C-

terminally fused to yellow fluorescent protein

This article has been accepted for publication and undergone full peer review but has not

been through the copyediting, typesetting, pagination and proofreading process which may

lead to differences between this version and the Version of Record. Please cite this article as

doi: 10.1111/bph.13977

Abstract

Background and purpose

Gain of function mutations in TRPC6 can cause autosomal dominant forms of focal

segmental glomerulosclerosis (FSGS). Validated inhibitors of TRPC6 that are biologically

active on FSGS-related TRPC6 mutants are eagerly sought.

Experimental approach

We synthesized new TRPC6-inhibiting modulators from larixol, a resiniferous constituent of

Larix decidua, and tested the potency and selectivity in cell lines stably expressing various

2+

TRP channel isoforms. Channel activation was followed by Ca influx analyses and

electrophysiological recordings. The most promising compound larixyl carbamate (LC) was

tested on native TRPC6 and TRPC6 constructs carrying FSGS-related point mutations.

Key results

LC exhibited an about 30-fold and 5-fold preference for TRPC6 versus TRPC3 and TRPC6

versus TRPC7, respectively. Six FSGS-related TRPC6 mutants, including the highly active

M132T and R175Q variants, were strongly inhibited by 1 µM larixyl carbamate. We next

2+

sought to identify TRPC6-like Ca signals in primary murine podocytes, or in acutely

2+

isolated glomeruli. Surprisingly, no TRPC6-related Ca signals were detectable in these

preparations. Quantitative PCR revealed a 20- to 50-fold lower abundance of TRPC6

transcripts in rat or mouse podocytes compared to pulmonary artery smooth muscle cells

from the same species. Accordingly, electrophysiological recordings demonstrated that

diacylglycerol-induced currents in murine podocytes are very small, but sensitive to LC.

Conclusion and implications

We conclude that, despite low abundance in native podocytes, the native TRPC6 channel is

targetable using larixol-derived TRPC6 inhibitors. Like wild-type TRPC6, FSGS-related

TRPC6 mutants are sensitive to the newly developed inhibitors, thus, paving the way for

experimental therapies.

List of hyperlinks used in MS# 2016-BJP-0904-RP.R2

Targets:

TRPC3: http://www.guidetopharmacology.org/GRAC/ObjectDisplayForward?objectId=488

TRPC6: http://www.guidetopharmacology.org/GRAC/ObjectDisplayForward?objectId=491

TRPC7: http://www.guidetopharmacology.org/GRAC/ObjectDisplayForward?objectId=492

Ligands:

angiotensin II: http://www.guidetopharmacology.org/GRAC/LigandDisplayForward?ligandId=2504

bradykinin: http://www.guidetopharmacology.org/GRAC/LigandDisplayForward?ligandId=649

carbachol: http://www.guidetopharmacology.org/GRAC/LigandDisplayForward?ligandId=298

flufenamic acid: http://www.guidetopharmacology.org/GRAC/LigandDisplayForward?ligandId=4191

SKF-96365: http://www.guidetopharmacology.org/GRAC/LigandDisplayForward?ligandId=2441

thapsigargin: http://www.guidetopharmacology.org/GRAC/LigandDisplayForward?ligandId=5351

Introduction

Focal segmental glomerulosclerosis (FSGS) is one of the leading causes of steroid-resistant

nephrotic syndrome and terminal kidney failure in children and adolescents. FSGS is a

histopathological finding which in most cases correlates with podocyte injury. In FSGS, the

slit diaphragm formed by intercalating foot processes of podocytes is severely disturbed. As

a consequence, the filtration barrier cannot retain proteins, and a progressive proteinuria

develops, finally culminating in end-stage renal disease and nephrotic syndrome. Inheritable

forms of FSGS can be caused by missense or nonsense mutations in proteins that are

enriched in podocyte foot processes (e.g. nephrin, podocin, or CD2-associated protein) or in

proteins that impact on podocyte function in more complex and indirect ways (Sprangers et

al., 2016). Of them, the identification of missense mutations in the TRPC6 gene in autosomal

dominant FSGS establishes TRPC6 as a potential pharmacological target (Winn et al., 2005;

El & Reiser, 2011). Both gain of function and loss of function mutations in TRPC6 have been

identified in autosomal dominant forms of FSGS (Riehle et al., 2016). Of note, the onset of

disease seems to correlate with functional properties of the mutated TRPC6 (Gigante et al.,

2

011). A M132T mutant, for example, causes an early onset of FSGS in the childhood,

which correlates with a massively increased channel activity of this mutant (Heeringa et al.,

009).

2

2+

TRPC6, a member of the canonical TRP channels TRPC1-TRPC7, is a Ca -permeable, but

poorly selective cation channel. It is regulated in a phospholipase C-dependent manner,

including a direct activation by diacylglycerols {Hofmann, 1999 8 /id}{Dietrich, 2014 14 /id}.

Like TRPV1, a functional TRPC6 channel complex presumably consists of four subunits. In

cells that primarily express TRPC6, four subunits assemble into a functionally active pore

complex. If co-expressed with TRPC3 or TRPC7, heteromeric TRPC6-bearing channels

complexes may form and mediate receptor-stimulated cation currents (Hofmann et al.,

2002). In TRPC6-linked FSGS, the autosomal dominant inheritance of the disease implies

that mutated TRPC6 subunits may either form separate channel complexes, or they may

oligomerise together with the gene product of the wild-type TRPC6 allele to form channel

complexes that contain 1-3 mutated TRPC6 subunits.

TRPC6 is expressed in a variety of tissues and cell types, including vascular smooth muscle

cells, mesangial cells and podocytes. Although the presence of TRPC6 in podocytes has

been demonstrated by several groups, there is considerable variability concerning the

abundance of the channel protein and the size of functional responses to TRPC6 activators

(

Kalwa et al., 2015; Anderson et al., 2014; Abkhezr et al., 2015). Since dysregulation of

TRPC6 expression in podocytes can occur in response to various stimuli, TRPC6 may also

play a seminal role in secondary forms of FSGS. Conditions that cause TRPC6 upregulation

or positively modulate its activation include puromycin treatment, hyperglycaemia, long-term

administration of angiotensin II, TNF- signaling, reperfusion injury or exposure to oxidative

stress (Ilatovskaya & Staruschenko, 2015; Kim et al., 2012; Abkhezr et al., 2015; Wang et

al., 2009; Szabo et al., 2015).

We have recently identified larixyl acetate as a potent and TRPC6-prevalent inhibitor (Urban

et al., 2016). We now investigate whether biological activity can be demonstrated on native

TRPC6 channel complexes in podocytes and on TRPC6 variants carrying mutations that are

linked to FSGS. In addition, to further improve the properties of the compound, we sought to

replace the instable acetyl ester by a carbamate moiety. The resulting semisynthetic

compound larixyl carbamate was equally potent as larixyl acetate, but its selectivity for

TRPC6 versus TRPC3 was slightly enhanced. In primary podocyte cultures or in freshly

2+

isolated glomeruli, fluorometric Ca assays failed to demonstrate functional expression of

TRPC6. Applying quantitative RT-PCR and Western blot analysis, a very low abundance of

TRPC6 expression was found in podocytes compared to pulmonary artery smooth muscle

cells. Finally, electrophysiological recordings were applied to demonstrate TRPC6 currents

in podocytes and their susceptibility to pharmacological inhibition by larixol derivatives.

Materials and Methods

Isolation and derivatisation of larixol

Isolation of larixol: larch turpentine (25 g) from Larix decidua (Kremer Pigmente, Aichstetten,

Germany) was dissolved in 100 ml n-hexane, and 50 ml of 1 M KOH were added. The

formed precipitate was filtrated off over Celite. The organic phase was washed with 1 M

NaOH, dried over magnesium sulfate, and n-hexane was removed under reduced pressure.

The yellow, oily residue was dissolved in 1 M KOH in EtOH (300 ml) and refluxed for 3 h.

The solution was evaporated to half of its volume, 100 ml of water were added, and the

solution was evaporated to an oily residue. To the latter, diethyl ether (100 ml) was added.

The aqueous phase was separated and extracted with diethyl ether. The combined organic

phases were dried over magnesium sulfate and, after filtration, solvents were removed

yielding after crystallization from cyclohexane (30 ml) 2.99 g of larixol.

Larixol derivatives were prepared according to the following protocols. All compounds were

purified by chromatography on silica to >95%. Purity was determined by TLC, NMR

spectroscopy and for the larixyl-6-carbamate, -formate and methyl ether by elemental

analysis. NMR data were assigned by using 2D experiments, namely H,H-COSY, HSQC,

and HMBC experiments.

Larixyl-diacetate: Larixol (50 mg, 0.16 mmol, 1.0 equiv.) was dissolved under nitrogen in

dimethylaniline (1.5 ml), and acetyl chloride (50 µl, 0.65 mmol, 4.0 equiv.) was added. The

solution was stirred at room temperature, and further acetyl chloride (25 µl) was added after

2

h and 4 h. After 16 h, further dimethylaniline (1 ml) and acetyl chloride (25 µl) were added,

and the mixture was heated to 90°C for 3 h. The reaction was quenched by addition of

water, extracted with dichloromethane, washed with saturated sodium bicarbonate solution,

dried over magnesium sulfate, and the solvent was removed at reduced pressure. The

obtained remainder was purified by chromatography on silica (n-hexane: ethyl acetate 95:5

→

90:10 (v:v)), yielding 29.7 mg (48%) of the diacetate product. R = 0.64 (n-hexane: ethyl

f

1

acetate 4:1 (v:v)). H-NMR (300 MHz, CDCl ) δ [ppm] = 0.74 (s, 3H, 20-CH ), 0.87 (s, 3H,

3

1

7

8/19-CH ), 1.01 (s, 3H, 18/19-CH ), 1.01-1.56 (m, 12H), 1.52 (s, 3H, 16-CH ), 1.41 (d, J =

3

2 Hz, 1H, 5-CH), 2.01 (s, 3H, Ac-CH ), 2.03 (s, 3H, Ac-CH ) 2.68 (dd, J = 12 Hz, 6 Hz, 1H,

3

-CH), 4.62 (m, 1H, CH ), 4.92 (m, 1H, CH ), 5.03 (dt, J = 12 Hz, 6 Hz, 1H, 6-CH) 5.10

2

13

(

s,1H), 5.14 (m, 1H), 5.96 (dd, J = 18 Hz, 12 Hz, 1H, 14-CH); C-NMR (75 MHz, CDCl3) δ

[

ppm] = 16.2 (20-CH ), 17.9 (11-CH ), 19.3 (2-CH ), 22.2 (Ac-CH ), 22.4 (Ac-CH ), 22.7

3

2

3

(

18/19-CH ), 23.8 (16-CH ), 33.7 (4-C), 36.4 (18/19-CH ), 39.2 (1-CH ), 39.4 (10-CH), 40.1

3 3 3 2

(

12-CH ), 43.7 (3-CH ), 44.4 (7-CH ), 56.5 (9-CH), 57.8 (5-CH), 73.5 (6-CH), 83.5 (13-C),

2 2 2

1

09.7 (17-CH ), 113.4 (15-CH ), 142.0 (14-CH), 145.4 (8-C), 170.2 (Ac-CO), 170.3 (Ac-CO).

2 2

+

HRMS [M+Na ] 413.26678, found: 413.26623 m/z.

Larixyl-6-propionate: Larixol (50 mg, 0.16 mmol, 1 equiv.) was dissolved under nitrogen in

dichloromethane (3 ml), and the solution was cooled to 0°C. After addition of propionyl

chloride (20 µl, 0.24 mmol, 1.5 equiv.) and diisopropylethyl amine (100 µl), the reaction was

allowed to proceed at room temperature under stirring for 16 h. The reaction was quenched

by addition of water, and the mixture was extracted with diethyl ether. Combined organic

phases were washed with 1 M HCl, saturated bicarbonate solution and brine, and dried over

magnesium sulfate. After filtration, solvents were removed and the product was purified by

silica chromatography (n-hexane: ethyl acetate 9:1 (v:v)), yielding 23.8 mg (41%) of the

1

larixyl-6-propionate. R = 0.46 (hexane: EE 9:1 (v:v)). H-NMR (300 MHz, CDCl ) δ [ppm] =

f

3

0

.74 (s, 3H, 20-CH ), 0.86 (s, 3H, 18/19-CH ), 1.00 (s, 3H, 18/19-CH ), 1.03-2.00 (m), 1.13

3 3 3

(

t, J = 6 Hz, 3H, propionyl-CH ) 1.27 (s, 3H, 16-CH ), 1.40 (d, J = 12 Hz, 1H, 5-CH), 2.01 (m,

3 3

1

H, 7-CH ), 2.28 (dq, J = 6 Hz, 3 Hz, 2H, propionyl-CH ) 2.66 (dd, J = 12 Hz, 6 Hz, 1H, 7-

2

CH), 4.63 (s, 1H, 17-CH ), 4.77 (s, 1H, 17-CH ), 4.97-5.06 (m, 2H, 15-CH , 6-CH), 5.20 (dd,

2

13

J = 18 Hz, 1H, 15-CH ), 5.90 (dd, J = 18 Hz, 12 Hz, 1H, 14-CH); C-NMR (75 MHz, CDCl )

2

3

δ [ppm] = 9.3 (propionyl-CH ), 16.2 (20-CH ), 18.2 (11-CH ), 19.2 (2-CH ), 22.2 (18/19-CH ),

3

2

3

2

8.0 (16-CH ), 28.6 (propionyl-CH ), 33.3 (4-C), 36.4 (18/19-CH ), 39.4 (1-CH ), 40.1 (10-

3 2 3 2

CH), 41.6 (12-CH ), 43.7 (3-CH ), 44.4 (7-CH ), 56.6 (9-CH), 57.8 (5-CH), 73.3 (6-CH), 73.8

2

(

13-C), 109.7 (17-CH ), 112.0 (15-CH ), 144.6 (14-CH), 145.4 (8-C), 173.7 (propionyl-CO);

2 2

+

HRMS [M+Na ] 385.2719, found: 385.27132 m/z.

The same protocol furnished 4.6 mg (7%) of the dipropionate. R = 0.60 (n-hexane: ethyl

f

1

acetate 9:1 (v:v)). H-NMR (300 MHz, CDCl ) δ [ppm] = 0.74 (s, 3H, 20-CH ), 0.87 (s, 3H,

3

1

2

8/19-CH ), 1.01 (s, 3H, 18/19-CH ), 1.01-1.56 (m, 12H), 1.13 (m, 6H, 2 x propionyl-CH ),

3

.52 (s, 3H, 16-CH ), 1.41 (d, J = 12 Hz, 1H, 5-CH), 2.25-2.35 (m, 4H, 2 x propionyl-CH ),

3

2

.68 (dd, J = 12 Hz, 6 Hz, 1H, 7-CH), 4.60 (m, 1H), 4.93 (m, 1H) 5.00 -5.16 (m, 3H), 5.96

13

(

dd, J = 18 Hz, 12 Hz, 1H, 14-CH); C-NMR (75 MHz, CDCl ) δ [ppm] = 8.0 (propionyl-CH ),

3 3

8

.3 (propionyl-CH ), 16.7, 18.0, 21.4, 22.6, 27.4, 27.6, 28.3, 28.7, 35.2, 38.0, 38.2, 38.8,

3

4

2.4, 43.2, 55.3, 56.5, 72.0 (6-CH), 81.9 (13-C), 109.4 (17-CH ), 112.1 (15-CH ), 140.9 (14-

2

+

CH), 143.3 (8-C), 172.3 (propionyl-CO), 172.5 (propionyl-CO); HRMS [M+Na ] 441.29808,

found: 441.29753 m/z.

Larixyl-6-carbamate: Larixol (200 mg, 0.653 mmol, 1.0 equiv.) was dissolved in

dichloromethane (3 ml), 2,2,2-trichloroacetyl isocyanate (78 µl, 0.653 mmol, 1.0 equiv.) was

added and stirred at room temperature for 1 h. Then, the solvents were removed with a

rotary evaporator and the dry product was dissolved in methanol (3 ml). Potassium

carbonate was added (97 mg, 0.702 mmol, 1.08 equiv.) and the suspension was stirred at

room temperature for 1 h. Again, the reaction mixture was evaporated to dryness. The

remainder was suspended in 5 ml water, and extracted three times with diethyl ether (10 ml

each). Combined organic phases were washed twice with water and once with brine, dried

over sodium sulfate, and evaporated to dryness. The solid remainder was purified by flash

chromatography over silica with ethyl acetate /n-hexane 1:1 as eluent, furnishing the product

as a white powder (167 mg, 73% yield). Mp: 122°C. R = 0.52 (n-hexane: ethyl acetate 1:1

f

1

(

v:v)); H NMR (500 MHz, CDCl ) δ [ppm] = 0.73 (s, 3H, 20-CH ), 0.91 (s, 3H, 18/19-CH ),

3

0.99 (s, 3H, 18/19-CH ), 1.00-1.08 (m, 1H, 12-CH ), 1.18-1.38 (m, 8H), 1.41-1.62 (m, 5H),

3 2

1

.69-1.78 (m, 2H), 1.99-2.08 (m, 1H, 7-CH ), 2.71 (dd, J = 12.3 Hz, 5.2 Hz, 1H, 7-CH ), 4.62

2

(

q, J = 1.6 Hz, 1H, 17-CH ), 4.67 (s, 2H, OCONH ), 4.87-4.94 (m, 2H, 17-CH , H-6), 5.04

2 2 2

dd, J = 10.8 Hz, 1.2 Hz, 1H, 15-CH ), 5.19 (dd, J = 17.4 Hz, 1.3 Hz, 1H, 15-CH ), 5.89 (dd,

2

13

J = 17.4 Hz, 10.8 Hz, 1H, H-14); C NMR (126 MHz, CDCl ) δ 16.04 (20-CH ), 18.10, 19.16,

3

2

4

1

2.51 (18/19-CH ), 27.78, 33.65, 36.31 (18/19-CH ), 39.24 (12-CH ), 39.90 (10-CH), 41.42,

3

2

3.67, 44.68, 56.45, 57.97, 73.58 (C-13), 74.17 (6-CH), 109.51 (17-CH ), 111.77 (15-CH ),

2

+

44.47 (C-8), 145.28 (14-CH), 156.10 (C=O); HRMS [M+Na ] 372.251, found: 372.253 m/z;

Anal. Calcd for C H NO : C, 72.17; H, 10.09; N, 4.01; Found: C, 71.54; H, 9.858; N, 3.864.

21

35

3

Larixyl-6-formate: Larixol (50 mg, 0.16 mmol, 1.0 equiv.) was dissolved in dry

dichloromethane (5 ml), formic acid (13µL, 0.326 mmol, 2.0 equiv.) and DMAP (2 mg, 0.016

mmol, 0.1 equiv.) was added. The solution was cooled to 0°C, DCC (44 mg, 0.21 mmol, 1.3

equiv.) was added and the solution was stirred for 15 min at 0°C. Thereafter, the icebath

was removed and the reaction stirred overnight at room temperature. The precipitated urea

was filtered off, washed with DCM and the filtrate evaporated under reduced pressure. The

obtained remainder was purified by flash chromatography on silica (n-hexane: 85:15 (v:v)),

yielding 48 mg (88%) of the larixyl-6-formate as white crystals. Mp. 90 °C. R = 0.48 (n-

f

1

hexane: ethyl acetate 3:1 (v:v)). H-NMR (700 MHz, CDCl ) δ [ppm] = 0.77 (d, J = 0.7 Hz,

3

1

–

H, 20-CH ), 0.91 (s, 3H, 18/19-CH ), 1.06 (s, 3H, 18/19-CH ), 1.09 (td, J = 12.9 Hz, 4.2 Hz,

3 3 3

H), 1.24 – 1.33 (m, 5H), 1.34 – 1.42 (m, 3H), 1.47 – 1.52 (m, 2H), 1.53 – 1.63 (m, 2H), 1.63

1.66 (m, 1H), 1.74 – 1.80 (m, 2H), 2.09 – 2.15 (m, 1H, 7-CH ), 2.73 (dd, J = 12.3 Hz, 5.1

2

Hz, 1H, 7-CH ), 4.69 (d, J = 1.6 Hz, 1H, 17-CH ), 4.97 (d, J = 1.6 Hz, 1H, 17-CH ), 5.08 (dd,

2

J = 10.8 Hz, 1.2 Hz, 1H, 15-CH ), 5.16 – 5.21 (m, 1H, H-6), 5.22 (dd, J = 17.4 Hz, 1.2 Hz,

2

13

1

H, 15-CH ), 5.93 (dd, J = 17.4 Hz, 10.8 Hz, 1H, H-14), 8.10 (d, J = 0.9 Hz, 1H, COH);

C

2

NMR (176 MHz, CDCl ) δ 15.93 (20-CH ), 18.01, 18.97, 22.37 (18/19-CH ), 27.80 (16-CH ),

3

3.70, 36.31 (18/19-CH ), 39.10, 39.93, 41.27, 43.50, 44.24, 56.32, 57.47 (9-CH), 73.30 (5-

3

CH), 73.33 (C-6), 73.46 (C-13), 109.85 (17-CH ), 111.72 (15-CH ), 144.02 (8-C), 145.14 (14-

2

+

CH), 160.64 (CHO); HRMS [M+Na ] 357.24, found: 357.242 m/z; Anal. Calcd for C H O :

21

34

3

C, 75.41; H, 10.25; Found: C, 75.43; H, 9.66.

Larixyl-6-phenylacetate: Larixol (50 mg, 0.16 mmol, 1.0 equiv.) was dissolved in dry

dichloromethane (3 mL), phenylacetic acid (33 mg, 0.245 mmol, 1.5 equiv.) and DCC (50

mg, 0.245 mmol, 1.5 equiv.) was added. After 5 min, DMAP (4 mg, 0.032 mmol, 0.2 equiv.)

was added and the solution was stirred overnight at room temperature. The precipitated urea

was filtered off, washed with DCM and the organic phase extracted with 15 mL NaHCO₃.

The organic phase was dried over MgSO and evaporated under reduced pressure. The

4

product was purified by flash chromatography on silica (n-hexane: ethyl acetate 85:15 (v:v)),

yielding 37 mg (53%) of the larixyl-6-phenylacetate as colorless oil. R = 0.52 (n-hexane:

f

1

ethyl acetate 3:1 (v:v)). H-NMR (700 MHz, CDCl ) δ [ppm] = 0.75 (s, 3H, 20-CH ), 0.87 (s,

3

1

H, 18/19-CH ), 0.98 (s, 3H, 18/19-CH ), 1.07 (td, J = 12.7 Hz, 4.1 Hz, 1H, 12-CH ), 1.22 –

3

2

.32 (m, 5H), 1.32 – 1.41 (m, 3H), 1.44 (d, J = 11.3 Hz, 1H, H-5), 1.46 – 1.51 (m, 1H), 1.52 –

.64 (m, 4H), 1.72 – 1.79 (m, 2H), 1.96 – 2.02 (m, 1H, 7-CH ), 2.66 (dd, J = 12.3 Hz, 5.1 Hz,

2

1

H, 7-CH ), 3.57 – 3.66 (m, 2H, CH -Ph), 4.65 (d, J = 1.6 Hz, 1H, 17-CH ), 4.92 (d, J = 1.7

2

Hz, 1H, 17-CH ), 5.03 – 5.12 (m, 2H, 15-CH , H-6), 5.22 (dd, J = 17.3 Hz, 1.2 Hz, 1H, 15-

2

CH ), 5.92 (dd, J = 17.3 Hz, 10.8 Hz, 1H, H-14), 7.26 – 7.32 (m, 3H, Ph), 7.32 – 7.36 (m, 2H,

2

13

Ph); C-NMR (176 MHz, CDCl ) δ 15.99 (20-CH ), 18.01, 19.01, 22.45 (18/19-CH ), 27.74

3

(

16-CH ), 33.51, 36.15 (18/19-CH ), 39.13 (12-CH ), 39.89 (10-CH), 41.30, 42.28, 43.45,

3 3 2

4

1

3.96 (7-CH ), 56.36 (9-CH), 57.45 (5-CH), 73.48 (C-13), 73.88 (6-CH), 109.56 (17-CH ),

2

11.69 (15-CH ), 127.01 (Ph-C-4), 128.50 (Ph-C-3/3´), 129.29 (Ph-C-2/2´), 133.95 (Ph-C-1),

2

+

44.24 (C-8), 145.16 (C-14), 170.64 (C=O); HRMS [M+Na ] 447.28, found 447.287 m/z;

Larixyl-6-methylether: Larixol (113 mg, 0.367 mmol, 1.0 equiv.) was dissolved in dry

tetrahydrofuran (3 ml) and cooled to 0°C. Then sodium hydride (29 mg, 60% suspension in

mineral oil, 0.735 mmol, 2.0 equiv.) was added and the solution stirred for 5 min, followed by

the addition of iodmethane (69 µL, 1.10 mmol, 3.0 equiv.). After 5 min, the icebath was

removed and the reaction stirred overnight at room temperature. Additional 20 mg sodium

hydride (20 mg) and iodomethane (69 µL) was added and stirred for additional 24h. The

solvent was evaporated under reduced pressure. The obtained remainder was suspended in

dichloromethane and extracted with water (15 mL). The organic phase was dried over

MgSO and evaporated under reduced pressure. The product was purified by flash

4

chromatography over silica (n-hexane: ethyl acetate 85:15 (v:v)), yielding 42 mg (36%) of a

62:38 (determined by NMR integration) mixture of larixyl-6-methylether (A) and larixyl-16-

1

methylether (B) as colorless crystals. R = 0.4 (n-hexane: ethyl acetate 4:1 (v:v)). H NMR

f

(

700 MHz, CDCl ) δ [ppm] = 0.70 – 0.72 (m, 5H, 20-CH of A+B), 0.94 (s, 3H, 18/19-CH of

3 3 3

A), 1.03 (s, 2H, 18/19-CH of B), 1.06 (td, J = 12.7, 4.0 Hz, 2H, 12-CH of A+B), 1.10 – 1.14

3

2

(

m, 4H), 1.18 (s, 2H, 18/19-CH of B), 1.23 – 1.41 (m, 1H), 1.44 – 1.52 (m, 2H), 1.52 – 1.63

3

m, 4H), 1.71 – 1.81 (m, 3H), 1.88 – 1.95 (m, 1H, 12-CH of A), 2.06 (td, J = 11.5 Hz, 10.6,

2

1.8 Hz, 0.6 H, 12-CH of B), 2.69 (dd, J = 12.2 Hz, 4.9 Hz, 0.6H, 12-CH of B), 2.86 (dd, J =

2 2

1

2.2 Hz, 4.7 Hz, 1H, 12-CH of A), 3.15 (s, 2H, OCH of B), 3.26 (td, J = 10.6 Hz, 4.7 Hz, 1H,

2

3

H-6 of A), 3.34 (s, 3H, OCH of A), 3.84 (td, J = 10.7 Hz, 4.9 Hz, 0.6H, H-6 of B), 4.60 (q, J =

3

1.6 Hz, 0.6H, 17-CH of B), 4.62 (q, J = 1.6 Hz, 1H, 17-CH of A), 4.89 (q, J = 1.6 Hz, 0.6H,

2 2

1

7-CH of B), 4.91 (q, J = 1.6 Hz, 1H, 17-CH of A), 5.07 (dd, J = 10.8 Hz, 1.2 Hz, 1H, 15-

2

CH of A), 5.13 (dd, J = 17.7 Hz, 1.3 Hz, 0.6H, 15-CH of B), 5.18 (dd, J = 10.9 Hz, 1.3 Hz,

2

0.6H, 15-CH of B), 5.22 (dd, J = 17.4, 1.3 Hz, 1H, 15-CH of A), 5.76 (dd, J = 17.7, 11.0 Hz,

2 2

13

0

.6H, H-14 of B), 5.93 (dd, J = 17.4 Hz, 10.8 Hz, 1H, H-14 of A); C NMR (176 MHz, CDCl )

3

δ [ppm] = 16.08 (20-CH of B), 16.29 (20-CH of A), 17.70, 18.10, 19.14, 21.61, 22.38,

3

22.66, 27.82, 29.71, 33.82, 33.90, 36.46, 36.66, 38.50, 39.32, 39.41, 39.56, 41.40, 43.25 (7-

CH of A), 43.66, 43.79, 49.23 (7-CH of B), 50.07 (OCH of B), 55.76, 56.60, 56.75, 58.91

2

3

(

OCH of A), 60.59, 71.71 (C-6 of A), 73.53, 77.56, 80.76 (C-6 of B), 108.21 (17-CH of A),

3 2

1

3

08.27 (17-CH of B), 111.62 (15-CH of A), 114.63 (15-CH of B), 143.03 (14-CH of A),

2 2 2

+

45.21 (14-CH of B), 145.70 (C-8 of B), 145.78 (C-8 of A); HRMS [M+Na ] 343.26, found:

43.265 m/z; Anal. Calcd for C H O : C, 78.70; H, 11.32; Found: C, 78.78; H, 11.44.

21

36

2

Mutagenesis of human TRPC6

To obtain pcDNA3-based expression plasmids, encoding FSGS-related point-mutated

TRPC6, site-directed mutagenesis of human TRPC6-YFP was performed with the

QuikChange kit (Agilent Technologies, Santa Clara, CA, USA), following the manufacturer´s

recommendations and applying suitable forward and reverse primers (sequences are

provided in Supplementary Table 1). After PCR, mutated plasmids were transformed in XL2-

Blue competent cells. Clones were selected, and the mutation was verified by cDNA

sequencing.

Cell culture and transfection of HEK293 cells

HEK293 cells were cultured in Earle´s Minimum Essential Medium (MEM; Sigma, Munich,

Germany), supplemented with 10% fetal calf serum (Gibco Thermo Fisher Scientific,

Darmstadt, Germany), 2 mM L-glutamine, 100 U/ml penicillin, and 0.1 mg/ml streptomycin.

To characterize human TRPC6 mutants and to verify the efficiency of larixol derivatives,

parental HEK293 cells were seeded on 25-mm glass coverslips, and transfected with 4 µl

Fugene HD (Promega, Mannheim, Germany) and 2 µg of DNA encoding wild-type or

mutated variants of TRPC6-YFP. For FRET assays, HEK293 cells were co-transfected with

.5-0.8 µg pcDNA3-hTRPC6-CFP wild-type and 1.2-1.5 µg pcDNA3-hTRPC6-YFP mutant to

achieve a 1:1 molar ratio between YFP- and CFP-fused channel constructs. After 24 hours,

0

2+

cells were directly used for [Ca ] imaging and FRET measurements, or singularised with

i

0.25% trypsin for electrophysiological experiments. Fluorescence plate imaging analyses of

cells expressing TRPC6 mutants were realized with stably transfected polyclonal cell lines,

selected by adding 1 mg/ml geneticin (G418) to the medium. All cells were maintained at

3

7°C and in a 5% CO -aerated, humidified atmosphere.

2

Isolation of glomeruli and podocyte primary cultures

Kidneys were isolated by dorsal dissection from 10-days-old (P10) or 8-week-old Wistar rats

or from 3- to 4-week-old mice. Mouse strains were C57BL/6J or B6.Cg-Tg(NPHS2-Trpc6)

F419Walz/J transgenic mice intercrossed on C57BL/6J background, purchased via Charles

River Laboratories from The Jackson Laboratory (Bar Harbor, MA, USA; Stock number

0

18293). After removing the fibrous renal capsule, cortical tissue was obtained, cut in small

pieces and washed in ice-cold buffer, containing 127 mM NaCl, 5.9 mM KCl, 1.2 mM MgCl₂,

.8 mM glucose, and 10 mM HEPES adjusted to pH 7.4. Then, the collected tissue was

enzymatically digested for 12-15 minutes at 37°C in the same buffer, additionally containing

.75 mg/ml collagenase (Sigma), 50 µM CaCl and 0.5% bovine serum albumin. The

1

0

2

digested tissue was recovered by short centrifugation, and resuspended in a podocyte

culture medium, containing RPMI 1640 supplemented with 10% fetal calf serum, 2 mM L-

glutamine, 100 U/ml penicillin, and 0.1 mg/ml streptomycin. The suspension, which

contained decapsulated glomeruli, was repeatedly sieved through cell strainers (Greiner)

with decreasing pore sizes (100 µm, 70 µm and 40 µm). Glomeruli-enriched fractions were

obtained by turning and rinsing the sieve size 70 µm (adult rat) or 40 µm (P10 rats and mice)

2+

with medium. For Ca measurements, isolated glomeruli were directly loaded with 5 µM

fura-2/AM (AAT Bioquest, Hamburg, Germany) or 4 µM fluo-4/AM (Invitrogen, Karlsruhe,

Germany) in podocyte culture medium for 30 min at 37°C, centrifuged (100 x g; 5 min) and

measured 30 min after allowing glomeruli to settle in collagen-coated 24-well cell culture

plates or pigmented clear-bottom 384-well plates (Corning, Kaiserslautern, Germany) in HBS

containing 0.1% BSA. In case of isolation of primary podocyte cultures, glomeruli suspended

in podocyte culture medium were seeded into 24-well plates which were freshly coated with

collagen A (Biochrom, Berlin, Germany) or into collagen type I-coated 60-mm dishes

(

Greiner, Frickenhausen, Germany). Analyses were done 5 to 8 days after growing out of

podocytes from the glomeruli. The fraction of cells with cobblestone-like morphology was

over 80%.

PASMC isolation

Pulmonary artery smooth muscle cells (PASMC) were obtained from 3-week-old Wistar rats

or from 3- to 4-week old mice, following a previously described procedure (Urban et al.,

2016). PASMC were cultured for 1 to 2 weeks in DMEM/Ham’s F12 medium (1:1 vol/vol),

supplemented with 10% fetal calf serum, 2 mM L-glutamine, 2.7 g/l D-glucose, 25 mM

HEPES, 100 U/ml penicillin, and 0.1 mg/ml streptomycin before preparation of RNA was

2+

done. For protein expression analysis, functional TRPC6 [Ca ] imaging and membrane

i

potential measurements, PASMC were used up to the seventh passage.

2+

Ca measurements in multiwell plates

2+

Fluorometric Ca assays in HEK293 cell lines and glomeruli were done in HEPES-buffered

saline (HBS), containing 132 mM NaCl, 6 mM KCl, 1 mM MgCl , 1mM CaCl , 5.5 mM D-

2

glucose, and 10 mM HEPES, 0.03% bovine serum albumin (BSA) adjusted to pH 7.4 with

NaOH. Isolated glomeruli or trypsin-detached HEK293 cell lines were loaded with 4 µM fluo-

4/AM (Invitrogen) in culture medium for 30 min at 37°C. After centrifugation (100 x g; 3 min),

glomeruli and HEK293 cells were resuspended in HBS and dispensed into pigmented clear-

bottom 384-well plates (Corning, USA). Fluorescence signals were recorded in a custom-

made fluorescence imaging plate reader (FLIPR) built into a robotic liquid handling station

(

Freedom Evo 150, Tecan, Männedorf, Switzerland) as described (Urban et al., 2016). First

TRPC6 inhibitors were added to stably transfected HEK293 cells at the indicated final

concentrations, and incubated for 5 min. Then, cells were stimulated with 50 µM OAG during

continuous fluorescence recording, controlled by a Micromanager software (Edelstein et al.,

2010). Fluorescence intensities were calculated from image stacks for each well with ImageJ

(

Abramoff & Magalhaes, 2004), corrected for background signals and normalized to the

initial intensities (F/F ). Concentration response curves were generated by fitting the data to

0

Hill equations. If freshly isolated glomeruli were examined, larixyl carbamate was incubated

for 10 min, and TRPC6 activation was indirectly triggered with 10 µM angiotensin II.

2+

Microfluorometric [Ca ] imaging analysis

i

Transfected HEK293 cells or PASMC grown on 24-mm coverslips or primary podocyte

cultures grown in 24-well plates were loaded with 4 µM fura-2/AM (AAT Bioquest) in HBS

containing 0.2% BSA for 30 min at 37°C. After washing cells, podocytes and PASMC were

allowed to equilibrate in HBS containing 0.1% BSA and 1 µM bisindolylmaleimide 1 (BIM) to

prevent activation of protein kinase C. HEK293 cells on coverslips were mounted in a bath

2+

chamber and superfused with HBS containing 0.1% BSA. [Ca ] measurements were

i

performed on an inverted epifluorescence microscope with a Fluar 10x/0.5 objective (Carl

Zeiss, Jena, Germany) and calibrated as described (Urban et al., 2016;Lenz et al., 2002).

Heterologously expressed TRPC6 in HEK293 cells was either activated with 50 µM OAG or

indirectly stimulated by challenging cells with a mixture of 1 mM carbachol, 300 µM ATP and

2+

0

.5 units/ml thrombin after internal Ca store depletion with 2 µM thapsigargin. To evaluate

the inhibitory potency of new TRPC6 blockers, HEK293 cells were pretreated with 1 µM of

respective compound. Activation of endogenous TRPC6 in PASMC, podocytes or freshly

isolated glomeruli was attempted with 100 µM OAG, 20 µM GSK1702934A, 10 µM

angiotensin II, 100 µM ATP, 10 µM bradykinin, 100 µM carbachol or 0.5 units/ml thrombin. In

2

+

some experiments, Ca mobilisation from the endoplasmic reticulum was abrogated by

adding 2 µM thapsigargin to the bath 5 min before application of the respective agonists.

Fluorescence resonance energy transfer (FRET) measurements

FRET was assessed and quantified by measuring donor (TRPC6-CFP) unquenching during

selective photobleaching of the FRET acceptor (YFP fused to the C terminus of wild-type or

mutant TRPC6) essentially as described earlier (Hellwig et al., 2005). To account for the

expected 1:1 gene dose in FSGS patients, the molar ratio between co-expressed CFP- and

YFP-tagged TRPC6 constructs was determined as described, and adjusted to achieve only

a slight (10-30%) excess of the YFP-fused FRET acceptor over the co-expressed CFP-fused

donor.

Genotyping and quantitative RT-PCR

After reproduction of B6.Cg-Tg(NPHS2-Trpc6) F419Walz/J transgenic mice, genotyping of 2

week old littermates was performed, applying a REDExtract-N-Amp Tissue PCR Kit (Sigma),

a Trpc6 exon 8 forward primer 5’-gacactgttctgggctatct and a Trpc6 3’ UTR reverse primer 5’-

cagtgtgatggagctcga as reported (Krall et al., 2010). As an internal control, the endogenous

Trpc6 gene was amplified, using a Trpc6 intron 8 reverse primer 5’-cccattttcctctccccaaa.

Amplicons were detected in gel electrophoresis as 830 bp product (transgene), and 530 bp

(

endogenous TRPC6) bands, respectively.

To quantify the TRPC6 transcript abundance, total RNA was isolated from primary

podocytes and PASMC with a Trizol reagent (VWR Life Science, Erlangen, Germany). First

strand cDNA was synthesized from RNA templates with a RevertAid first strand cDNA

synthesis kit (Thermo Scientific) and random hexamer primers, following the manufacturer´s

instructions. Finally, cDNA was applied directly as template in real-time PCR using DyNAmo

ColorFlash SYGR Green qPCR Kit and a Piko Real 24 thermocycler (Thermo Scientific).

Primers for detecting native TRPC6 and smooth muscle -actin transcripts were

independently designed for rat and mouse TRPC6. For quantitative RT-PCR analysis of

podocin expression, common mouse/rat primers were used. Transgenic TRPC6 RNA was

detected with a Jackson Laboratory genotyping primer and a reverse hemagglutinin (HA) tag

primer. PCR was performed with a three-step protocol (7 min at 95°C; and 45 cycles of 15 s

at 95°C, 20 s at 56°C/ 63°C/ 66°C, 30s at 72°C; final extension of 2 min at 72°). Results

CT(reference) –

were normalized to 18s RNA, using the ∆C method (ratio reference/target = 2

T

CT(target)

)

. All applied primers are listed in the Supplementary Table 1.

Western blot and immunofluorescence analysis

Cultured podocytes and PASMC were washed twice with PBS, harvested with a cell scraper

and homogenized with a RIPA buffer, containing 150 mM NaCl, 50 mM Tris-HCl pH 7.4, 1

mM EDTA, 1% NP-40, 0.25% natrium deoxycholate, 1 mM NaF, 1 mM natrium

orthovanadate, 1 mM PMSF, 1 µg/ml aprotinine, 1 µg/ml leupeptine, and 1 µg/ml pepstatin.

Homogenisation of glomeruli in RIPA was done directly after isolation and two washing steps

with PBS. To assess protein yield of the samples, a BCA protein assay was used (Thermo

Fisher Scientific). Then, 30 µg protein of glomeruli or HEK293 cells, and 18 µg protein of

podocytes or PASMC were lysed in Laemmli buffer, and subjected to SDS-PAGE (8%

acrylamide/bis-acrylamide). Separated proteins were electroblotted onto PVDF membranes.

Membranes were blocked for 30 min in 5% nonfat dry milk TBST (20 mM Tris pH 7.4, 137

mM NaCl, 0.1% Tween-20) and incubated overnight with anti-TRPC6 (Alomone, Jerusalem,

Israel; 1:200), anti-podocin (Sigma; 1:2,000) or anti-α smooth muscle actin (α-SMA) (Sigma;

1:1,000) antibodies in TBS, 1% dry milk, 0.5% Tween-20. As a housekeeping protein,

GAPDH was analysed by 2 hours incubation with anti-GAPDH (Millipore, Darmstadt,

Germany; 1:5,000) or anti-ß-actin (Sigma; 1:5,000). Same-day or next day, membranes

were washed, followed by incubation with secondary peroxidase-coupled anti-rabbit

(

TRPC6, podocin, GAPDH) or anti-mouse (α-SMA, ß-actin) antibodies (Sigma; 1:5,000) for 2

hours. Peroxidase activity was detected using a Novex ECL chemiluminescence reagent

Invitrogen). For TRPC6 analysis, a more sensitive peroxidase substrate SuperSignal West

(

Femto (Thermo Fisher Scientific) was necessary. Between detecting the different target

proteins, the PVDF membranes were stripped with a 1 M glycine solution adjusted to pH 1.8

with HCl.

For immunofluorescence, freshly isolated glomeruli were transferred onto glass slides, dried

quickly under a ventilator and fixed with methanol. Primary isolated podocytes were split

onto freshly collagen A-coated µ slides (ibidi) for 2 days and fixed with 4%

paraformaldehyde. Then, samples were blocked and permeabilized with PBS, supplemented

with 3% BSA, 0.1% Triton-X100 and 10% goat serum. Primary antibody decoration of slides

was followed overnight with polyclonal rabbit anti-podocin (Sigma; 1:250) and monoclonal

mouse anti-α-SMA (Sigma; 1:500). After washing slides three times for 5 minutes in PBS,

bonded primary antibodies were decorated with Alexa Fluor 488-conjugated goat anti-rabbit

antibody (Invitrogen A11008; 5 µg/ml) and Cy3-conjugated donkey anti-mouse antibody

(

Jackson ImmunoResearch Laboratories; 1:400) for 2 hours. Finally, nuclei were

counterstained with Hoechst 33258 (10 µg/ml), slides were washed again and mounted with

Mowiol 4-88 (Roth) or ibidi mounting medium (ibidi). Fluorescence detection was performed

with a LSM 510 META confocal microscope (Carl Zeiss), using Plan-Apochromat 63x/1.3 or

20x/0.8 objectives.

MTT assay/ Detection of cell viability

Cell viability was assessed in parental HEK293 cells with a (3-(4,5-dimethylthiazol-2-yl)-2,5-

diphenyltetrazolium bromide) (MTT) assay. Cells were seeded in poly-L-lysine-coated 96-

well plates and grown for 24 hours until monolayers reached 50% confluency. Then medium

was exchanged, and larixyl derivatives or the corresponding DMSO solvent controls were

added and incubated for another 24 hours. After 24 hours, the medium was replaced with

fresh one, but now containing 0.5 mg/ml MTT. After 3 hours of incubation, supernatants

were removed, cells were lysed and formazan crystals were completely dissolved with

undiluted DMSO. Finally, absorbance measurements at 560 nm and 670 nm were done with

a plate reader (Polarstar Omega, BMG Labtech, Germany). The differences of both

extinctions were calculated, and values were normalized to the 0.25% solvent control.

Membrane potential measurements

The determination of membrane potential changes by LC and LMA were executed in

PASMC using the fluorescence dye DiBAC (3). PASMC grown on 24-mm coverslips were

4

mounted in a bath chamber and loaded for 15 to 20 min at 37°C with 500 nM DiBAC (3)

4

(

AAT Bioquest) in HBS, containing 2 mM CaCl . DiBAC (3) fluorescence of PASMC was

2 4

continuously detected on an inverted microscope (Axiovert 100 with a Fluar 10x/0.5

objective, Carl Zeiss, Germany) equipped with a monochromator (Polychrome II, TILL-

Photonics, Germany), and a 12-bit cooled CCD camera (Flowmaster, LaVision) by exciting

the dye at 480 nm and collecting emitted light through a 512-nm long pass filter. The test

compounds were added after a baseline recording of 1 minute.

Electrophysiological procedures

HEK_h_T_R_P_C₆_-_Y_F_Pcells and podocytes were seeded at low density on PLL-coated glass

coverslips. After 6-24 hours, TRPC6 membrane currents were recorded in the whole-cell

patch clamp configuration, using a Multiclamp 700B amplifier with a Digidata 1440A digitizer

controlled by the PClamp 10 software (Molecular Devices, Sunnyvale, CA, USA). Coverslips

were mounted in a chamber (Warner Instruments, Hamden, CT; USA), placed onto the

stage of an inverted microscope, and perfused with standard extracellular solution contained

1

40 mM NaCl, 5 mM CsCl, 2 mM MgCl , 1 mM CaCl , and 10 mM HEPES, 0.1% BSA

2 2

adjusted to pH 7.4 with NaOH. In case of podocyte stimulation with OAG, the bath solution

additionally contained 1 µM BIM to suppress simultaneous activation of PKC. Patch pipettes

were pulled from borosilicate glass capillaries BG150F-8P (Science Products) with a pipette

puller PIP6 (HEKA Elektronik, Lambrecht, Germany) and had a resistance of 4-10 MΩ when

filled with the standard pipette solution: 140 mM CsCl, 4 mM MgCl , 10 mM EGTA, and 10

2

mM HEPES adjusted to pH 7.2 with CsOH. In all whole-cell measurements, serial resistance

was lower than 13 MΩ, and compensated by 70%. To record current/voltage (I/V) curves,

-1

voltage ramps (400 mV s ) ranging from -100 mV to +100 mV were applied. Currents were

filtered at 3 kHz with a four-pole Bessel filter and sampled at 10 kHz.

Nomenclature of Targets and Ligands

Key protein targets and ligands in this article are hyperlinked to corresponding entries in

http://www.guidetopharmacology.org, the common portal for data from the IUPHAR/BPS

Guide to PHARMACOLOGY (Southan et al., 2016), and are permanently archived in the

Concise Guide to PHARMACOLOGY 2015/16 (Alexander et al., 2015).

Results

Derivatisation of larixol

Since monoacetylation of larixol at its secondary alcohol was known to enhance the potency

of TRPC6 inhibition in larixyl monoacetate (LMA) (Urban et al., 2016), we sought to add a

second acetylation at the tertiary alcohol of larixol, and/or to replace the acetate by other

substituents to further improve the properties of the TRPC6-inhibiting pharmacophore. Under

harsher reaction conditions, various carboxylate esters were formed, which included

modification of both hydroxyl groups of larixol. Of them, larixyl diacetate and larixyl

dipropionate were less potent in inhibiting wild-type TRPC6 in a stably transfected

HEK_h_T_R_P_C₆_-_Y_F_Pcell line than the previously characterised LMA or a larixyl mono-6-propionate

(

Fig. 1A-D). Despite its reasonably TRPC6-prevalent mode of action, higher IC₅₀values

argued against a favourable modification at the tertiary alcohol of the 3-hydroxy-3-

methylpent-4-enyl group of larixol. Instead, we modified the hydroxyl group at the C6 atom of

the diterpene ring. Here, the addition of a methyl ether, or of formyl or propionyl groups did

not significantly improve the potency of TRPC6 inhibition compared to LMA. Likewise, the

selectivity to prevent the OAG-induced activation of TRPC6 compared to the activation of

stably transfected TRPC3 or TRPC7 was similar or lower than that of LMA. Of note, larixyl

methyl ether, larixyl formate, LMA, and larixyl propionate elicited a concentration-dependent

biphasic effect on TRPC7, with a slight potentiation at lower concentrations and channel

inhibition at higher concentrations. With an IC of 36 µM, a larger phenylacetate-substituted

50

larixol was only poorly active and selective (Fig. 1H). Reasoning that the ester bond in LMA

may be readily cleaved by intracellular esterases e.g. in the liver, we replaced the acetate

group in LMA by a carbamate group. Larixyl carbamate (LC) displayed a favourable potency

with an IC to inhibit wild-type TRPC6 of 0.48 µM (Fig. 1C). Of note, the selectivity to inhibit

50

TRPC6, but not TRPC3 was 30-fold and, thus, slightly higher than that of the previously

reported LMA. For comparison, we assessed the inhibition of TRPC3, TRPC6 and TRPC7

by the compound SKF-96365 in the same assay. As expected, SKF-96365 was moderately

potent and poorly specific. The corresponding IC₅₀values were 5.0 µM, 7.1 µM, and 19 µM

to inhibit TRPC6, TRPC3, and TRPC7, respectively (Fig. 1I).

To assess a potential cytotoxicity of the new larixol derivatives, we executed an MTT assay

in parental HEK293 cells, applying the three highest compound concentrations used in the

FLIPR analysis (Fig.1B-H). All compounds stocks were dissolved in DMSO at a

concentration of 10 mM. Compared to the respective solvent control, none of the TRPC6

inhibitors significantly affected the cell viability at a concentration of 25 µM, with the

exception of larixyl diacetate (Fig.1J). Of note, LC concentrations necessary for full TRPC6

2+

block we determined at only 5 µM in Ca measurements and 1 µM in electrophysiology.

Generation and functional expression of FSGS-related TRPC6 mutants

Plasmids which encode TRPC6 constructs that carry FSGS-related point mutations were

generated by site-directed mutagenesis and confirmed by cDNA sequencing. Upon transient

2+

heterologous expression in HEK293 cells, fluorometric single cell [Ca ] analyses

i

reproduced the gain of function phenotype in the P112Q, M132T, R175Q, and Q889K point-

mutated TRPC6 constructs. Here, the basal as well as the OAG (50 µM)-induced peak

2+

[

Ca ] were up to 3-fold higher than in cells expressing the wild-type TRPC6. Of the two

i

remaining constructs, R895C and E897K, only R895C gave rise to a slightly higher basal

2+

[

Ca ], but the stimulated peak [Ca ] displayed only a statistically not significant trend to

i i

higher values (Table 1).

Since TRPC6-linked FSGS is an autosomal dominant inheritable disease, patients have a

second allele, which encodes wild-type TRPC6. Using fluorescence resonance energy

transfer (FRET) analyses, we tested whether YFP-fused mutated TRPC6 constructs can still

form oligomeric assemblies with wild-type TRPC6, which was C-terminally fused to CFP.

When transiently co-expressed with a 1.1- to 1.3-fold molar excess of YFP-fused constructs

over TRPC6-CFP, all investigated TRPC6 mutants gave rise to robust FRET signals,

ranging from FRET efficiencies of 15.9% to 19.7%. Under the same experimental conditions,

the wild-type TRPC6-CFP:TRPC6-YFP complex developed a FRET coupling efficiency of

18.0%. From these data, we conclude that all investigated TRPC6 mutations are still capable

of forming channel complexes together with TRPC6 subunits that are encoded by the wild-

type allele (Table 2).

2+

Inhibition of Ca entry through FSGS-mutated TRPC6 by larixol derivatives

The efficiency of LMA to inhibit wild-type or mutated TRPC6 in a GPCR-mediated activation

2+

mode was tested by calibrated single-cell [Ca ] analysis in fura-2-loaded HEK293 cells,

i

which were transiently transfected to express wild-type TRPC6 or FSGS-related mutant

TRPC6 variants. Cells grown on glass coverslips were mounted into bath chambers,

exposed to HBS, containing either 1 µM LMA or the solvent, and then incubated with 2 µM

2+

thapsigargin for 5 min to deplete inositol 1,4,5-trisphosphate-sensitive Ca

stores.

Thereafter, an agonist mix, comprising maximally effective concentrations of ATP (300 µM),

carbachol (1 mM), and thrombin (0.5 units/ml) was added to the bath to activate G-protein-

sensitive phospholipases C and, in turn, TRPC6. Irrespective of the expression of wild-type

2+

or mutated TRPC6 variants, [Ca ] signals that were elicited by the agonist mix were

i

strongly and significantly attenuated if cells were pretreated with 1 µM LMA (Fig. 2). When

compared the findings in the OAG-activated mode (see Table 1), the relative order of

channel hyperactivity was similar with P112Q, M132T, R175Q and Q889K mutants causing

the strongest enhancement of signals compared to wild-type TRPC6 or the R895C and

2+

E897K mutants. Of note, the elevated basal [Ca ] in cells overexpressing the M132T

i

variant (Fig. 2C) was efficiently lowered in the presence of 1 µM LMA. A remaining, small

2+

[

Ca ] signal seen in LMA-treated cells expressing the P112Q, M132T, or R175Q variants

i

may either indicate partial inhibition of the channel or an incomplete store depletion.

To quantitatively assess the potency of inhibition of FSGS-related TRPC6 variants by LMA

and by the newly synthesized LC, we generated stably transfected cell lines for each of the

2+

six TRPC6 channel mutants, and applied a semi-quantitative multiwell Ca analysis in fluo-

2+

4

-loaded cell suspensions. Here, a complete inhibition of OAG (50 µM)-induced Ca signals

was seen in cells pretreated with 7.5-15 µM of either LMA or LC. Concentration response

curves obtained by applying various concentrations of LMA or LC revealed IC₅₀values of

0.35-0.65 µM LMA for most TRPC6 constructs. The R175Q mutant appeared to be inhibited

with a slightly lower potency (IC₅₀= 0.91 µM). Applying LC as the TRPC6 inhibitor, we

obtained very similar results with IC₅₀values of 0.41-0.66 µM for five TRPC6 variants, and

.03 µM for the R175Q construct (Fig. 3).

1

Electrophysiological properties of TRPC6 inhibition

As expected, ionic currents through mutant TRPC6 constructs resembled those of wild-type

TRPC6 in terms of current-voltage-relationship, and their activation by 50 µM OAG (Fig. 4).

With the exception of the E897K mutant, peak current densities were markedly higher than

those seen in cells expressing the wild-type TRPC6. In all investigated channel constructs,

the acute addition of 0.2 µM LC partially and 1 µM LC almost completely abolished the OAG-

induced inward and outward currents measured at -100 mV and +100 mV, respectively. The

block by 0.2 µM or 1 µM LC developed within a few seconds, which was sufficiently fast to

qualitatively discern the channel inhibition from the slower channel inactivation, which was

observed in cells that were not exposed to the inhibitor (see grey traces in Fig. 4B-H).

To obtain more detailed information about the inhibition of TRPC6 currents by LC in an

equilibrium state of inhibitor binding, we exposed stably transfected HEK_h_T_R_P_C₆_-_Y_F_Pcells to

various concentrations of LC prior to stimulating channel activity with 50 µM OAG. In voltage

ramp protocols applied in 1-s intervals, both inward currents measured at -100 mV and

outward currents at +100 mV were concentration-dependently suppressed by LC (Fig. 5A-

C). The corresponding IC values were 118 and 120 nM, respectively. Accordingly, currents

50

measured during application of voltage ramps showed no signs of a voltage-dependent

inhibition, and reversal potentials remained unchanged (Fig. 5B), indicating no major impact

on the selectivity of conducted cations. In HEK293 cells expressing mutated TRPC6

constructs, preincubation with 1 µM LC for 1 min prior to OAG application significantly

inhibited all tested TRPC6 variants (Fig. 5D). Since the inhibition amounted 70%-95%, we

conclude that, like in wild-type TRPC6, the IC₅₀of LC to inhibit FSGS-related TRPC6

mutations is in a sub-micromolar concentration range.

Larixyl carbamate does not cause changes of the membrane potential and of TRPC6

expression

Since some pimarane compounds, which are chemically related to larixol, have been

2+

+

identified as modulators of large conductance Ca -activated K (BK) channels, we

investigated whether LC can alter the membrane potential (E ) in rat PASMC that

m

endogenously express BK channels and voltage-dependent calcium channels (VDCC).

2+

Since the acute addition of 5 µM LC did not cause an increase in [Ca ] in fura-2-loaded

i

PASMC (Fig. 6A), an efficient activation of VDCC could be excluded. In PASMC loaded with

the E -sensitive fluorescent dye DiBAC (3), the direct TRPC6 activator GSK1702934A (10

m

4

µM) caused a significant increase in DiBAC (3) fluorescence by 34.4% ± 17.3% (n = 8),

4

indicating a depolarisation. The known BK channel activator NS1619 (30 µM) caused a

decrease in DiBAC (3) fluorescence by 42.9% ± 11.4% (n

=

7), indicating

a

4

hyperpolarisation. By contrast, neither 5 µM LMA nor 5 µM LC caused a change in

DiBAC (3) fluorescence compared to the solvent controls (Fig. 6B,C).

4

To analyse a possible impact of long-term incubation with LC on the global expression of

TRPC6 in recombinantly and in natively TRPC6 expressing cells, Western blot analyses

were performed in samples obtained from stably transfected HEK_h_T_R_P_C₆_-_Y_F_Pcells and from rat

PASMCs (three independent experiments, each) that were grown for 24 h in the presence of

5

and 10 µM LC. The global abundance of heterologously expressed or of native TRPC6

remained largely unaffected by the TRPC6 inhibitor (Fig. 6D-F). Furthermore, since OAG-

2+

induced increases in [Ca ] measured in PASMC after long-term (24 h) treatment with LC,

i

followed by an acute wash-out of the inhibitor were not significantly different from those

measured without preincubation of the TRPC6 inhibitor (see Fig. 6A), we conclude that the

availability of TRPC6 channels at the cell surface is presumably not grossly affected by long

term treatment with LC.

Isolation of glomeruli and of primary podocyte cultures

The isolation of mouse or rat glomeruli from the renal cortex usually involves a fractional

sieving technique. Critical differences may rely on the way how glomeruli are being detached

from surrounding tissue. Different options have been described: first a mechanical

dissociation method by squeezing renal cortex through a mesh, and second an enzymatic

digestion of minced pieces of renal cortex. In our hands, the latter one was suitable to obtain

short-term cultures, which are highly enriched for podocytes. This observation was

supported by the following evidence: (i) glomeruli fractions displayed a lower contamination

with tubuli, (ii) Bowman’s capsules were more efficiently removed, causing a strong increase

in the relative abundance of podocin transcripts in quantitative RT-PCR and (iii) an efficient

enrichment of podocin as seen in Western blot analyses (Fig. 7A,B). Finally,

immunofluorescence staining of glomeruli with antibodies raised against the podocyte-

specific marker podocin and the mesangial-specific marker α-SMA also confirmed the

existence of podocytes and a well preserved structure of decapsulated rat or mouse

glomeruli (Fig. 7C-E). By culturing them, we obtained podocin-positive podocyte cultures that

exhibited the typical cobblestoned and, later on, a beginning arborized morphology (Fig. 7F-

H). A few mesangial cells also grew out of glomeruli, but they show a more elongated

morphology and were positive for α-SMA staining (Fig. 7I-K).

2+

Ca signalling in isolated rat and mouse podocytes and glomeruli

In short-term cultures of enzymatically isolated isolated rat and mouse podocytes, activation

2+

of P2Y, PAR1 or bradykinin receptors yielded the most robust [Ca ] signals (Fig. 8A-E). As

i

described earlier, functional expression of AT receptors is rapidly lost in podocytes during

1

the culturing. Accordingly, only in 2 out of 9 experiments, a fraction of rat podocytes (up to

2+

3

5%) still responded to 10 µM angiotensin II with a small Ca mobilisation signal. In mouse

2+

podocytes, no increases in [Ca ] were elicited by angiotensin II. During acute exposure to

i

2

+

1

00 µM OAG, we did not observe a statistically significant increase in [Ca ] in rat or mouse

i

podocytes (see Fig. 8C-E). As expected, preincubation of mouse podocytes with

2+

thapsigargin induced a transient increase in [Ca ], but a subsequent challenge with 100 µM

i

ATP, 10 µM bradykinin or 0.5 U/ml thrombin failed to cause a significant second increase in

2+

[

Ca ] (Supplemental Figure 1), which would be expected if cells were expressing

i

substantial amounts of TRPC6. Likewise, the acute application of 5 µM LMA did not cause

2+

rapid decline of the [Ca ] in ATP-, bradykinin- or thrombin-stimulated mouse podocyte

i

2+

cultures (Fig. 8G) and even if 5 µM LC was preincubated, the GPCR-induced [Ca ] peaks

i

did not diminish (Fig. 8F). The recently disclosed compound GSK1702934A is currently the

most potent and effective direct activator of TRPC3 and TRPC6 channels. In multiwell

2+

calcium measurements, we obtained an EC₅₀of 1.8 µM for GSK1702934A to activate Ca

influx in stably transfected HEK_h_T_R_P_C₆_-_Y_F_Pcells whereas the EC of OAG and flufenamic acid

50

(

FFA) were 21.4 µM and over 200 µM (Supplemental Figure 2). Moreover, FFA seemed to

provoke unspecific effects on parental HEK293 cells at high concentrations (see

2+

Supplemental Fig. 2A). Unfortunately, also 20 µM GSK1702934A failed to elicit [Ca ]

i

signals in rat and mouse podocytes like demonstrated in PASMC from Wistar rats. By

contrast, in PASMC prepared from the same animals, GSK1702934A induced a robust

2+

[

Ca ] signal (see Supplemental Fig. 2C-F), demonstrating that the compound indeed

i

activates native TRPC6- and/or TRPC3-bearing channel complexes.

Since TRPC6 expression may be quickly down-regulated after the cell isolation, we also

2+

sought to observe TRPC6-related [Ca ] signals in acutely dissociated mouse or rat

i

glomeruli. Rat glomeruli robustly responded upon stimulation with 100 µM carbachol (Fig.

2+

9

A). However, acute addition of 5 µM LMA during the plateau phase of [Ca ] elevation did

i

2+

not significantly lower the [Ca ] (Fig. 9B). Similar results were obtained with glomeruli

i

isolated from transgenic B6.Cg-Tg(NPHS2-Trpc6) F419Walz/J mice (Fig. 9C,D). In glomeruli

from these mice, pretreatment with 5 µM LC also failed to exert a discernible impact on the

2+

size or on the kinetic properties of angiotensin II (10 µM)-induced [Ca ] signals (Fig. 9E-G).

i

To exclude that our isolation procedure with collagenase may cause the lack of a discernible

TRPC6 activation, we also used the mechanical squeezing method. Regardless of the

2+

isolation method, the TRPC6 activator GSK1702934A did not cause a discernible Ca influx

in any cell type of freshly isolated rat or mouse glomeruli. By contrast, stimulation with

carbachol or with angiotensin II was still effective (see Supplemental Fig. 2E-F), indicating

that small peptides and chemical compounds can access the cells. Increasing the driving

force for calcium entry through TRPC6 by elevating the CaCl concentration in the buffer

2

from 1 mM to 2 mM had no effect. Since LMA-sensitive TRPC6-like responses were

previously seen in short-term cultures of rat PASMC (Urban et al., 2016), which also

responded to stimulation with the TRPC6 activator GSK1702934A (Supplemental Fig. 3C),

the obvious question arose, how TRPC6 expression in podocytes compares to that in

PASMC.

TRPC6 expression in rat and mouse podocyte cultures and glomeruli compared to

PASMC

First, quantitative RT-PCR was applied to analyse the abundance of TRPC6-encoding

transcripts in podocyte cultures obtained from rats, from wild-type mice or from transgenic

mice, selectively overexpressing mouse TRPC6 in podocytes (under the control of a human

podocin promoter/enhancer sequence). PCR products with the expected sizes for

endogenously expressed rat or mouse TRPC6 were found in podocytes as well as in

PASMC isolated from several investigated animals (Fig. 10A,B). The selective detection of

the transgene was positive in the TRPC6-overexpressing transgenic mouse strain, but not in

wild-type mice (Fig. 10C). Quantitative PCR normalised to ribosomal 18s RNA revealed that

in podocytes from 10 days-old or adult rats, TRPC6 mRNA was markedly less abundant

than in rat PASMC (Fig. 10D). Podocin expression was stronger in podocytes isolated from

young animals, compared to podocytes from adult rats or to PASMC, which also showed the

highest relative expression of -SMA (Fig. 10E). In podocytes isolated from wild-type mice,

TRPC6 expression was again lower than in PASMC isolated from the same animals. As

expected, the mRNA abundance of mouse TRPC6 tended to be higher in podocytes from

mice expressing the TRPC6 transgene than in age-matched wild-type mice, but was still

about 6-fold lower than endogenous TRPC6 expression in mouse PASMC (Fig. 10F). The

expression profile of podocin and -SMA in wild-type or transgenic mice was similar to that

observed in adult rats (see Fig. 10G).

At the protein level, demonstration of TRPC6 expression is limited by the quality of the

antibodies. Western blot analyses clearly demonstrated TRPC6 expression in transiently

transfected HEK293 cells, but two additional, unspecific bands were also seen in

untransfected cells, or in control-transfected HEK293 cells expressing TRPC2-HA, TRPC3-

HA or a soluble enhanced yellow fluorescent protein (Fig. 10H-J). Accepting the single

specific band as an intraexperimental size marker for TRPC6, lysates of acutely dissociated

rat or mouse glomeruli, podocyte or PASMC cultures were judged. Specific signals were

found in rat PASMC cultures, but absent or markedly less intense in lysates from glomeruli

(

Fig. 10H,I) or podocyte cultures (Fig. 10J). Controls for the tissue and cell preparations

were performed by probing the membranes with anti-podocin, and anti--SMA antibodies. In

addition, a loading control was included (anti-GAPDH) in the experiments. From these data,

we conclude that both mRNA and protein abundance of TRPC6 are highly dissimilar

2+

between PASMC and podocytes or glomeruli. Thus, the lack of TRPC6-like global [Ca ]

i

signals may be explained by the limited sensitivity of fluorometric analyses under conditions

of low channel expression, which was not overcome by the moderate, podocyte-specific

TRPC6 overexpression in a transgenic B6.Cg-Tg(NPHS2-Trpc6)F419Walz/J mouse line.

Ionic currents in rat and mouse podocytes

In electrophysiological recordings, the sensitivity of detecting minute TRPC6 currents may

2+

be higher than in fluorometric [Ca ] analyses. We therefore attempted to assess the

i

sensitivity of native TRPC6 to inhibition by LMA and LC by measuring whole-cell currents in

primary cultures of mouse and rat podocytes. Applying a voltage ramp protocol, an increase

in ionic currents could be elicited by stimulation with OAG in 10 out of 34 patched rat

podocytes. Figure 9A depicts current densities in one of these cells, more examples are

shown in Supplementary Figure 1. The reversal potential was close to 0 mV, indicating a

non-selective conductance. Leak-subtracted increases in current densities upon stimulation

with 50 µM OAG were 0.648 ± 0.126 pA/pF and 1.267 ± 0.359 pA/pF, measured at -100 mV

and + 100 mV, respectively. In these podocytes, the acute addition of 5 µM LMA (see

Supplementary Fig. 1A) or of 2.5 µM LC (Fig. 11A and Supplementary Fig. 3B-D) caused a

rapid decline of inward currents at -100 mV by 1.069 ± 0.321 pA/pF, and of outward currents

at +100 mV of 2.161 ± 0.746 pA/pF at +100 mV. The LMA- or LC-sensitive current

components again displayed a reversal potential close to 0 mV and outwardly rectifying

properties, which is compatible with TRPC6 being the modulated channel entity.

In podocytes cultures obtained from TPRC6-overexpressing transgenic mice, similar results

were obtained (Fig. 11B and Supplementary Fig. 3E-H). Here, OAG induced increases in

current densities in 20 out of 45 patched cells. Current density increases over basal currents

amounted 0.563 ± 0.162 pA/pF at -100 mV, and 1.126 ± 0.213 pA/pF at +100 mV. The acute

addition of LMA or LC caused decreases in inward and outward current densities by

0.717 ± 0.173 pA/pF and 1.753 ± 0.412 pA/pF, measured at -100 mV and +100 mV,

respectively. Like in rat podocytes, all mouse podocytes that displayed a discernible

increase in current densities when challenged with OAG also responded to the addition of

LMA or LC with a current inhibition. From these data we conclude that TRPC6-like

conductances in primary rodent podocyte cultures are small, but sensitive to the

semisynthetic larixol derivatives. For comparison, the OAG-activated non-selective ion

current densities measured under the same conditions in rat PASMC were -2.46 ± 0.93 and

21.27 ± 5.23 pA/pF at -100 mV and +100 mV, respectively (data not shown).

4. Discussion

Within canonical TRPC channels, gain of function mutations of TRPC6 in familiar forms of

FSGS represent the best characterised channelopathy, which leads to an as yet untreatable

focal segmental glomerulosclerosis, culminating in end-stage kidney disease. Adding to this

still unmet medical need, lung diseases, such as pulmonary hypertension, allergic airway

hyperresponsiveness, and ischemia/reperfusion-induced lung oedema, are potential fields of

application for inhibitory pharmacological modulators of TRPC6 activity (Bon & Beech,

2013;Mene et al., 2013). We here propose larixyl carbamate as a semisynthetic larixol

derivative, which is easily available, highly efficient and potent in suppressing TRPC6-

2+

mediated Ca influx and ionic currents in both heterologous expression systems as well as

in native cell models of TRPC6 activity. The biological activity of LC was reassessed and

approved in various mutated TRPC6 constructs, carrying FSGS-related nonsynonymous

substitutions. During the process of validating the blocker efficiency in native cells, we faced

the problem that native TRPC6-like responses were hardly detectable in primary podocyte

cultures or in freshly isolated glomeruli from rats or mice, including a transgenic mouse

strain, which moderately overexpresses TRPC6 under the control of a podocyte-specific

promoter (Krall et al., 2010). To elucidate the apparent discrepancy, we assessed the mRNA

and protein abundance in podocytes and glomeruli compared to pulmonary artery smooth

muscle cells. The about 20- to 50-fold lower abundance of TRPC6 mRNA in podocytes and

glomeruli provided an explanation for the lack of strong TRPC6-like signals in fluorometric

2+

[

Ca ] assays. In electrophysiological patch clamp experiments, the expected low current

i

densities were seen, and the efficiency of larixyl carbamate to inhibit these currents could be

demonstrated. Thus, despite its weak expression, up-regulated wild-type TRPC6 or mutated

TRPC6 variants seem to play seminal roles in causing podocyte injury in diseased states.

We expect that more selective and validated TRPC6 inhibitors will be instrumental to explore

the therapeutic potential of pharmacological interference with this target.

Our development of new TRPC6 inhibitors originated from larixyl monoacetate (Urban et al.,

2016), and revealed larixyl carbamate as a highly potent derivative, which is biologically

active with respect to native TRPC6 as well as to six human TRPC6 variants carrying

mutations known to cause FSGS. In comparison to some other TRPC6 inhibitors like

norgestimate (Miehe et al., 2012) or compound 8009-5364 (Urban et al., 2012), the

selectivity of LMA or LC to inhibit TRPC6 but not their closest related channels TRPC3 and

TRPC7 is considerably higher. The highly potent TRPC6 inhibitor (Maier et al., 2015)

displays an even higher potency and selectivity than LC, but its efficiency and potency to

inhibit FSGS-related TRPC6 mutations has not been confirmed so far. Experiments with

2+

single cell [Ca ] measurements and electrophysiological trials demonstrated that high

i

nanomolar or low micromolar concentrations of either LMA or LC are sufficient to inhibit the

activity of FSGS-related TRPC6 channel constructs to levels of wild-type TRPC6 or below.

Since TRPC6-associated FSGS is an autosomal dominant disease in which a wild-type

allele coexists, and since mutations are located in intracellular domains that may affect the

channel assembly, we explored if mutated TRPC6 channels still maintain their ability to form

channel complexes together with the wildtype subunits. FRET analyses clearly

demonstrated that the investigated point mutations do not grossly affect the oligomerisation

properties of the channel. Thus, although some mutated TRPC6 variants seemed to require

slightly higher LMA or LC concentrations for half-maximal inhibition, a heteromeric complex

formed by the gene products of the mutated and wild-type alleles may further reduce these

differences. Sub-micromolar IC₅₀values provide a promising starting point for a more

detailed investigation in preclinical disease models examinations and experimental therapies

that critically involve TRPC6 activity. Featuring a carbamyl group rather than an ester-

bonded acetate, LC may resist intracellular esterases. Of note, the pharmacophore of LC is

chemically unrelated to that of SAR7334. Thus, observing similar effects with both inhibitors

would strengthen the evidence for an involvement of TRPC6 in biological activities that are

modulated by either drug.

A medical need for TRPC6 inhibitors is best exemplified by hereditary forms of FSGS that

involve gain of function mutations in TRPC6. We therefore aimed at characterising six of the

previously described TRPC6 mutations that have been reported to cause hyperactive

channel complexes (Reiser et al., 2005;Winn et al., 2005;Heeringa et al., 2009;Hofstra et al.,

2013;Zhu et al., 2009). Our findings confirmed the phenotype of the P112Q-, M132T-,

R175Q- and Q889K-variants that displayed significantly higher basal activity and OAG-

2+

induced peaks in single-cell [Ca ] measurements, and larger current densities in

i

electrophysiological patch-clamp experiments. Interestingly, the apparent activity of the

2+

R895C variant showed a discrepancy between applied experimental methods. Basal [Ca ]

i

and the OAG-induced peak currents in electrophysiological experiments were significantly

increased compared to wild-type TRPC6, which is compatible with previous reports (Reiser

et al., 2005;Schlöndorff et al., 2009;Hofstra et al., 2013). However, we did not observe

2+

significantly elevated [Ca ] values after OAG addition in single cell [Ca ] measurements. In

i

our experiments, the E897K mutation displayed no significant hyperactivity in either method.

In carbachol-stimulated HEK293-M1 cells that stably express the M1 muscarinic receptor,

Reiser et al. reported higher current densities for E897K-mutated TRPC6 (15924139).

Applying the same construct, Schlöndorff et al. have described a trend to lower basal current

2+

amplitudes, and an elevated basal [Ca ] was only seen in a subset of HEK293-M1 cells that

i

express the R895C or E897K variants (Schlöndorff et al., 2009). It is known that the

receptor-activated signals are significantly larger than those after direct TRPC6 activation

with the membrane-permeable DAG mimic OAG (Estacion et al., 2004). Hence, the different

modes of TRPC6 activation and differences in the cellular background may explain the

2+

differences. Since LC efficiently inhibits Ca entry and ionic currents through the E897K-

mutated TRPC6, its use would still be justified e.g. in a preclinical disease model given by a

transgenic mouse strain that overexpress the E897K variant and develops FSGS-like

glomerular lesions (Krall et al., 2010).

2+

In podocytes, TRPC6 is the predominant Ca -permeable channel that promotes stress fibre

formation via RhoA activity and thereby mediates the contractile phenotype in these cells

(

Tian et al., 2010). Furthermore, TRPC5 has been implicated to be involved in podocyte

migration by increasing Rac1 activation and a loss of stress fibres (Tian et al.,

010;Schaldecker et al., 2013). Under physiological conditions, both TRPC6 and TRPC5 are

2

activated by angiotensin II downstream of the AT receptor (Tian et al., 2010;Eckel et al.,

1

2011), whereupon DAG generated from phosphatidylinositol-4,5-bisphosphate by PLC

directly and activates TRPC6 (Hofmann et al., 1999;Sonneveld et al., 2014). Performing

2+

microfluormetric [Ca ] imaging experiments with podocytes, we therefore initially aimed at

i

activating endogenous TRPC6 channels with the membrane-permeable DAG analog OAG.

2+

Surprisingly, we could not detect a significant increase in [Ca ] in primary podocyte cultures

i

2+

from rat and mice. Applying angiotensin II instead of OAG, we found that the [Ca ] signals

i

were inconsistent and not significantly impaired by preincubation with LMA or LC. To

activate other GPCRs that are known to be expressed in podocytes (Nitschke et al.,

2001;Madhusudhan et al., 2012;Ilatovskaya et al., 2013;Roshanravan & Dryer, 2014), we

also tested ATP, bradykinin, carbachol or thrombin as activators, but TRPC6-like activities

2+

remained undetectable in microfluorometric [Ca ] analyses. It is known that podocytes

i

rapidly de-differentiate under cell culture conditions. This process is characterised by a loss

of foot processes and of podocyte-specific proteins, such as synaptopodin, podocin or the

2+

AT receptor (Shankland et al., 2007). Hence, a small and inconsistent [Ca ] signal upon

1

i

addition of angiotensin II was not unexpected. Other groups have faced the same problem

and, to circumvent it, sometimes re-introduced the AT receptor by transient transfection

1

(

Tian et al., 2010). The lack of a discernible TRPC6 activation by OAG nonetheless intrigued

us, because endogenously expressed TRPC6 is functionally detectable in freshly isolated

PASMCs (Weissmann et al., 2006;Fuchs et al., 2011;Urban et al., 2016). We also obtained

the transgenic mouse strain which overexpresses TRPC6 in podocytes and develops an

2+

FSGS-like phenotype (Krall et al., 2010). Nevertheless, a TRPC6-like Ca influx was also

not detectable in isolated podocytes from this mouse strain, applying OAG in fluorometric

assays. In line with the author’s description of a 2-fold elevation of TRPC6 mRNA levels in

glomeruli from this mouse, and of low abundance of the TRPC6 protein (Krall et al., 2010), a

mild podocyte-specific increase in TRPC6 expression may still not suffice to augment

2+

TRPC6-like Ca influx signal above detection thresholds. Of note, all studies that

2+

demonstrate a robust OAG-mediated Ca influx in podocyte cultures used immortalised

mouse or human podocyte cell lines rather than primary cultures (Möller et al.,

2007;Nijenhuis et al., 2011;Yang et al., 2013;Sonneveld et al., 2014;Ambrus et al., 2015).

Cultured podocytes lose their cell microarchitecture, receptors or ion channels in vitro. We

therefore next focused on freshly isolated intact glomeruli. Here, carbachol- and angiotensin

2+

II-induced increases in [Ca ] have been reported for rat and mouse glomeruli (Nitschke et

i

al., 2000;Nitschke et al., 2001). Our data are in good agreement with these findings.

2+

However, the recently published inhibition of angiotensin II-induced [Ca ] signals by SKF-

i

96365 (Ilatovskaya et al., 2014) was not reproducible with the TRPC6 inhibitors LMA or LC.

In our hands, 1 µM SKF-96365 is not sufficiently effective to block recombinant TRPC6, and

single channel properties of TRPC6 are characterised by markedly shorter open times, very

low spontaneous activity, an almost instantaneous opening after GPCR stimulation, a more

rapid deactivation, a voltage-dependence of their open probability, and a larger unitary

conductance (Hofmann et al., 1999, Urban et al. 2016). In rat glomeruli, the more long-

2+

lasting carbachol-induced [Ca ] signal enabled an acute application of LMA, but again the

i

signal did not seem to critically depend on TRPC6. Finally, our analyses of TRPC6 mRNA

and protein in freshly isolated glomeruli and in cultured podocytes provided an explanation

2+

for the absence of robust TRPC6-related [Ca ] signals. The about 20 to 50-fold lower

i

abundance of TRPC6 mRNA in podocytes compared to PASMCs matched with

electrophysiological findings. OAG-induced current densities in PASMCs were about 20-fold

higher than in podocytes, which only reached about 1 pA/pF. Interestingly, few studies

utilizing freshly isolated podocytes instead of the immortalized cell lines confirmed these very

small current amplitudes upon stimulation with angiotensin II (Eckel et al., 2011;Kalwa et al.,

2015).

The partial inhibition of TRPC6 activity by low concentrations LMA or LC may be exploited to

design experimental therapies in pulmonary and renal diseases. For example, about 0.5-2

µM LC suppressed ionic currents through the most strongly overactive TRPC6 mutations

(

P112Q, M132T, and R175Q) to a level that corresponds to those of wild-type TRPC6. In

pulmonal diseases or to prevent ischemia-reperfusion-induced pulmonary oedema e.g. in a

scenario of lung or combined heart-lung transplantation, a beneficial effect of TRPC6

inhibition still needs to be demonstrated. Consequently, the required systemic plasma

concentrations or tissue concentrations of TRPC6 inhibitors are not yet established. In the

lung or in organ transplantation, however, topical application or addition to the preservation

fluid may limit systemic effects of TRPC6 inhibitors and thereby increase the tolerability and

safety of use. In familial forms of FSGS, a lifelong application of TRPC6 inhibitors e.g. in

carriers of TRPC6 mutations may be considered to prevent or at least delay the onset of

nephrotic syndrome and requirement of kidney transplantation. On the other hand, since

TRPC6 may also exert protective functions in complement-dependent forms of glomerular

injury (Kistler et al., 2013), the risk to benefit ratio of pharmacological TRPC6 inhibition

needs to be critically evaluated. Appropriate animal models of human kidney or lung

diseases together with validated new TRPC6 inhibitors, such as LC or aminoindanes (Maier

et al., 2015) now provide a rational basis for the validation of TRPC6 as a pharmacological

target by designing preclinical models, assessment of the safety profile, and by testing their

efficacy and tolerability in experimental therapies.

Conflict of interest

The authors declare no conflicts of interest.

Acknowledgments

We thank Anne-Kathrin Krause for expert technical assistance and animal care and Thomas

Rudolph for technical support of the chemical synthesis and analysis. This work was

supported by the Deutsche Forschungsgemeinschaft (TRR 152 to M.S. and TRR 67 to J.R.).

Article Doi

Pharmacological inhibition of focal segmental glomerulosclerosis-related, gain of function mutants of TRPC6 channels by semi-synthetic derivatives of larixol

DOI: 10.1111/bph.13977

Source and publish data:

Authors:

Article abstract of DOI:10.1111/bph.13977

Full text of DOI:10.1111/bph.13977

Products guided by the article

R&D Labs maybe for 1226871-01-7

Relevant to this article

Hot Product