Derivatization of common antidepressant drugs increases inhibition of acid sphingomyelinase and reduces induction of phospholipidosis

Article abstract of DOI:10.1007/s00702-018-1923-z

In recent studies, major depressive disorder (MDD) was linked to an increase in acid sphingomyelinase (ASM) activity. Several drugs that are commonly used to treat MDD functionally inhibit the lysosomal enzyme ASM and are called functional inhibitors of ASM (FIASMAs). These drugs are classified as cationic amphiphilic drugs (CADs) that influence the catalytic activities of different lysosomal enzymes. This action results in the side effect of phospholipidosis (PLD), which describes a detrimental increase in the phospholipid content in lysosomes. FIASMAs differ only slightly in their physico-chemical properties, but their effects on ASM activity and induction of the lysosomal phospholipid content vary significantly. In this study, we systematically induced minor chemical modifications to the FIASMAs imipramine, desipramine and fluoxetine. We generated a library of 45 new CADs with slightly different log P (logarithmic partition coefficient) and pKa (logarithmic acid dissociation constant) values. The effects of the compounds on the ASM activity and lysosomal phospholipid content were assessed in cell culture assays. We identified four compounds with beneficial effects, i.e., increased ASM activity inhibition and reduced PLD induction compared with the original drugs. The compounds HT04, RH272B and RH272D outperformed the original imipramine, whereas RH281A performed better than desipramine. Thus, minor chemical variations of CADs impact lysosomal metabolism in a specific manner and can lead to antidepressant drugs with less deleterious side effects.

Full text of DOI:10.1007/s00702-018-1923-z

Journal of Neural Transmission
https://doi.org/10.1007/s00702-018-1923-z
PSYCHIATRY AND PRECLINICAL PSYCHIATRIC STUDIES - ORIGINAL ARTICLE
Derivatization of common antidepressant drugs increases inhibition
of acid sphingomyelinase and reduces induction of phospholipidosis
Cosima Rhein1 · Stefan Löber2 · Peter Gmeiner2 · Erich Gulbins3 · Philipp Tripal1,4 · Johannes Kornhuber1
Received: 4 June 2018 / Accepted: 28 August 2018
© Springer-Verlag GmbH Austria, part of Springer Nature 2018
Abstract
In recent studies, major depressive disorder (MDD) was linked to an increase in acid sphingomyelinase (ASM) activity. Sev-
eral drugs that are commonly used to treat MDD functionally inhibit the lysosomal enzyme ASM and are called functional
inhibitors of ASM (FIASMAs). These drugs are classified as cationic amphiphilic drugs (CADs) that influence the catalytic
activities of different lysosomal enzymes. This action results in the side effect of phospholipidosis (PLD), which describes
a detrimental increase in the phospholipid content in lysosomes. FIASMAs differ only slightly in their physico-chemical
properties, but their effects on ASM activity and induction of the lysosomal phospholipid content vary significantly. In this
study, we systematically induced minor chemical modifications to the FIASMAs imipramine, desipramine and fluoxetine. We
generated a library of 45 new CADs with slightly different log P (logarithmic partition coefficient) and pKa (logarithmic acid
dissociation constant) values. The effects of the compounds on the ASM activity and lysosomal phospholipid content were
assessed in cell culture assays. We identified four compounds with beneficial effects, i.e., increased ASM activity inhibition
and reduced PLD induction compared with the original drugs. The compounds HT04, RH272B and RH272D outperformed
the original imipramine, whereas RH281A performed better than desipramine. Thus, minor chemical variations of CADs
impact lysosomal metabolism in a specific manner and can lead to antidepressant drugs with less deleterious side effects.
Keywords Acid sphingomyelinase · Functional inhibitors of ASM activity (FIASMAs) · Imipramine · Desipramine ·
Fluoxetine · Phospholipidosis · Major depression
Introduction
as a potential target for antidepressant action (Gulbins et al.
2013; Kornhuber et al. 2005; Rhein et al. 2017). ASM is a
mainly lysosomal located enzyme that optimally functions
at a pH of 5. ASM hydrolyzes the abundant membrane lipid
sphingomyelin to ceramide, a bioactive lipid messenger
that affects downstream signaling in the cell (Gulbins and
Grassme 2002; Gulbins and Kolesnick 2002). Interestingly,
several antidepressant drugs indirectly inhibit ASM activ-
ity (Albouz et al. 1986) and are called ‘functional inhibi-
tors of ASM’ (FIASMAs) (Kornhuber et al. 2011, 2010,
2008). Therefore, the role of ASM in the etiology of MDD
was investigated in human and murine studies. Depressed
patients displayed significantly increased levels of ASM
activity in their cultivated blood cells compared with a con-
trol group. In addition, the severity of depression as indi-
cated by the Hamilton Scale of Depression (HAMD) and
ASM activity levels of patients were positively correlated
(Kornhuber et al. 2005). In a genetic mouse model over-
expressing ASM, the ASM/ceramide system mediates the
effects of antidepressant drugs (Gulbins et al. 2013). Thus,
Major depressive disorder (MDD) is a psychiatric disease
with an unresolved etiology. In recent studies, the enzyme
acid sphingomyelinase (ASM, EC 3.1.4.12) was investigated
Philipp Tripal and Johannes Kornhuber contributed equally.
* Cosima Rhein
cosima.rhein@uk-erlangen.de
1
Department of Psychiatry and Psychotherapy, University
Hospital Erlangen, Friedrich-Alexander Universität
Erlangen-Nürnberg (FAU), Erlangen, Germany
2
Pharmaceutical Chemistry, Friedrich-Alexander Universität
Erlangen-Nürnberg (FAU), Erlangen, Germany
3
Department of Molecular Biology, University
of Duisburg-Essen, Essen, Germany
4
Present Address: Optical Imaging Centre Erlangen (OICE),
Friedrich-Alexander Universität Erlangen-Nürnberg (FAU),
Erlangen, Germany
Vol.:(0123456789)
1 3

C. Rhein et al.
strong ASM inhibition should be achieved by a FIASMA to
be beneficial in this context (Kornhuber et al. 2014), without
inducing side effects.
not in a specific or differential manner. But experimental
data showed that the inhibitory potential of FIASMAs on
ASM activity is broad (Kornhuber et al. 2011), and that only
a few of the lysosomal enzymes were affected by FIASMA
application (Kornhuber et al. 2010), whereas other enzymes
retained their function. These data led to the hypothesis that
CADs can indeed exert specific effects on selected lysosomal
enzymes. For an overview of the different lysosomal param-
eters, see Fig. 1. It could be experimentally demonstrated
that the FIASMAs imipramine, desipramine, and fluoxetine
exert differential patterns of influence on ASM activity
(Kornhuber et al. 2011) and phospholipid accumulation
(Muehlbacher et al. 2012), that point to specific lysosomal
action of CADs with need for optimized parameters in the
context of MDD.
The FIASMAs imipramine, desipramine, and fluoxetine
are among the most prominent antidepressant drugs. For
desipramine, experimental data showed that it accumu-
lates at the lysosomal membrane and causes displacement
of ASM from the membrane (Kölzer et al. 2004). ASM is
released into the lysosomal lumen and subjected to degrada-
tion by lysosomal mechanisms via proteases (Hurwitz et al.
1994). This effect on ASM activity can be measured and
quantified by an enzyme activity assay (Gulbins and Kole-
snick 2000) in cell culture systems (Kornhuber et al. 2011),
clinical studies on cultured human blood cells (Kornhuber
et al. 2005) or murine tissues (Gulbins et al. 2013).
However, the application of FIASMAs also results in the
side effect of phospholipidosis (PLD), which is the accumu-
lation of phospholipids (Anderson and Borlak 2006), and
is probably caused by the degradation of lysosomal phos-
pholipases (Halliwell 1997). This effect occurs in lysosomal
storage disorders, but is also induced by different drugs, and
then called drug-induced phospholipidosis (Lüllmann et al.
1978). Affected organs develop inflammatory reactions and
histopathological changes (Anderson and Borlak 2006).
Phospholipid content in cells can be monitored using spe-
cific staining (Muehlbacher et al. 2012).
Therefore, we aimed to generate modified derivatives of
imipramine, desipramine, and fluoxetine that should increase
ASM activity inhibition and reduce PLD induction com-
pared with the original drugs. The systematic modification
of imipramine, desipramine, and fluoxetine resulted in the
generation of a library of 45 new CADs. All substances
were synthesized and characterized. In addition, the effects
of the compounds on ASM activity and PLD induction were
assessed in cell culture systems. The results demonstrate
that minor chemical modifications changed the properties
regarding ASM activity and PLD induction for some of the
compounds significantly. This approach will help to define
specific actions exerted by CADs and could lead to improved
antidepressants with a reduced spectrum of side effects.
Antidepressant drug development thus aims to generate
potent ASM inhibitors with a low potential of inducing PLD.
Relevant for this approach are those factors that influence
lysosomal targeting of compounds. Cationic amphiphilic
drugs (CADs) are known for their optimal lysosomotro-
pic characteristics. FIASMAs are a subgroup of CADs and
also feature high lysosomal accumulation as predicted by a
chemo-theoretic approach (Trapp et al. 2008).
Experimental section
Synthesis of imipramine derivatives
CADs enter the lysosomal cell compartment by a mecha-
nism known as lysosomotropism or lysosomal trapping (de
Duve et al. 1974; Trapp et al. 2008). The efficiency of this
mechanism is highly dependent on the chemical properties
of the respective compound (Kornhuber et al. 2011; Trapp
et al. 2008). The log P value (logarithmic partition coef-
ficient) of a chemical compound is a measure of its lipo-
philicity, and the pKa value (logarithmic acid dissociation
constant) describes its ability to become protonated in a pH-
dependent manner. These two characteristics are the main
discriminators of lysosomal trapping (Trapp et al. 2008).
Lysosomotropic compounds are rather lipophilic (log P
between 2 and 9) with a pKa value > 4; thus, these com-
pounds are protonated at a low pH level.
Test compounds RH272B, RH272D, RKZ01, RKZ06,
RKZ07 and HT04 were obtained by reductive alkylation
of desipramine (free base) using 5 equivalents of propanal,
pentanal, hexanal, cyclohexanol, acetone or 6-hydroxyhexa-
nal, respectively. Na(OAc)3BH (3 eq.) served as the reducing
agent, and CH2Cl2 served as a solvent. After stirring the mix-
ture at room temperature for 16 h, the solution was treated
with brine and extracted with CH2Cl2. After evaporation, the
crude product was purified by flash chromatography.
For the synthesis of the derivative HT23
5-(4-bromobutyl)-10,11-dihydro-5H-dibenz[b,f]azepine
(Schmidt et al. 2008) was treated with 1-hexylamine (5 eq.),
Na2CO3 (1 eq.) and NaI (0.2 eq.) and stirred at reflux temper-
ature in acetonitrile for 16 h. The cooled mixture was treated
with brine and extracted with CH2Cl2. After evaporation, the
crude product was purified by flash chromatography.
Therefore, CADs were supposed to influence the lyso-
somal metabolism by unspecific, functional mechanisms
(Hurwitz et al. 1994; Kölzer et al. 2004; Kornhuber et al.
2010) that affect lysosomal hydrolases like ASM and phos-
pholipases in general (Kodavanti and Mehendale 1990), but
The amide derivatives LP05 and HT13 were synthe-
sized by acylation of desipramine (free base) with 6-bromo
1 3

Derivatization of common antidepressant drugs increases inhibition of acid sphingomyelinase…
Fig. 1 Overview of relevant lysosomal parameters. The FIASMAs
imipramine, desipramine, and fluoxetine show values for lipophilicity
(log P) and protonation (pKa) that allow efficient trapping in the lyso-
some. The inner surface charge of the lysosomal membrane is nega-
tive (Gallala and Sandhoff 2011; Kolter and Sandhoff 2010) due to
the high abundance of bis-(monoacylglycero)-phosphate (Kobayashi
et al. 1998; Wilkening et al. 1998). ASM is positively charged at lyso-
somal pH and binds to lysosomal membrane structures (Kölzer et al.
2004). The drugs enter the lysosome, get protonated, accumulate in
the lysosomal membrane and cause the displacement of ASM from
the membrane (Arrow 1). ASM is released into the lysosomal lumen
and subjected to degradation by lysosomal mechanisms via proteases
(Arrow 2) (Hurwitz et al. 1994). This indirect inhibition of ASM
activity can be measured by an enzyme activity assay (“Assay 1”).
The degradation of ASM by FIASMAs results in the accumulation
of phospholipids (“Phospholipidosis”; arrow 3) that can be measured
using LipidTOX (“Assay 2”). FIASMAs are supposed to influence
lysosomal enzymes other than ASM, e.g., phospholipases (“Enzyme
A”; arrow 4), that leads to further phospholipid accumulation (Arrow
5). In contrast, other lysosomal enzymes might be unaffected by
FIASMAs (“Enzyme B”). Basic motifs from © motifolio.com
hexanoic acid chloride (1.5 eq) in the presence of trieth-
ylamine (2 eq.) in CH2Cl2 at room temperature for 2 h.
The resulting bromo hexanoylamide was purified by flash
chromatography and reacted in DMF at 50 °C for 20 h with
dimethylamine or piperidine (4 eq.), respectively. The prod-
ucts were purified by flash chromatography.
The piperazine derivative RKZ11 was derived from
5-(3-chloropropyl)-10,11-dihydro-5H-dibenzo[b,f]azepine
1 3

C. Rhein et al.
(Dollinger et al. 2006) by reaction with N-methyl pipera-
zine following the procedure described for the synthesis of
HT23.
Germany) supplemented with 10% (v/v) fetal bovine serum
(FBS) and 4 mM glutamine (Biochrom, Berlin, Germany).
The cell line was maintained at 37 °C in a humidified atmos-
phere containing 8.5% CO2 and was routinely split at a ratio
of 1:6. The cells were regularly tested for mycoplasma con-
tamination by an ELISA-based assay (Lonza, Basel, Swit-
zerland), and the tests were always negative.
Synthesis of desipramine derivatives
Test compounds RH281A, HT21 and HT25 were obtained
by nucleophilic substitution of 5-(4-bromobutyl)-10,11-di-
hydro-5H-dibenz[b,f]azepine (Dollinger et al. 2006) or
5-(4-bromobutyl)-10,11-dihydro-5H-dibenz[b,f]azepine
(Schmidt et al. 2008), respectively, with n-propylamine,
methylamine or n-hexylamine following the procedure
described for the synthesis of HT23.
Acid sphingomyelinase activity assay
The activity of ASM was determined in whole cell lysates,
as previously described (Gulbins and Kolesnick 2000). After
the substance was added to the growth medium at a final
concentration of 10 µM, cells were kept at 37 °C in a humid-
ified atmosphere at 8.5% CO2 for 30 min. Results are given
as residual ASM activity (%) normalized to control cells
treated with the solvent alone and represent mean values of
Synthesis of fluoxetine derivatives
The fluoxetine derivatives HT30, HT31, LP21, LP23 and
LP24 were synthesized according to the depicted scheme.
Mannich reaction of acetophenone with the hydrochloride salts
of dimethylamine or N-methylhexylamine in ethanol at reflux
temperature (2.5 h) yielded the respective β-aminoketones.
Reduction of the ketone functionality with NaBH4 in MeOH/
H2O (5:1) at room temperature (16 h) followed by treatment
with thionylchloride in CH2Cl2 for 4 h at reflux temperature
resulted in the respective benzyl chlorides. Finally, reaction
with 3- or 4-trifluoromethyl phenol, 4-hydroxy benzaldehyde
or 4-hydroxy benzonitrile, respectively, in DMF at 80 °C in
the presence of K2CO3 (3 h) yielded the desired fluoxetine
derivatives. After treatment with aqueous NaOH, extraction
with CH2Cl2 and evaporation of the solvent, the crude products
were purified by flash chromatography.
three independent experiments, each with a standard devia-
tion of approx. 16%. A residual ASM activity ≤ 50.0% in
H4-cells was rated as positive.
Phospholipidosis assay
H4 cells were seeded in 96-well white dishes (Nunc, Lan-
genselbold, Germany) at a density of 4×103 cells per well.
After 48 h, the medium was replaced with fresh medium
that included the test substances and HCS LipidTOX Green
phospholipidosis detection reagent (Invitrogen, San Diego,
US) at their respective final concentrations. Each test sub-
stance was diluted from the stock solution with medium
and was applied at 2.5 µM for an additional 24 h. All tests
were performed in quadruplicate. During the test period,
the cells were cultured at 37 °C in a humidified atmosphere
containing 8.5% CO2. To quantify the lysosomal phospho-
lipid content, the cells were washed with phosphate-buff-
ered saline (PBS), counterstained with DAPI (Carl Roth
GmbH, Karlsruhe, Germany) and fixed with 10% neutral
buffered formalin (AppliChem GmbH, Darmstadt, Ger-
many). Fluorescent signals were quantified with a fluores-
cence reader (Perkin Elmer, Waltham, US) at an excita-
tion wavelength of 485 nm and an emission wavelength of
All compounds were dissolved in water or DMSO at a
concentration of 10 mM and stored at −20 °C. For an over-
view of chemical structures of test compounds, see Fig. 2.
Cell culture
Human brain neuroglioma H4 cells were purchased from
Promochem (Wesel, Germany). The cells were cultivated
in Dulbecco’s modified Eagle’s medium (Biochrom, Berlin,
1 3

Derivatization of common antidepressant drugs increases inhibition of acid sphingomyelinase…
Fig. 2 Chemical structures of
imipramine, desipramine and
fluoxetine derivatives
535 nm for the lysosomal dye (LipidTOX Green) and an
excitation wavelength of 355 nm and an emission wave-
length of 470 nm for the nuclear dye (DAPI). The results
were corrected by subtraction of the background and are
reported as percent fluorescence of the corresponding con-
trol averaged over four experiments. A chemical-induced
reduction in cell number greater than 25% (DAPI fluores-
cence < 75% compared with untreated control) was rated
as a toxic effect and prompted the exclusion of the respec-
tive compound from further analysis. Maximal doubling
of the cellular phospholipid content (LipidTox fluores-
cence ≤ 200.0% compared with control) was rated as an
absent PLD side effect and thus a non-toxic result (Muehl-
bacher et al. 2012).
Calculation of pKa and log P values
The pKa and log P values for the substances were calculated
using ACD/Log D Suite (program version 10.02, Toronto,
Canada) as previously described (Kornhuber et al. 2008).
Briefly, the pKa prediction algorithms of the ACD program
are based on a fragmental method using generic substruc-
tures and Hammett type equations to cover the most popular
ionizable functional groups using either a database or the
estimation of electronic substituent constants. The log P
prediction algorithms of the ACD program are based on a
fragmental method with correction factors. Further details
of the prediction algorithms are presented on the ACD web-
page (http://www.whocc.no/atcddd).
1 3

C. Rhein et al.
Statistical analysis
SPSS software (Statistical Package for the Social Sciences,
Chicago, IL, for Windows, Version 21.0) was used for sta-
tistical analysis. Correlations between ASM activity and
PLD induction were calculated using Pearson’s correlation
coefficients. Correlations were considered significant with
p values<0.05 (two-tailed).
Results
A subset of derivatives exhibits beneficial potential
regarding residual ASM activity and PLD induction
A reduction in ASM activity of at least 50%, thus resulting
in the classification of the substance as a FIASMA (Korn-
huber et al. 2008), and a non-toxic PLD induction with less
than 200% induction compared with control (Muehlbacher
et al. 2012) comprise the beneficial range for substances in
our study. Of the 45 compounds generated, 18 compounds
exhibited effects in a beneficial range after application in
our cell culture model (Fig. 3). Of note, a significant nega-
tive correlation (r = − 0.5, p < 0.05) between ASM inhibi-
tion and PLD induction was observed. This finding indi-
cates that increased functional inhibition of ASM activity
correlates with increased PLD induction.
Fig. 3 Derivatives of imipramine, desipramine, and fluoxetine with
beneficial effects regarding the residual ASM activity and PLD induc-
tion (n=18). H4 cells were treated with 10 µM of antidepressant
drug derivatives for 30 min, harvested and subjected to ASM activity
assays. In parallel, cells were treated with 2.5 µM lipidTOX for 24 h,
stained with DAPI and analyzed for PLD induction. The original drug
values are indicated in red. Only compounds that exhibit ASM inhibi-
tion≥50% and PLD induction≤200% are presented. Both parameters
are significantly negatively correlated (r = − 0.5; p<0.05)
Table 1 Derivatives with
Derivative of
Compound
Residual ASM activ- PLD induction (% log P
pKa
Molar
mass (g/
mol)
reduced residual ASM
ity (% control SD)
control SD)
activity or/and decreased PLD
induction levels compared
with the original drugs, and
corresponding log P, pKa and
molar mass values
–
Imipramine
HT04
RH272B
RH272D
RKZ-11
LP05
RKZ-7
HT13
RKZ-6
RKZ-1
HT23
Desipramine
RH281A
HT21
HT25
Fluoxetine
HT31
48.1 10.6
26.8 6.1
33.8 8.2
28.4 12.6
41.6 10.9
6.2 2.0
115.0 5.5
101.3 8.3
110.5 8.1
112.1 3.9
143.5 3.6
132.8 7.4
142.1 7.5
197.7 16.0
182.6 6.7
174.6 14.9
163.0 3.6
123.6 7.3
110.4 2.1
146.0 2.9
180.3 8.6
141.3 15.7
113.5 9.3
106.5 2.4
110.9 3.3
104.2 4.4
103.0 1.9
4.8
5.5
5.9
7.5
3.8
4.6
5.7
5.8
6.6
6.9
7.5
4.1
5.2
4.4
7.0
4.1
4.8
5.0
6.0
6.1
7.7
9.5
9.5
9.6
9.7
8.2
9.7
9.7
9.9
5.4
280.4
366.5
308.5
350.5
335.5
407.6
308.5
447.7
342.5
336.5
364.6
266.4
294.4
280.4
350.5
309.3
323.4
323.4
353.5
350.5
393.5
Imipramine
32.7 5.4
6.9 0.6
25.8 8.1
30.3 12.6
13.0 3.2
25.1 10.9
17.8 4.9
3.3 0.5
9.6
9.9
–
10.4
10.6
10.6
11.0
10.1
9.3
9.2
9.5
9.5
9.5
Desipramine
7.5 1.6
–
13.0 2.8
30.0 4.4
24.9 4.1
47.3 3.6
29.4 2.9
47.9 6.5
Fluoxetine
HT30
LP23
LP24
LP21
1 3

Derivatization of common antidepressant drugs increases inhibition of acid sphingomyelinase…
drug (Table 1). Compound RH281A exhibited increased
beneficial effects for both parameters: an approximately 10%
reduced PLD risk compared with desipramine, and ASM
activity inhibition was increased by approximately 10%
compared with the original drug (Fig. 4). This compound
features a propyl residue that results in a similar pKa value
but in an increased log P value compared with desipramine
(Fig. 2; Table 1).
Fluoxetine derivatives with increased beneficial
potential compared with fluoxetine
None of the fluoxetine derivatives exhibited enhanced
ASM activity inhibition compared with fluoxetine. Five
compounds exhibited approximately 20–40% reduced PLD
induction compared with fluoxetine. However, the efficacy
of PLD induction was achieved at the expense of residual
ASM activity levels that increased disproportionally by 3- to
4-fold compared with fluoxetine treatment (Table 1).
Fig. 4 Derivatives with increased beneficial effects compared with
their original drugs. H4 cells were treated with 10 µM of the respec-
tive substance for 30 min, harvested and subjected to ASM activ-
ity assays. In parallel, cells were treated with 2.5 µM lipidTOX for
24 h, stained with DAPI and analyzed for PLD induction. Only those
derivatives that exhibit both reduced residual ASM activity levels and
reduced PLD induction compared with their original drug are pre-
sented. The results are presented as the residual ASM activity (%)
normalized to control cells representing the mean values of three inde-
pendent experiments with the respective standard deviations or PLD
induction (%) normalized to control cells representing the mean values
of quadruplicate measurements with the respective standard deviations
Discussion
In this study, we demonstrate that small changes in com-
pound structure can markedly change the properties of
the indicated drugs. Compared with their original drugs,
derivatives exhibited varied potential to inhibit ASM activ-
ity and PLD induction. As expected, according to earlier
studies on FIASMAs and PLD (Muehlbacher et al. 2012),
strong functional inhibition of ASM correlated with high
levels of PLD induction. Drug-induced PLD is a side effect
of CADs, especially with regard to the common practice of
polypharmacy in the aged population (Glock et al. 2016;
Kukreja et al. 2013). Several lysosomal mechanisms that
cause PLD are currently discussed, including direct or
indirect inhibition of different phospholipases, induction
of non-degradable complexes of lipids and drugs, increased
lipid synthesis, and inhibition of enzyme transport pro-
cesses in the lysosome (Nonoyama and Fukuda 2008).
Interestingly, it was predicted in silico that different CADs
induce PLD via distinct underlying lysosomal mechanisms
(Lowe et al. 2010). This prediction is consistent with our
observations given that some of the derivatives did not
induce high levels of PLD despite their significant effect
on ASM activity. These chemical compounds indicate
the presence of a PLD-inducing mechanism independent
of ASM biology. Thus, our data demonstrate that CADs
indeed seem to exert effects on specific lysosomal enzymes.
Regarding the commonly used antidepressant fluoxetine,
none of the derivatives yielded significantly better results
for ASM inhibition and PLD induction given that both cel-
lular mechanisms vary among the derivatives. In contrast,
Imipramine derivatives with increased beneficial
potential compared with imipramine
Compared with the original drug imipramine, 10 compounds
exhibited reduced residual ASM activity, a reduced risk of PLD
induction, or both. All 10 compounds exhibited enhanced ASM
inhibition compared with imipramine itself. Three compounds
resulted in less PLD induction compared with imipramine
(Table 1). The compounds HT04, RH272B and RH272D
exhibited increased beneficial effects for both parameters
(Fig. 4). RH272B and RH272D share a tricyclic structure with
imipramine, but have an elongated residual chain (propyl and
hexyl, respectively) that results in a similar pKa value but in
an increased log P value compared with imipramine (Fig. 2;
Table 1). Compound HT04 seems to be a promising compound
given that its application resulted in an approximately 20%
reduced PLD level and it inhibited ASM activity approximately
twice effectively compared with its original drug imipramine. A
(CH2)6OH residue is added to the tricyclic structure, resulting
in a similar pKa value but in an increased log P value compared
with imipramine (Fig. 2; Table 1).
Desipramine derivatives with increased beneficial
potential compared with desipramine
Of the desipramine derivatives, three compounds inhibited
ASM activity more efficiently compared with the original
1 3

C. Rhein et al.
derivatives of the tricyclic antidepressant drugs imipramine
and desipramine exhibited beneficial results. Three imipra-
mine derivatives, HT04, RH272B and RH272D, and one
desipramine derivative, RH281A, exhibit improved values
compared with the original drugs. The imipramine deriva-
tive HT04 seems to be a promising compound as it reduces
PLD by 20% and inhibits ASM twice as efficiently as the
original drug imipramine. This substance could therefore
be an interesting molecule with therapeutic potential.
Given that both ASM inhibition and PLD induction require
proper lysosomal accumulation of the CAD, modifications
that increase the lysosomotropic potential of the respective
compound could account for the different effects. Of note,
similar lysosomal accumulation was demonstrated for all
derivatives by a chemo-informatic prediction algorithm in
silico and thus does not explain those effects. Interestingly,
beneficial derivatives could be identified only for tricyclic
antidepressants. Fluoxetine seems to be the most efficient
structure in our context and does not benefit from modifica-
tion. When assessing the properties of the beneficial tricy-
clic compounds HT04, RH272B, RH272D and RH281A, it
is obvious that these compounds exhibit similar pKa values
compared with their original drugs, and the modifications
do not seem to influence their effects. In contrast, the log P
value was increased for all compounds compared with their
original drugs. HT04 features the longest residue. However,
the increase in log P value is limited to the beneficial range
due to the OH group. Thus, the lipophilic property is a
main factor in our context. Additional explanations for the
beneficial potential of these compounds remain elusive.
Limitations of the study are that the new compounds were
not tested for other side effects and additional properties.
More studies are needed to specify the mechanisms of CAD
action in the lysosome. Our CAD library serves as a valuable
tool for in vitro studies that seek to vary ASM inhibition and
PLD induction independently.
References
Albouz S, Le Saux F, Wenger D, Hauw JJ, Baumann N (1986) Mod-
ifications of sphingomyelin and phosphatidylcholine metabo-
lism by tricyclic antidepressants and phenothiazines. Life Sci
38:357–363
Anderson N, Borlak J (2006) Drug-induced phospholipidosis. Febs
Lett 580:5533–5540
de Duve C, de Barsy T, Poole B, Trouet A, Tulkens P, Van Hoof F
(1974) Commentary. Lysosomotrop Agents Biochem Pharmacol
23:2495–2531
Dollinger S, Löber S, Klingenstein R, Korth C, Gmeiner P (2006)
A chimeric ligand approach leading to potent antiprion active
acridine derivatives: design, synthesis, and biological investi-
gations. J Med Chem 49:6591–6595. https://doi.org/10.1021/
jm060773j
Gallala HD, Sandhoff K (2011) Biological function of the cellular
lipid BMP-BMP as a key activator for cholesterol sorting and
membrane digestion. Neurochem Res 36:1594–1600. https://
doi.org/10.1007/s11064-010-0337-6
Glock M, Muehlbacher M, Hurtig H, Tripal P, Kornhuber J (2016)
Drug-induced phospholipidosis caused by combinations of com-
mon drugs in vitro. Toxicol In Vitro 35:139–148. https://doi.
org/10.1016/j.tiv.2016.05.010
Gulbins E, Grassme H (2002) Ceramide and cell death receptor clus-
tering. Biochim Et Biophys Acta 1585:139–145
Gulbins E, Kolesnick R (2000) Measurement of sphingomyelinase
activity. Methods Enzymol 322:382–388
Gulbins E, Kolesnick R (2002) Acid sphingomyelinase-derived cera-
mide signaling in apoptosis. Subcell Biochem 36:229–244
Gulbins E et al (2013) Acid sphingomyelinase-ceramide system
mediates effects of antidepressant drugs. Nat Med 19:934–938.
https://doi.org/10.1038/nm.3214
Halliwell WH (1997) Cationic amphiphilic drug-induced phospho-
lipidosis. Toxicol Pathol 25:53–60
Hurwitz R, Ferlinz K, Sandhoff K (1994) The tricyclic antidepres-
sant desipramine causes proteolytic degradation of lysosomal
sphingomyelinase in human fibroblasts. Biol Chem Hoppe Sey-
ler 375:447–450
Kobayashi T, Stang E, Fang KS, de Moerloose P, Parton RG, Gru-
enberg J (1998) A lipid associated with the antiphospholipid
syndrome regulates endosome structure and function. Nature
392:193–197. https://doi.org/10.1038/32440
Kodavanti UP, Mehendale HM (1990) Cationic amphiphilic drugs
and phospholipid storage disorder. Pharmacol Rev 42:327–354
Kolter T, Sandhoff K (2010) Lysosomal degradation of membrane
lipids. FEBS Lett 584:1700–1712
Acknowledgements We thank Michaela Henkel and Alice Konrad
for excellent technical assistance. We are grateful to Lydia Petter-
mann and Heike Thomas for generating the chemical compounds. We
thank Christiane Mühle for helpful comments on the manuscript and
preparation of Fig. 3. This work was supported by funding from the
Forschungsstiftung Medizin at the University Hospital Erlangen, the
German Federal Ministry of Education and Research (BMBF, 01
EE1401C, to J. K.), and DFG grants KO 947/13-1 and GU 335/29-1
(to J.K. and E.G.).
Kölzer M, Werth N, Sandhoff K (2004) Interactions of acid sphin-
gomyelinase and lipid bilayers in the presence of the tricyclic
antidepressant desipramine. FEBS Lett 559:96–98
Kornhuber J, Medlin A, Bleich S, Jendrossek V, Henkel AW, Wilt-
fang J, Gulbins E (2005) High activity of acid sphingomyelinase
in major depression. J Neural Transm 112:1583–1590
Kornhuber J, Tripal P, Reichel M, Terfloth L, Bleich S, Wiltfang J,
Gulbins E (2008) Identification of new functional inhibitors of
acid sphingomyelinase using a structure-property-activity rela-
tion model. J Med Chem 51:219–237. https://doi.org/10.1021/
jm070524a
Author contributions PT and JK conceived and designed the experi-
ments; PT and SL designed and performed the experiments and wrote
parts of the paper; CR, JK and EG analyzed the data; SL, PG and JK
contributed reagents/study materials/analysis tools; CR and PT wrote
the paper. All authors read the paper and provided intellectual input.
Kornhuber J et al (2010) Functional Inhibitors of acid sphingomyeli-
nase (FIASMAs): a novel pharmacological group of drugs with
broad clinical applications. Cell Physiol Biochem 26:9–20. https
://doi.org/10.1159/000315101
Kornhuber J et al (2011) Identification of novel functional inhibi-
tors of acid sphingomyelinase. PloS One 6:e23852. https://doi.
org/10.1371/journal.pone.0023852
Compliance with ethical standards
Conflict of interest The authors declare no conflict of interest.
1 3

Derivatization of common antidepressant drugs increases inhibition of acid sphingomyelinase…
Kornhuber J, Müller CP, Becker KA, Reichel M, Gulbins E
(2014) The ceramide system as a novel antidepressant target.
Trends Pharmacol Sci 35:293–304. https://doi.org/10.1016/j.
tips.2014.04.003
Rhein C et al (2017) Alternative splicing of SMPD1 coding for acid
sphingomyelinase in major depression. J Affect Disord 209:10–
15. https://doi.org/10.1016/j.jad.2016.09.019
Schmidt M, Teitge M, Castillo ME, Brandt T, Dobner B, Langner
A (2008) Synthesis and biochemical characterization of new
phenothiazines and related drugs as MDR reversal agents. Arch
Pharm 341:624–638. https://doi.org/10.1002/ardp.200800115
Trapp S, Rosania GR, Horobin RW, Kornhuber J (2008) Quantitative
modeling of selective lysosomal targeting for drug design. EBJ
37:1317–1328. https://doi.org/10.1007/s00249-008-0338-4
Wilkening G, Linke T, Sandhoff K (1998) Lysosomal degradation on
vesicular membrane surfaces. Enhanced glucosylceramide deg-
radation by lysosomal anionic lipids and activators. J Biol Chem
273:30271–30278
Kukreja S, Kalra G, Shah N, Shrivastava A (2013) Polypharmacy in
psychiatry: a review. Mens Sana Monogr 11:82–99. https://doi.
org/10.4103/0973-1229.104497
Lowe R, Glen RC, Mitchell JBO (2010) Predicting phospholipidosis
using machine learning. Mol Pharm 7:1708–1714
Lüllmann H, Lüllmann-Rauch R, Wassermann O (1978) Lipidosis
induced by amphiphilic cationic drugs. Biochem Pharmacol
27:1103–1108
Muehlbacher M, Tripal P, Roas F, Kornhuber J (2012) Identifica-
tion of drugs inducing phospholipidosis by novel in vitro data.
ChemMedChem. https://doi.org/10.1002/cmdc.201200306
Nonoyama T, Fukuda R (2008) Drug-induced phospholipidosis—
pathological aspects and its prediction. J Toxicol Pathol 21:9–24
1 3