Targeted Manipulation of Brain Serotonin: RNAi-Mediated Knockdown

Subscriber access provided by CAL STATE UNIV BAKERSFIELD

Article

Targeted Manipulation of Brain Serotonin – RNAimediated Knockdown of Tryptophan hydroxylase 2 in Rats Susann Matthes, Valentina Mosienko, Elena Popova, Marion Rivalan, Michael Bader, and Natalia Alenina ACS Chem. Neurosci., Just Accepted Manuscript • DOI: 10.1021/acschemneuro.8b00635 • Publication Date (Web): 12 Apr 2019 Downloaded from http://pubs.acs.org on April 15, 2019

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 28 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Chemical Neuroscience

Targeted Manipulation of Brain Serotonin – RNAi-mediated Knockdown of Tryptophan hydroxylase 2 in Rats Susann Matthes1,2, Valentina Mosienko1,3, Elena Popova1, Marion Rivalan4, Michael Bader1,2,4,5,6*, Natalia Alenina1,5,7* 1

Max-Delbrück Center for Molecular Medicine (MDC), Robert-Rössle- Straße 10, 13125 Berlin-Buch,

Germany 2

University of Lübeck, Institute for Biology, Ratzeburger Allee 160, 23562 Lübeck, Germany

3

College of Medicine and Health, Institute of Biomedical and Clinical Sciences, University of Exeter,

Hatherly Building, Prince of Wales Rd., EX4 4PS, Exeter, UK 4

Charité University Medicine, Charitéplatz 1, 10117 Berlin, Germany

5

German Center for Cardiovascular Research (DZHK), Partner Site Berlin, Berlin, Germany

6

Berlin Institute of Health (BIH), Anna-Louisa-Karsch-Straße 2, 10178 Berlin, Germany

7

Institute of Translational Biomedicine, St. Petersburg State University, Saint Petersburg, 199034,

Russia * shared last authorship

corresponding author: [email protected]

I ACS Paragon Plus Environment

ACS Chemical Neuroscience 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 28

Abstract Tryptophan hydroxylase (TPH) is the rate-limiting enzyme in the biosynthesis of the biogenic monoamine serotonin (5-hydroxytryptamine, 5-HT). Two existing TPH isoforms are responsible for the generation of two distinct serotonergic systems in vertebrates. TPH1, predominantly expressed in gastrointestinal tract and pineal gland, mediates 5-HT biosynthesis in non-neuronal tissues, while TPH2, mainly found in the raphe nuclei of the hindbrain, is accountable for the production of 5-HT in the brain. Neuronal 5-HT is a key regulator of mood and behaviour and its deficiency has been implicated in a variety of neuropsychiatric disorders, e.g. depression and anxiety. To gain further insights into the complexity of central 5-HT modulations of physiological and pathophysiological processes, a new transgenic rat model, allowing an inducible gene knockdown of Tph2 was established based on doxycycline-inducible shRNA-expression. Biochemical phenotyping revealed a functional knockdown of Tph2 mRNA expression following oral doxycycline administration, with subsequent reductions in the corresponding levels of TPH2 enzyme expression and activity. Transgenic rats showed also significantly decreased tissue levels of 5-HT and its degradation product 5-Hydroxyindoleacetic acid (5-HIAA) in the raphe nuclei, hippocampus, hypothalamus, and cortex, while peripheral 5-HT concentrations in the blood remained unchanged. In summary, this novel transgenic rat model allows inducible manipulation of 5-HT biosynthesis specifically in the brain and may help to elucidate the role of 5-HT in the pathophysiology of affective disorders.

Keywords: serotonin, tryptophan hydroxylase, depression, RNA interference, transgenic rat

II ACS Paragon Plus Environment

Page 3 of 28 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Introduction Serotonin (5-hydroxytryptamine, 5-HT) is an evolutionary ancient molecule, widespread throughout the animal and plant kingdoms. In mammals, 5-HT acts as a neurotransmitter within the central and peripheral nervous systems (CNS, PNS), and as a local hormone in various other non-neuronal tissues, including the gastrointestinal tract, the cardiovascular system and immune cells. This functional duality of the 5-HT system is typical for all vertebrates 1. In the periphery, 5-HT is mainly produced by enterochromaffin cells (EC) of the gut and released to the circulation where it is taken up by thrombocytes and stored in specific vesicles 2. Because of its hydrophilic properties, 5-HT is not able to penetrate the blood-brain barrier (BBB) and is independently synthesized in the brain, by the serotonergic neurons of raphe nuclei in the brainstem. Central 5-HT is important for normal brain development 3, regulation of physiological functions, and nociception. In addition, 5-HT affects nearly all behavioural patterns, including memory, mood, stress response, aggression, fear, appetite, addiction as well as maternal and sexual behaviours

4-9.

An imbalance in

the 5-HT system has been implicated in a multitude of neuropsychiatric diseases. The biosynthesis of 5-HT is a highly regulated two-step process, starting with the essential amino acid L-tryptophan (Trp). The first and rate-limiting step comprises the hydroxylation of Trp to 5-hydroxytryptophan (5-HTP). This reaction is carried out by the enzyme tryptophan hydroxylase (TPH) and requires Fe2+ ions as a cofactor and molecular oxygen (O2) and tetrahydrobiopterin (BH4) as cosubstrates

10.

Secondly, 5-HTP is immediately decarboxylated to 5-hydroxytryptamine (5-HT) by

the ubiquitously expressed aromatic amino acid decarboxylase (AADC). For several decades, confusing data about divergent protein/mRNA ratios and biochemical characteristics of TPH proteins from peripheral sources and from the CNS strongly implied the existence of different TPH isoforms 1. In 2003, targeted ablation of the only known Tph gene (now Tph1) in mice, unravelled the existence of a second isoform, encoded by a novel Tph2 gene with distinct chromosomal localization 11. TPH1 and TPH2 proteins in vertebrates are highly homologous, sharing an overall 70% amino acid sequence identity in humans, but differ in their kinetic properties and, more remarkably, in their tissue distribution

1,12.

Further studies of mRNA and protein levels in rodent and human tissues confirmed

TPH2 to be the neuronal isoform, predominantly expressed in the neurons of raphe nuclei in the brainstem

1,13-15

and in myenteric neurons in the gut, while it is absent in peripheral organs, such as

lung, heart, kidney or liver 16-19. On the other hand, TPH1 is mainly found in the gastrointestinal system as well as in the pineal gland, where it produces 5-HT serving as a precursor molecule for melatonin biosynthesis 1,20. The inability of 5-HT to cross the BBB enforces the dualistic character of the 5-HT system by creating two physiologically separated 5-HT pools in the body. In fact, both 5-HT systems are defined by the TPH1 and TPH2 isoforms and characterized by distinct physiological functions and independent regulatory mechanisms. Consequently, both systems can be targeted in an autonomous fashion to pharmacologically or genetically manipulate central and peripheral 5-HT functions 1,21,22. We previously used RNA interference technology to generate a transgenic rat with doxycycline (Dox)inducible diabetes type 2, by inhibiting insulin receptor (InsR) expression

23.

This inhibition was

achieved by a shRNA homologous to InsR mRNA under the control of the TetR/tetO system in all cells of the animals. The knockdown of InsR was reversible and its normal protein expression resumed

3 ACS Paragon Plus Environment


after Dox treatment was stopped

23.

Page 4 of 28

These rats have already proven their suitability for the study of

pathogenetic processes during diabetes

24,25

and for the testing of novel antidiabetic drugs

26.

In order

to allow such studies also for diseases based on central 5-HT deficiency, we generated a novel transgenic rat model with Dox-inducible expression of an shRNA against Tph2 (tetO-shTPH2 transgenic rats). The tetracycline (Tet)-controlled system is a drug-mediated gene activation strategy, which ensures a temporal control of a transgene expression

27.

The development of Tet-responsive systems originated

from the regulatory elements of the Transposon 10 (Tn10) Tet resistance operon of Escherichia coli, in which the Tet repressor (TetR) binds to the operator sequences (tetO) located within the promoter region of the operon. This interaction negatively regulates the expression of Tet resistance genes. When Tet or its synthetic analogue Dox is present, the TetR undergoes a conformational change and dissociates from the tetO sequence, thereby facilitating the initiation of gene transcription

28.

The

reversible Tet-controlled system used in this study comprises a bimodal expression system, combining both tetR and tetO sequences within one single construct, functionally distinguished by two different promoters 29-31. The lipophilic inducer molecule of choice in transgenic animals is Dox, which has been shown to be effective at low doses without triggering any cytotoxic response. Due to its excellent cell and tissue penetration properties, the required Dox concentrations can be easily achieved, even in compartments like the placenta, which are separated by specific barriers, though it was also observed that Dox penetrates the BBB less efficiently. Among the various routes by which Dox may be administered, the supply via drinking water has been proven to be the most simple and reliable one 27. In contrast to existing constitutive Tph2 knockout models

32,33,

tetO-shTPH2 transgenic rats lack the

risk of adaptive changes, since the CNS can develop normally and in the presence of 5-HT. Furthermore, rats have noticeable experimental advantages over mice, as their biology is more similar to humans

34

and a broader spectrum of behavioural tasks is readily available for testing rat social

behaviours and complex cognitive abilities, compared to less sophisticated experimental tools yet available for testing similar functions in mice.


Page 5 of 28 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Results and Discussion Generation of a tetO-shTPH2 transgenic rat line The shTph2 oligonucleotides were designed to target a non-conserved region in exon 2 of the Tph2 gene, thereby generating isoform specificity and cloned in a vector with the ubiquitously active U6 promoter. To test their efficiency in vitro, Cos7 cells were co-transfected with plasmids coding for rat Tph2 and either siRNA against Gfp and Tph2 or plasmids coding for shEgfp and shTph2. Western blot analysis showed a significant reduction of TPH2 protein expression in Cos7 cells co-transfected with shTph2 expression plasmids, compared to shEgfp transfected control cells (Figure 1). Furthermore, with a decrease of TPH2 protein by 63%, Tph2 shRNA was significantly more efficient than siTph2 (46%) (Figure 1). Thereafter, this shRNA was used to generate the pH1tetO-shTPH2-CAGGStetR expression cassette, which was microinjected into zygotes of Sprague Dawley (SD) rats to generate transgenic rats (TG). Out of 70 animals born after microinjection, genotyping PCR confirmed one positive transgenic rat (#9321; Figure 2A). This male founder rat was further bred with SD females to assess successful germline transmission and to establish a transgenic line. Hemizygous animals of the first filial (F1) offspring were interbred to create homozygous progeny for long-term maintenance of the line (Figure 2B). While absent in wildtype (WT) controls, tetR protein was expressed in different central and peripheral tissues of shTph2 transgenic rats (Figure 2C). This strongly implied that the construct is transcriptionally active and has not been silenced. Tph2 expression is reduced in Dox-treated tetO-shTPH2 transgenic rat First, we analysed the effect of 14 days of Dox treatment (20mg/kg body weight) on TPH2 expression in homozygous TG and WT rats. To analyse Tph2 mRNA levels, qRT-PCR was performed from raphe nuclei of Dox-treated transgenic rats and controls, using GAPDH as reference. Tph2 expression showed a significant 60% reduction exclusively in Dox-treated TG animals, while there was no effect of Dox in WT rats (Figure 3A). In order to exclude a potential loss of serotonergic neurons by toxic effects of the shRNA we measured the expression of a general neuronal marker, class III β-tubulin (TUBB3, Figure 3B), and two specific markers for 5-HT neurons, the serotonin transporter, Sert (Figure 3C), and the transcription factor, Pet1 (Figure 3D). None of these marker genes showed reduced expression. Next, the TPH2 protein expression in the raphe nuclei was analysed by immunoblotting, which revealed a significant reduction of up to 50% solely in Dox-treated TG males, while protein levels in all other groups remained comparable to untreated WT controls (Figure 4). Importantly, Dox treatment of WT and TG rats did not affect their drinking and eating behaviour: liquid consumption and body weight gain did not differ between Dox-treated and untreated rats (Supplementary Figure S1). Tph2 knockdown results in lowering brain, but not peripheral 5-HT levels To verify the physiological impact of shRNA-induced reduction of Tph2 mRNA and protein levels, the quantities of central and peripheral 5-HT and its catabolite 5-HIAA were evaluated by HPLC analysis of raphe nuclei, hypothalamus, hippocampus, cortex and blood of TG and WT rats. After 14 days of Dox treatment, male TG rats showed significant reductions of 5-HT quantities in all brain areas analysed (Figure 5), with an average 5-HT decrease of about -24% in raphe nuclei, -20% in hypothalamus and hippocampus and -13% in cortex (Table 1). Similarly, significantly reduced 5-HT levels



Page 6 of 28

could also be observed in raphe nuclei, hypothalamus and cortex of Dox-treated TG female rats (Table 1). As expected, peripheral 5-HT levels in the blood in TG animals were not affected by Dox administration (Figure 5). HPLC data further confirmed no changes of the 5-HT levels in brain and blood of WT males (Figure 5), after 14 days of Dox treatment. Analysis of the major 5-HT catabolite 5-HIAA generally reflected the decrease observed in 5-HT quantities of 14 days Dox-treated TG rats (Table 1). 5-HIAA levels were significantly reduced in raphe (-28%), hypothalamus (-20%) and hippocampus (-10%) of Dox-treated TG males. In the cortex however, the 13% reduction in 5-HT levels did not result in subsequent decrease of 5-HIAA. In Dox-treated TG females all brain areas displayed significantly diminished 5-HIAA levels, with the strongest reduction of about 35% in the raphe (Table 1). We also checked if the observed effects are persistent over a longer period of time: 28 days of Dox administration had comparable effects, leading to a 30% reduction in the brain 5-HT and 5-HIAA levels, whereas Trp levels were not affected by the treatment (Supplementary Figure S2). Dox-administration in tetO-shTPH2 rats leads to in-vivo and in-vitro reduction in TPH activity in the brain. We further evaluated, if Tph2 knockdown effectively downregulates 5-HT synthesis rate in vivo by analysing 5-HTP accumulation after administration of the AADC inhibitor, NSD. Indeed, 28 days of Dox treatment resulted in significant reduction in 5-HTP levels in different brain areas (in raphe nuclei by 50%, hippocampus by 40%, hypothalamus by 53%, cortex by 40%, striatum by 52%; Figure 6A), but not in pineal gland (Figure 6B) or liver (data not shown). Moreover, in vitro analysis of TPH2 activity revealed a 28% decrease in 5-HTP accumulation in raphe nuclei lysates of Dox-treated TG rats under physiological Trp concentrations (Figure 6C). TPH2 activity was 10 times higher when substrate Trp was added in excess to the reaction, and was decreased by half (48%) in lysates of TG animals treated with Dox in comparison to untreated TG rats. Addition of the TPH inhibitor LX1606 35 to the reaction completely blocked the 5-HTP synthesis, confirming the specificity of the reaction. Relevance of an inducible transgenic rat model The RNAi approach used in this study aimed at the creation of a novel animal model that will be of use to investigate the role of serotonin in affective disorders, and resulted in the successful generation of an inducible shTph2 transgenic rat line. Genetically modified animals are essential tools to study in vivo gene function and to model human diseases. Due to the availability of more sophisticated technologies for introducing genetic modifications, such as embryonic stem (ES) cell-mediated gene targeting, mice became the more attractive model among rodents, allowing specific genes to be knocked out, knocked in or conditionally switched on or off

36.

However, the

laboratory rat represents the first animal species specifically bred for scientific purposes and has become an indispensable model organism for biomedical research. Being more similar to humans, rats provide important contributions to many health-related research fields, including cardiovascular, metabolic and autoimmune diseases, and are further important for toxicity testing and pharmacological studies. Their larger size makes rats more suitable for physiological manipulations, such as neurosurgery and because they are more sociable and intellectually skilled, rats are also the preferred animal model in many behavioural assessments 34.


Page 7 of 28 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


While transgenic technologies for rats were already established in 1990

37,

targeted alterations of the rat

genome became possible only 10 years ago, when finally germline competent rat embryonic stem cells became available

38,

shortly followed by the development of new technologies, e.g. molecular scissors, such

as zinc-finger nucleases (ZFN), transcription activator-like effector nucleases (TALENs), and the CRISPR/Cas9 system 39-43. However, one major limitation in the generation of germline knockouts in general, is the risk of embryonic lethality, as observed in e.g. Lmx1b or Vmat2 knockouts, that die in utero or shortly after birth

44,45.

Methods

of (inducible) conditional gene knockout, using site-specific recombinases, such as the Cre/loxP or Flp/FRT system, have helped to overcome the obstacle of a lethal phenotype, by allowing in vivo gene inactivation in a time- and tissue-specific manner throughout development or in adulthood

40.

Hence, the generation of

conditional mice with (inducible) deletion of Lmx1b exclusively in Pet-1 expressing cells (Lmx1bPet1-cre Lmx1bPet1-icre

47),

or specific ablation of Vmat2 in Sert-expressing cells (Vmat2Sert-cre

48),

46,

resulted in viable

mouse lines, displaying strongly reduced 5-HT levels in the brain. In mice, the constitutive knockout of Tph2 is not lethal and has been successfully attempted by several research groups

7,15,49-53.

Alongside parallel findings on drastically reduced central 5-HT concentrations and

mostly unaltered formation of 5-HT neurons, inconsistent results have been reported on aspects of anxietyand depression-like behaviour, as well as on levels of other neurotransmitters. This might be influenced by the genetic background of the animals and by discrepancies within the methodologies used, but may also reflect the influence of adaptive mechanisms occurring during animal development. Another difference, indicating compensatory circuitries arising upon the developmental loss of 5-HT neurons, is observed in autonomic functions, such as respiration and thermoregulation. Mice with complete depletion of central 5-HT (Tph2-/-)

7

or near-complete loss of central 5-HT neuron (Lmx1bf/f/p)

6

display a normal body temperature at

basal levels, but develop severe hypothermia during a cold challenge, while adult animals subjected to acute perturbation of 5-HT neurons already show pronounced disruptions of body thermoregulation at room temperature

54.

These findings suggest that alternative strategies, such as sequence-specific gene-silencing

with the help of RNAi

55,

may address the limitations of animal models with complete gene knockout by

circumventing the risk of adaptive changes, which might help to gain new insights into such complex disorders as depression. Historically, rats with targeted shRNA expression were mostly generated by lentiviral-mediated transgenesis 56,57.

This strategy requires extensive and time consuming animal breeding, since founders are born with

genetic mosaicism of transgene expression and frequently lack germline transmission

58.

Therefore, the

procedure of choice in this study, was the pronuclear microinjection of a linearized DNA construct into fertilized oocytes. This method has been reportedly successful in knocking down gene expression in mice and rats, in numerous tissues and at any developmental stage

59.

Even though the mechanical process of

pronuclear microinjection is more challenging and less efficient in rats than the lentiviral system, the technique is well established in the group shRNA-based rat model

23.

60

and was recently used for the successful generation of another

However, with the rate of 1.43%, in transgenesis for the shTph2 construct was

slightly below the usual efficiency of 2.7-10.0%, and the generation of only one germline competent founder animal proved the difficulty of pronuclear microinjection, independent of the transgene-construct or the animal strain used, since SD rats display a favourable overall transgenic efficiency, when compared to other strains e.g. Wistar and Lewis rats 60,61.



Page 8 of 28

It was also suggested that a low output of transgenic founders carrying shRNA constructs might be the result of embryotoxicity caused by high shRNA expression in embryos

23,62.

Yet this problem was avoided by

employing a Tet-inducible system, where shRNA expression remained suppressed until the oral administration of Dox. Other advantages of Tet-inducible systems are the reversibility of gene-silencing, the feasibility of dose-response studies over a broad range of shRNA expression levels and the functional exploration of numerous mammalian genes in-vitro and in-vivo

63.

Various systems of Tet-inducible

transgenic gene silencing have been described in mice, successfully targeting e.g. ΔfosB, CREB kinase

65,

ABCA1

66,

tumor suppressor Trp53

genes, like 5-HT1A receptor

67,

or insulin receptor

23,29,

64,

PI3-

as well as specifically neuronal

68.

The construct for the generation of Tet-inducible transgenic rats targeting Tph2, was technically based on the bimodal system that was previously used to create an inducible rat model with shRNA-mediated knockdown of the insulin receptor

23,

by exchanging the shRNA sequence with specific and highly efficient

shTph2 oligos. Following pronuclear microinjection, the random fashion of transgene integration can lead to position site-dependent effects, such as chromatin-mediated silencing, that may affect transgene expression 36.

However, the detection of TetR protein expression in various peripheral and central tissues constitutes

convincing evidence for the ubiquitous transcriptional activity of the transgene. Following 14 days of oral Dox treatment, shTph2 transgenic rats display significant reductions in TPH2 mRNA and protein levels. This is in line with earlier studies, where Dox concentrations of 20mg/kg body weight (200μg/ml) showed to be efficient in silencing gene expression, without inducing an inflammatory response 23,29,57. Furthermore, mRNA and protein levels in untreated transgenic rats were comparable to WT animals (Figure 3), indicating that the H1 promoter is not leaky in the uninduced state and shRNA expression remains latent until Dox is added to the drinking water. However, Dox was reported to be unevenly distributed throughout the animal body and very often only 20–30% of the serum concentration of Dox was detected in the brain 27,63. Yet, since TPH2 expression is restricted to the raphe nuclei neurons, the efficiency of the Dox-induced knockdown could not be compared with peripheral tissues. Furthermore, because only one founder animal was obtained, there is also no option to analyze several transgenic lines with different degrees of knockdown. However, we saw about the same shRNA-induced decrease of InsR mRNA protein levels in tetO rats (by 27-36%) in the brain, after receiving acutely high doses of Dox (2mg/ml in 10% sucrose) 23. Yet in another study, Tph2 knockdown in SD rats was induced by stereotaxic infusions of antisense phosphorodiamidate morpholino oligonucleotides into the caudal dorsal raphe nucleus. This resulted in 60% reduction in TPH protein levels, but the effect was restricted to the site of injection

69.

Others injected Tph2-

shRNA in the rostral ventromedial medulla (RVM) of SD rats, which resulted in a subsequent knockdown of TPH2 protein of max. 80% on day three after injection 70. Several RNAi-based knockdown strategies targeting other components of the 5-HT system in mice have been published. Acute intracerebral infusion or intranasal administration of siRNA against 5-HT1A autoreceptor was reported with knockdown efficiencies of 40%

71

and 44%

72

in the corresponding brain

areas. Other groups used the same delivery strategy for the siRNA-mediated knockdown of Sert, obtaining reductions of 40% at mRNA level and 18-60% 73 or 30-40% 74 reduced SERT binding. Even though there are notable differences in the concentrations and the duration of siRNA administration underlying these results, the knockdown was in any case sufficient to evoke increased serotonergic neurotransmission and


Page 9 of 28 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


hippocampal neurogenesis

74,

or antidepressant-like responses

71-73.

These findings confirm, that the degree

of Tph2 knockdown, obtained by Dox-induced shRNA expression, is comparable to non-transgenic RNAi approaches in mice and rats and might subsequently translate into behavioural alterations in this rat model. This notion is further supported by the fact that both of the previously mentioned injection-based rat models of Tph2 knockdown revealed behavioural phenotypes. In the first study the TPH2 downregulation was interfering with estrogen-dependent anxiolytic effects, while the overexpression of TPH2 had opposing effects

69.

However, Hiroi et al.

69

did not comment on any corresponding central 5-HT levels. In another

study, the TPH2 knockdown resulted in 70% depletion in 5-HT levels in the spinal dorsal horn, which was associated with attenuated formalin-induced spontaneous nocifensive responses, tissue or nerve injuryinduced allodynia and hyperalgesia, underpinning the role of 5-HT in pain facilitation

70.

Moreover, several

studies reported association of mild reductions in brain 5-HT levels induced by Trp-free diet (25-30 %) or TPH2 polymorphisms (10-40%) with behavioural alterations, such as changed aggression or anxiety

75-78.

Thus, the degree of 5-HT reduction in the Dox-treated shTph2 transgenic rats should be sufficient to evoke behavioural phenotypes. Following 14 days of oral Dox treatment, shTph2 transgenic male and female rats show significant reductions in central 5-HT and 5-HIAA tissue levels. The 5-HT levels in the blood, and pineal gland remained unchanged, which confirms the isoform specificity of the shRNA and indicates the absence of obvious offtarget effects resulting from the ubiquitous shTph2 expression, driven by the non-specific H1 promoter. Equivalent 5-HT levels in treated and untreated WT animals further assured no undesired effects mediated by Dox itself. Importantly, the effect of TPH2 knockdown on 5-HT levels was similar in males and females, assuring the suitability of this rat model for studies in both sexes. Different brain parts were chosen for HPLC analysis, based on their proposed functions and subsequent relevance for behavioural alterations, e.g. hippocampus (learning, memory), hypothalamus (autonomic, endocrine and behavioural control, e.g. stress and emotional responses) and cortex (impulse control, motivation). Surprisingly, in transgenic rats, the 50% knockdown of TPH2 only resulted in 20-30% reductions of 5-HT levels in different brain areas. Moreover, the degree of 5-HT decrease was not affected by the duration of the Dox treatment: the 30% drop in 5-HT levels persisted in transgenic rats also after 28 days of Dox administration without any signs of toxicity. These findings imply the influence of underlying regulatory mechanisms in place to partially compensate an induced 5-HT deficiency, which is in line with similar findings in heterozygous Tph2+/- mice

7,15,49,

or mice harbouring a hypomorphic Tph2 SNP

9,77,79,80.

Nevertheless, the reduction in brain 5-HT in shTph2 transgenic rats (30%) is more pronounced than in heterozygous Tph2+/- mice (10% reduction in 5-HT levels in whole brain lysates 9,81). Presumably the life-long downregulation of 5-HT synthesis in Tph2+/- mice starting already during embryogenesis, triggers long-term compensatory mechanisms, which maintain close to normal brain 5-HT levels in those mice. Obviously these developmental mechanisms regulating serotonin homeostasis do not apply in case of acute downregulation of 5-HT synthesis during adulthood in shTph2 transgenic rats. Probably, the rather short-term downregulation of TPH2 cannot be completely compensated, confirming the suitability of shTph2 transgenic rats as a model to study the consequences of acute depletions in brain 5-HT. The measured 5-HTP synthesis rates in raphe nuclei and other brain areas well reflected the 50% decrease in TPH2 protein. These suggest a reduced turnover of 5-HT to be responsible for the observed differences between 5-HT synthesis rate and content. However, the reduction of the 5-HT degradation product 5-HIAA



Page 10 of 28

was in most brain parts comparable to that of 5-HT, indicating that a diminished 5-HT catabolism is not primarily causing the effect. It is also likely that the minor differences in the 5-HT levels result from a background signal from residuals of blood, since it is technically not possible to completely remove it by cardiac perfusion. Hence, a closer investigation will be needed to explore other potential regulatory feedback mechanisms, such as changes in 5-HT transporter and receptor expression or sensitivity, as well as extracellular 5-HT. The main advantage of the shTph2 transgenic rat model is the inducibility of the knockdown, which allows monitoring influences on behaviour before and after Dox treatment. The direct application of RNAi agents faces additional obstacles, such as delivery-related toxicity, the lack of safe delivery to specific target cells and the difficulty to target selected neuronal populations. In addition, the efficiency of synthetic siRNA delivery in mammals is hampered by the lack of a systematic RNAi response that effectively transports siRNA across cellular boundaries

82.

Besides, less invasive knockdown techniques will avoid complications

that go along with neurosurgical interventions, e.g. tissue injury and infections, and might be more appropriate when testing animal behaviour. The intermediate phenotypes, created by shRNA-mediated knockdown, might better reflect the human disease situation than the complete absence of a protein

31

and can be a complementary tool to classical

knockout technology for the analysis of gene functions and to elucidate the consequences of developmental adaptation, like it has previously been described for the dopamine transporter

83.

The behavioural testing of

Dox-treated shTph2 transgenic rats is currently in progress, but was not within the scope of this study. Thus, the novel transgenic rat model described in this study represents a powerful new tool to analyse not only behavioural aspects, but also a broad range of hormonal and neurochemical changes, as well as the efficacy of pharmacological substances and will therefore help to unravel the complex role of 5-HT in the wide spectrum of affective disorders. Methods Animals and Doxycycline treatment Sprague Dawley (SD) rats (RjHAN:SD; from JANVIER, France) were used to generate the transgenic (TG) line and also served as wildtype (WT) controls. Animals were maintained in individually ventilated cages (Tecniplast, Italy) under specific pathogen-free, standardized conditions in accordance with the German Animal Protection Law. Rats were group-housed at a constant temperature of 21± 2°C with a humidity of 65 ± 5%, an artificial 12h light/dark cycle and unlimited access to standard chow (0.25 % sodium; Ssniff, Soest, Germany) and drinking water. Adult TG and WT rats were treated with doxycycline (Dox) (Fluka, Steinheim, USA) for 14 and 28 days to induce shRNA expression. The light-sensitive Dox solution (20-40 mg/kg body weight) was freshly prepared every second day and delivered in dark drinking bottles. During the experiments animals were frequently checked for body weight and liquid consumption, and Dox concentration was adopted to the drinking volume to maintain the Dox-dose at the constant level. On the final day of Dox treatment, rats were deeply anesthetized via isoflurane inhalation (Abott, Wiesbaden, Germany) and heparinized blood was collected via cardiac puncture and quickly precipitated with 710μM ascorbic acid and 2.4% perchloric acid (70%) (SigmaAldrich, Steinheim, Germany). After centrifugation, supernatants were kept at -80°C for HPLC measurement. Animals were transcardially perfused with PBS (Sigma-Aldrich, Steinheim, Germany) supplemented with


Page 11 of 28 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


300U/ml heparin (Braun, Melsungen, Germany). Organs and brain parts were collected, snap-frozen and stored at -80°C until further use. All experimental procedures were performed according to national and institutional guidelines and have been approved by responsible governmental authorities (Landesamt für Gesundheit und Soziales (LaGeSo), Berlin, Germany; G0300/13). shRNA oligo design and generation of an inducible transgenic rat line shTph2 was designed to target 5’-GGCAAAGACACTTCGGA-3’ sequence in exon 2 of the rat Tph2 gene. Mycoplasma-free Cos7 cells (ATCC, Manassas, VA, USA) were cultivated at 37°C and 5% CO2 in Dulbecco’s

modified

Eagle’s

medium

(DMEM),

containing

10%

fetal

bovine

serum

and

1%

penicillin/streptomycin (Gibco, Darmstadt, Germany). At 90% confluency, cells were co-transfected with 2µg of plasmid DNA coding for rat Tph2 (pCMVrTPH2) and either with siRNA against Gfp (commercially available FITC-conjugated siGfp), and Tph2 (5’-GCG GCA AAG ACA CUU CGG AUU CUC AAG AGA AAU CCG AAG UGU CUU UGC CGC-3’, BIOTEZ, Germany) or with plasmids pU6shTPH2Ex2 and pU6shGFPUbAsRed coding for shTph2 (5’- GCG GCA AAG ACA CTT CGG ATT CTC AAG AGA AAT CCG AAG TGT CTT TGC CGC-3’) and shEgfp (5’-GCA AGC TGA CCC TGA AGT TAT TCT CAA GAG AGA ACT TCA GGG TCA GCT TGC-3’), respectively, using LipofectamineTM2000 (Invitrogen, Darmstadt, Germany) according to the manufacturer’s instructions. Subsequently TPH2 protein expression was analysed by Western blot. Complementary sense and antisense oligos, coding for shTph2 and five thymidines for transcription termination (shTPH2Ex2 sense: 5’-TCC CGC GGC AAA GAC ACT TCG GAT TCA AGA GAT CCG AAG TGT CTT TGC CGC TTT TTT A-3’, shTPH2Ex2 antisense: 5’-CGC GTA AAA AAG CGG CAA AGA CAC TTC GGA TCT CTT GAA TCC GAA GTG TCT TTG CCG C-3’), were annealed and cloned 3’ of the H1tetO promoter sequence into the bimodal pINV-7 vector (Taconic, Köln, Germany) to generate the pH1tetOshTPH2-CAGGStetR expression plasmid. Transgenic rats were generated according to established techniques

23,60.

Briefly, a 4kb DNA fragment,

containing the H1tetO-shTPH2-CAGGStetR sequence, was released with PacI and KpnI restriction enzymes (New England Biolabs, MA, USA), dissolved in microinjection buffer (8mM Tris-HCl, pH 7.4, 0.15mM EDTA) to a final concentration of 3ng/µl and microinjected into fertilized oocytes of SD rats. Eggs were subsequently cultured for 2h and transferred into pseudo-pregnant SD foster mothers. Integration of the construct was determined by transgene-specific genotyping PCR (TetRfor: CAA GTT GCC AAG GAG GAG AG, TetRrev: AAC CGG TCT AGA ATC GAT GG) from genomic DNA, isolated from different tissues. Quantitative Realtime PCR Total RNA from cells and tissues was isolated using the TRIzol® Reagent (Invitrogen, Darmstadt, Germany), following the manufacturer’s instructions. Final RNA concentration was measured with a NanoDrop™ spectrophotometer (Peqlab, Erlangen, Germany) and 2µg of total RNA was subject to first-strand DNA synthesis, using random primers and M-MLV Reverse Transcriptase (Invitrogen, Darmstadt, Germany) according to the manufacturer’s protocol. Quantitative Realtime PCR (qPCR) was used to determine mRNA expression of Tph2 (rTPH2fw1: CCT TTG CAA GCA AGA AGG TC, rTPH2rev1: TTG GAA GGT GGT GAT TAG GC), Pet1 (ratPet1_Ex2_154_FW:



Page 12 of 28

ATG AGA CAG AGC GGC ACC T, ratPet1_Ex4_154_REV: ACT GCC ACA ACT GGA TCT GC), Sert (ratSERT_ex13_182_FW CCA GCT ACG GCT TTT CCA AT, ratSERT_ex14_182_REV: ACG GGA TTT CTG TGG GTG TT), class III β-tubulin (ratTubb3_Ex3_137_FW: CCA GAG CCA TTC TGG TGG AC, ratTubb3_Ex4_137_REV: ATA GTG CCC TTT GGC CCA GT) in the raphe nuclei and hippocampi of transgenic and wildtype rats, with Glyceraldehyde-3-phosphate dehydrogenase (GAPDH) mRNA (ratGAPDH_ex1_112_FW: ATG GTG AAG GTC GGT GTG AAC, ratGAPDH_ex2_112_REV: GGT CAA TGA AGG GGT CGT TG) serving as housekeeping/reference gene. The SYBR© green protocol together with the GoTaq® qPCR master mix (Promega, Mannheim, Germany) was applied according to the manufacturer’s instructions, using a 7900HT AbiPrism Real-Time PCR System (Applied Biosystems, Darmstadt, Germany). The qPCR program was followed by a melting curve analysis, controlling the specificity of the fluorescent signal. Gene expression was calculated based on the 2-∆∆CT method 84. Western blot analysis Total proteins from Cos7 cells were extracted using Cell lysis buffer (CellSignalling, MA, USA) supplemented with protease inhibitor cocktail (Roche, Mannheim, Germany), according to the manufacturer’s instructions. Supernatants of the cell extracts were collected and protein concentrations were measured with Bradford reagent (Sigma-Aldrich, Steinheim, Germany). 15µg of denaturated protein extracts were separated by SDSPAGE and subsequently transferred to methanol pre-activated PVDF membranes (Amersham Biosciences, Little Chalfon, UK). The membranes were blocked with 1x Roti-Block (Roth, Karlsruhe, Germany) and incubated with the primary antibodies (sheep anti-TPH 1:2500 (Abcam, Cambridge, UK) 1:1000 (BIOZOL, Eching, Germany); rabbit anti-β-actin 1:1000 (CellSignaling, MA, USA)) over night at 4°C. Next day, membranes were washed in TBST (50mM Tris, 150mM NaCl 0.05% Tween20) (Sigma-Aldrich, Steinheim, Germany) and incubated with peroxidase-conjugated secondary antibodies (rabbit anti-sheep HRP 1:10000 (Sigma-Aldrich, Steinheim, Germany); goat anti-rabbit HRP 1:2000 (Pierce Rockford, IL, USA)) for 1-2h at RT. Signal detection was performed using SuperSignaling™ West Dura Substrate (Thermo Scientific, Rockford, USA) and subsequent exposition and development of X-ray films (Fotochemische Werke GmbH, Berlin, Germany). To extract total proteins from different organs, tissue samples were homogenized in RIPA buffer (SigmaAldrich, Steinheim, Germany) supplemented with protease inhibitor cocktail (Roche, Mannheim, Germany), using a FastPrep™-24 device (MP Biomedicals, Eschwege). Protein concentration was measured with the Bicinchoninic Acid (BCA) Protein Assay Kit (Sigma-Aldrich, Steinheim, Germany), following the manufacturer’s protocol. 50µg of denaturated protein extracts were separated by SDS-PAGE and subsequently transferred to methanol pre-activated PVDF membranes (Amersham Biosciences, Little Chalfon, UK). The membranes were blocked with Odyssey® blocking solution (LI-COR Biosciences, Bad Homburg, Germany) and incubated with the primary antibodies (mouse anti-TetR 1:8000 (MoBiTec GmbH, Göttingen, Germany); mouse anti-TPH (WH3) 1:2000 (Sigma-Aldrich, Steinheim, Germany); rabbit antiGAPDH 1:1000 (CellSignaling, MA, USA)) over night at 4°C. Next day, membranes were washed in TBST (50mM Tris, 150mM NaCl 0.05% Tween20) (Sigma-Aldrich, Steinheim, Germany) and incubated with infrared dye-labeled secondary antibodies (IRDye coupled anti-rabbit/mouse 1:10000 (LI-COR Biosciences,


Page 13 of 28 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Bad Homburg, Germany)) for 1-2h at RT. Membrane exposure and signal quantification was performed by using the Odyssey®Infrared imaging system (LI-COR Biosciences, Bad Homburg, Germany). In-vivo 5-HTP synthesis rate To determine 5-HTP in-vivo synthesis rate, animals received the AADC inhibitor 3-hydroxybenzyl hydrazine (NSD1015, Sigma, 100mg/kg, ip). Animals were sacrificed 30 minutes later and 5-HTP tissue accumulation in different brain parts was quantified by HPLC analysis.

In-vitro TPH activity assay In-vitro TPH activity in raphe tissue was determined by the accumulation of 5-HTP in the presence of its substrate, cofactor and NSD1015

85.

Frozen raphe samples were homogenized in 75mM tris-acetate (pH

7.5) (Roth, Karlsruhe, Germany), sonicated and shortly centrifuged. The collected lysates were mixed with 2mg/ml catalase, 100μM ferrous ammonium sulfate, 25mM 1,4-dithiothreitol (Sigma-Aldrich, Steinheim, Germany) and incubated for 10min at 30°C. Next, 15mM tris-acetate (pH 6.4) (Roth, Karlsruhe, Germany), 50μM L-tryptophan, 50μM tetrahydrobiopterin and 2mM NSD1015 (Sigma-Aldrich, Steinheim, Germany) were added and incubated for 30min at 37°C. The reaction was stopped with 2.4% perchloric acid (70%) (Sigma-Aldrich, Steinheim, Germany) and the 5-HTP-containing supernatants were analysed by HPLC. HPLC analysis Frozen tissue samples were homogenized in 710μM ascorbic acid and 2.4% perchloric acid (70%) (SigmaAldrich, Steinheim, Germany), precipitated proteins were pelleted through centrifugation (20min, 20000xg, 4°C) and the collected supernatant was analyzed for serotonergic metabolites (Trp, 5-HTP, 5-HT and 5HIAA) using high-sensitive reversed-phase HPLC with fluorometric detection

86.

Samples were separated at

20°C in 10mM potassium phosphate buffer (pH 5.0) (Sigma-Aldrich, Steinheim, Germany) with 5% methanol (Roth, Karlsruhe, Germany), using a C18 reversed-phase column (LipoMare C18, AppliChrom, Oranienburg, and ProntoSil 120 C18 SH, VDS Optilab, Berlin) and a flow rate of 0.8-1.0 ml/min. Column eluates were excited at 295nm and the fluorescent signal was measured at 345nm. CLASS-VP software (Shimadzu, Tokyo, Japan) was used to analyse the peak parameters of chromatographic spectra and to quantify substance levels, based on comparative calculations with alternately measured external standards. Statistics All data was subject to statistical analysis using GraphPad PRISM 6 software (San Diego, CA, USA). The results shown in this study represent mean values  standard error of mean (SEM), unless stated otherwise. Tests defining the level of significance included Student’s t-test or Mann-Whitney U-test to compare independent pairs of means, while more than two groups were analysed by one-way ANOVA and Bonferroni post-hoc test. Results with p

Targeted Manipulation of Brain Serotonin: RNAi-Mediated Knockdown

Recommend Documents