Arch Toxicol (2013) 87:361–370 DOI 10.1007/s00204-012-0922-3

REPRODUCTIVE TOXICOLOGY

Effects of mycophenolic acid alone and in combination with its metabolite mycophenolic acid glucuronide on rat embryos in vitro Flavia Schmidt • Kathrin Eckardt • Mehdi Shakibaei Petra Glander • Ralf Stahlmann

•

Received: 16 June 2012 / Accepted: 31 July 2012 / Published online: 23 August 2012 Ó Springer-Verlag 2012

Abstract Mycophenolic acid (MPA) is an immunosuppressive agent that acts as a selective, non-reversible inhibitor of the enzyme inosine-50 -monophosphate dehydrogenase (IMPDH). Malformations have been described in children after maternal exposure to mycophenolate. However, the causal link is unclear in most cases because women had been treated with a combination of drugs and birth defects may have other causes. Therefore, it is important to study the action of this drug and its main metabolite on embryonic tissue. We studied the teratogenic potential of MPA and its major metabolite, the mycophenolic acid glucuronide (MPAG) in the rat whole-embryo culture. A total of 147 day 9.5 embryos were cultivated for 48 h in the standard medium containing 85 % serum. We tested MPA at concentrations of 0.1; 0.25; 0.5; 0.75 mg/l (0.31; 0.78; 1.56; 2.34 lM) and MPA glucuronide at concentrations of 3; 10; 30; 100 mg/l (6.04; 20.14; 60.43; 201.43 lM). Both substances are highly protein bound, and MPA glucuronide might displace MPA from protein binding. Therefore, we examined whether the effects of MPA can be enhanced when studied in combination with the glucuronide. Furthermore, the focus was on additional endpoints to the standard evaluation of cultivated embryos,

such as development of cranial nerves [trigeminal nerve (V), facial nerve (VII), glossopharyngeal nerve (IX), vagus nerve (X)] after staining with an antibody against 2H3 neurofilament. Ultrastructural changes were evaluated by electron microscopy. At a concentration of 0.75 mg MPA/l medium, all embryos showed dysmorphic changes. Embryos exposed to 0.25 mg MPA/l medium showed impaired development of nerves, and at 0.1 mg/l, no effects were detectable. Concentration-dependent ultrastructural changes, such as signs of apoptosis, were found by electron microscopy. The examination of the metabolite in this assay showed that at a concentration of 100 mg MPAG/l, the embryos exhibited distinct malformations. This is probably caused by MPA, which was detectable at 0.6 % in the material used for our experiments. The combination of the parent compound (0.03; 0.1; 0.25 mg/l) with its metabolite MPAG (3 mg/l) did not cause enhanced toxicity under our experimental conditions. IMPDH, the target enzyme of MPA, could be detected in rat embryos on day 9.5 of embryonic development as well as at the end of the culture period 48 h later. In summary, MPA impairs embryonic development at low, therapeutically relevant concentrations, but the glucuronide does not exhibit such a potential. Activity of MPA is not enhanced by MPAG.

F. Schmidt  K. Eckardt  R. Stahlmann (&) Institute for Clinical Pharmacology and Toxicology, Charite´, Universita¨tsmedizin, Luisenstraße 7, 10117 Berlin, Germany e-mail: [email protected]

Keywords Mycophenolic acid  Mycophenolic acid glucuronide  Whole-embryo culture  Whole immunostaining  Inosine-50 -monophosphate dehydrogenase activity  Apoptosis

M. Shakibaei Institute of Anatomy, Ludwig-Maximilian-University, Munich, Germany

Introduction

P. Glander Department of Internal Medicine, Nephrology, Charite´, Universita¨tsmedizin, Berlin, Germany

Mycophenolic acid (MPA), 6-(4-hydroxy-6-methoxy7-methyl-3-oxo-5-phthalanyl)-4-methyl-4-hexenoic acid, is

123

362

the active pharmaceutical agent of the immunosuppressive drugs mycophenolate mofetil (CellCept) and enteric-coated mycophenolic sodium (Myfortic), which are used for immunosuppression after solid organ transplantation and for the treatment of autoimmune diseases (Perez-Aytes et al. 2008; Neumann et al. 2008). Also, it is important to study compounds that are teratogenic in routinely performed teratogenicity tests and which probably cause birth defects in humans in the whole-embryo culture to examine whether this in vitro system is able to detect the teratogenic potential of such substances. The whole-embryo culture is a validated assay, and if it is used as an ‘‘alternative’’ test, for example, in the context of the REACH legislation to reduce and refine animal experiments, it is a prerequisite that it produces reliable results. Routinely performed animal studies showed that MPA is a teratogenic agent. Studies in pregnant rabbits and rats treated with mycophenolate mofetil revealed fetal malformations and increased incidence in abortions or resorptions (Tendron et al. 2002). Furthermore, a number of case reports demonstrate an increased risk of spontaneous abortions and major malformations in newborns after their mothers had been treated with MPA during pregnancy (Anderka et al. 2009; Merlob et al. 2009; Lin et al. 2011). As a consequence, the Food and Drug Administration (FDA) changed labeling of MPA from Pregnancy Category ‘‘C’’ to ‘‘D’’, which is defined as ‘‘positive evidence of human fetal risk, but potential benefits may warrant use of the drug in pregnant women despite the potential risk’’ (Sifontis et al. 2006; Anderka et al. 2009). The mechanism of the teratogenicity of MPA is unclear. The principal immunosuppressive effect of MPA is caused by inhibition of the enzyme inosine-50 -monophosphate dehydrogenase (IMPDH), which is essential for the de novo synthesis of guanosine. MPA induces a depletion of guanosine nucleotides and thereby impairs lymphocyte proliferation. In contrast to the parent compound, the main metabolite mycophenolic acid glucuronide (MPAG) does not show immunosuppressive activity. MPA is a highly protein-bound drug with a bound fraction of 97 %, and MPAG is bound to albumin at 82 %. Because the number of binding sites in plasma proteins for MPA and MPAG is restricted, it is possible that MPAG at high concentrations could displace MPA from protein binding. The human plasma concentration of MPAG is 20- to 100-fold higher than that of the parent compound. The unbound fraction of MPA is thought to be responsible for the immunosuppression, but also for the toxic effects (Bullingham et al. 1998). We examined whether the combination of both compounds causes more pronounced effects on the embryos than each substance alone, possibly as a result of deliberation of MPA from protein binding. New and co-workers developed the whole-embryo culture 40 years ago (Augustine-Rauch et al. 2010), and soon after, this method was discussed as a

123

Arch Toxicol (2013) 87:361–370

technique for the assessment of teratologic effects (New 1976). Today, the whole-embryo culture is a widely used in vitro assay to estimate embryotoxic potential of xenobiotics (Genschow et al. 2002; Flick and Klug 2006). The method can be optimized through additional endpoints for the evaluation of the embryonic changes and thus enhance the importance as a tool for the reproductive safety assessment of xenobiotics. The effects of MPA alone on cultivated rat embryos have been tested before in our laboratory. All embryos were evaluated by a routinely used score system of the wholeembryo culture assay (Eckardt and Stahlmann 2010). In the present study, we added specific endpoints to the standard evaluation of cultivated embryos to possibly increase the sensitivity of the assay. Neurofilaments of MPA-exposed embryos were marked with a specific antibody to evaluate the development of cranial nerves. In addition, cellular changes were studied by electron microscopy. The wholeembryo culture is a useful tool to study specific mechanism of action of xenobiotics. Therefore, we measured alterations in enzyme activity of IMPDH—the target enzyme of MPA in lymphocytes—in cultivated rat embryos and compared it with enzyme activity in embryos after in utero development.

Materials and methods Test compound Mycophenolic acid (MW: 320.3 g/mol; 1 mg/l = 3.12 lM) was purchased from Alexis Biochemicals (cat. no. 380-015M250, batch no. L13111/b) Lausen, Switzerland. Mycophenolic acid glucuronide (MW: 496.46 g/mol; 1 mg/ l = 2.01 lM) was ordered by Analytical Services International Ltd (Batch No. ASI/MPAG/006) Liverpool, United Kingdom. MPA was dissolved in DMSO and MPAG in methanol. The final solvent concentration in culture medium did not exceed 0.1 %. To examine the stability of the compounds in whole-embryo culture (WEC) medium, we analyzed the concentrations at the start and end of the culture period. Analytics were performed at the Department of Internal Medicine, Nephrology, Charite´ – Universita¨tsmedizin, Berlin, by using HPLC as described elsewhere (Czock et al. 2007). Animals housing and mating Wistar rats (Bor: Wistar Unilever/spf, TNO; Harlan-Winkelmann GmbH, Borchen, Germany) were kept under conventional conditions at a constant day/night cycle. They received a standard rodent laboratory diet (2018X Teklad Global, 18 % Protein Extruded, Rodent Diet, Harlan

Arch Toxicol (2013) 87:361–370

Laboratories, Rossdorf, Germany) and tap water ad libitum. Three female animals were caged with one male for 2 h in the early-morning hours. After mating, the following 24 h were called day 0 of gestation. Whole-embryo culture (WEC) Our laboratory uses the rat whole-embryo culture method as described in detail in earlier publications (Klug et al. 1985; Flick and Klug 2006; Eckardt and Stahlmann 2010). Briefly, pregnant rats were decapitated on day 9.5 of gestation, and the embryos were prepared under aseptic conditions. Four embryos with a developmental stage of 3–4 somites were cultured in one glass bottle with 7 ml sterilefiltered culture medium for 48 h. The medium was composed of 85 % serum (90 % bovine serum purchased from Quad Five, product code 962, Ryegate, Montana and 10 % rat serum from our laboratory pool) plus 15 % Hanks’ balanced salt solution, 75 mg/l methionine and 1.57 g/l glucose. At the initiation of the culture and after 36 h, the bottles were aerated with a defined gas mixture of O2, CO2 and N2. For all experiments described in this publication, a total number of 147 embryos were cultured. First, we applied the same concentrations of MPA (0.1; 0.25; 0.5; 0.75 mg/l) that had been previously tested in our laboratory (Eckardt and Stahlmann 2010). In addition to the routine evaluation, we studied the embryos by whole immunostaining and electron microscopy. We also tested the metabolite of MPA, MPAG, at concentrations of 3, 10, 30 and 100 mg/l. Control embryos were cultivated in a medium with the solvent (0.1 % methanol). Effects of the glucuronide were assessed by routine evaluation as described below. Finally, we studied the combination of these two substances. MPAG at the no observed adverse effect concentration (NOAEC = 3 mg/l) was combined with the parent compound at three concentrations (0.03, 0.1 and 0.25 mg/l), which caused no or moderate effects on growth and differentiation of the embryos.

363

The frequency of dysmorphic alterations in embryos of one treatment group is given in absolute number. Statistical analyses The statistical analyses were performed with SPSS version 16.0. In consideration of the type of variables, different analyses were performed to identify differences between test groups and controls: the crown-rump length and protein content were evaluated with ANOVA (analysis of variance) and Dunnett’s T3 post hoc test. The morphological score and number of somites were analyzed with the Mann–Whitney U test, and the yolk sac circulation and frequency of dysmorphogenesis were evaluated with the Kruskal–Wallis test and Fisher’s exact test, respectively. Immunostaining of 2H3 neurofilament Immunostaining of 2H3 neurofilament with a specific antibody has been described by Mark and co-workers (Mark et al. 1993). We evaluated cranial nerves of cultured rat embryos, which were exposed to various concentrations of MPA. After 48 h of culture, embryos were fixed in 2 % paraformaldehyde dissolved in phosphate-buffered saline (PBS) overnight. The next morning, the embryos were washed in 1 % FCS-PBS solution and dehydrated in an ascending methanol series (25 %, 50 %, 75 % dissolved in PBS and 100 % methanol). The dehydrated embryos were kept at -20 °C until staining. After thawing, rehydrating and several wash steps in 1 % FCS-PBS solution, the embryos were incubated with the primary antibody (2H3 concentrate, developmental studies Hybridoma Bank) for 24 h at 4 °C. Subsequently, the embryos were washed once again and incubated with the secondary antibody (Peroxidase-conjugated AffiniPure Goat Anti-Mouse IgG, Dianova, Jackson Immuno Research Laboratories) for 20 h at 4 °C. The staining was carried out with the substrate 4-chloro1-naphthol and the cranial nerves were evaluated and photo-documented with a stereomicroscope.

WEC data analyses

Transmission electron microscopy

For routine analysis of the embryos at the end of the culture, we used the method described by Klug and co-workers (Klug et al. 1985; Flick and Klug 2006). The WEC data are presented as mean values ± standard deviation. The yolk sac blood circulation is expressed as score points (I–III), and the crown-rump length is given in millimeters. The number of somites was also noted, and all other endpoints (embryonic rotation, neurulation, developmental stage and morphology of head, otic and optic vesicles, heart, fore and hind limbs anlagen, tail anlage and haematopoiesis) were summarized as morphological score.

After culture, the control and exposed embryos (0.1, 0.175, 0.25, 0.75 mg MPA/l) were separated in head, rump and tail and fixed in Karnovsky’s solution. Thereafter, they were washed repeatedly in 0.1 M phosphate buffer and postfixed in 1 % OsO4 solution. Before the embryos were embedded in EPONTM resin (diglycidyl ether of bisphenol F), they were dehydrated in an ascending ethanol series. After polymerization for 3 days, we prepared ultrathin slices with an Ultracut E (Reichert, Darmstadt, Germany). The slices were stained with 2 % uranyl acetate and lead citrate, and the ultrastructure was evaluated under a

123

364

transmission electron microscope (TEM 10, Zeiss, Jena, Germany). Target enzyme activity The activity of the enzyme IMPDH in cultured embryos was measured with a method that has been developed for detection in lymphocytes, thus enabling monitoring of the pharmacological action of MPA in transplanted patients. Because we were interested in changes in the enzyme activity in embryonic tissue over the relevant period of embryonic development, we measured the activity several times at day 10, day 10.5, day 11 and day 11.5 with and without yolk sac. Furthermore, we compared the IMPDH activity measured in embryos during development in utero with that of cultured embryos. Four embryos of identical developmental stage were pooled and frozen at -20 °C. After thawing, the embryos were homogenized in 250 ll water (HPLC grade) and frozen at -80 °C. Disrupted cells and insoluble fragments were removed by centrifugation at 1,000g for 2 min. The supernatant was transferred to a fresh reaction tube and used for enzyme assay. The IMPDH assay quantifies the metabolic turnover of inosine monophosphate (IMP) to xanthosine monophosphate (XMP) normalized to the AMP content of the sample and has been described in detail elsewhere (Glander et al. 2009).

Results Whole-embryo culture At a concentration of 0.25 mg MPA/l medium, the drug induced moderate alterations in some cultured embryos and in yolk sac circulation. Nearly all embryos showed pronounced effects and impaired development after exposure to 0.5 or 0.75 mg MPA/l medium. Results of the routine evaluation are shown in Table 1. Embryos cultivated at a concentration of 3 mg MPAG/l medium showed no changes in growth and development. Two of eight embryos exhibited dysmorphic changes at 10 mg MPAG/l medium and one of eight embryos at 30 mg MPAG/l medium. It is imperative to differentiate these alterations in the development of embryos from teratogenic effects in higher concentrations. The embryos showed minor growth retardation and an open cranial neuropore. The caudal neuropore was closed. Based on our evaluation score system and the normal neurulation in embryos, we characterized this alteration as dysmorphic changes. At a concentration of 100 mg MPAG/l medium, the differences were statistically significant in comparison with the control embryos. All embryos had malformations in head, truncated torso, deformed and fused branchial

123

Arch Toxicol (2013) 87:361–370

arches, and missing optic and otic vesicles. Embryos that were exposed to MPA plus MPAG showed no alterations in growth and development in comparison with the embryos that were exposed to MPA alone. Only one of each group showed an open cranial neuropore. All results of the routine evaluation of embryos are shown in Table 1 and Fig. 1. Analysis of the test compounds in WEC medium by HPLC is presented in Table 2. Evaluation of cranial nerves after 2H3 neurofilament staining To visualize the development of cranial nerves, we used a specific antibody against the 2H3 neurofilaments and stained MPA-treated embryos. The trigeminal (V), facial (VII), glossopharyngeal (IX) and vagus (X) nerves were stained and evaluated. After 48 h of culture, the trigeminal ganglion of control embryos is differentiated in an ophthalmic branch and a few ventral nerve fibers. The developmental stage of the facial nerve showed a ganglion and fibers which had reached the second branchial arch. Below the otic vesicle, the glossopharyngeal ganglion was visible with dorsal and ventral fine rootlets. A few thin fibers of the vagus nerve were detectable. All embryos cultivated in a medium with 0.1 mg MPA/l could not be distinguished from control embryos. At the next higher concentration of MPA, 0.25 mg/l, the development of the facial nerve was impaired. Furthermore, no fibers of the vagus nerve could be observed. The development of cranial nerves in embryos that were exposed to 0.5 or 0.75 mg/l was strongly affected: only thin root fibers were present above the trigeminal ganglion. The ganglia of the facial and glossopharyngeal nerve were visible, but the vagus nerve was absent (Fig. 2). Transmission electron microscopic evaluation Cells of embryos from the control group were embedded in loosely disturbed tissue, and a lot of pseudopodia were observed between the cells. Characteristic of embryonic cells is a thin cytoplasmic layer around the large nucleus with much loosely packed functionally active euchromatin. No differences could be observed between the cells from embryos exposed to 0.1 mg MPA/l and control embryos. In embryos exposed to 0.25 mg MPA/l, enhanced intercellular spaces between the embryonic cells were observed. Also, cellular as well as matrix alterations were detectable by electron microscopy at this concentration. Typical findings were degenerative changes such as irregular structure of cytoplasm. The rate of inactive heterochromatin in the nucleus increased. At the highest concentrations tested, 0.5 mg MPA/l and 0.75 mg MPA/l, the embryonic cells formed lipid vesicles and the rate of heterochromatin further increased. Sporadically, we observed

Arch Toxicol (2013) 87:361–370

365

Table 1 Effects of mycophenolic acid, mycophenolic acid glucuronide and combinations of both substances on rat embryos in vitro Approach

Concentration (mg/l)

(n)

Yolk sac circulation

Number of somites

Crown-rump length (mm)

Protein content (lg/embryo)

Score

Number of dys. embryos

MPA Control

6

3.0 ± 0.0

24.4 ± 1.1

2.96 ± 0.2

35.4 ± 0.9

0

0.1

8

3.0 ± 0.0

25.0 ± 0.9

2.75 ± 0.5

33.6 ± 4.5

2

0.25

8

2.3 ± 0.5**

24.7 ± 1.2

2.79 ± 0.4

31.6 ± 5.0

2

0.5

8

2.0 ± 0.5**

25.0 ± 1.0

2.45 ± 0.4

#

25.0 ± 5.4**

6*

0.75

8

1.4 ± 0.5**

23.5 ± 0.7

2.37 ± 0.4*

23.0 ± 6.6**

7**

MPAG Control

7

2.9 ± 0.4

25.3 ± 1.4

2.91 ± 0.8

143.7 ± 12.1

35.7 ± 1.2

0

3

7

3.0 ± 0.0

25.7 ± 0.6

3.38 ± 0.1

n.d.

38.6 ± 0.5

0

10

8

3.0 ± 0.0

23.9 ± 1.6

2.78 ± 0.2

135.3 ± 18.4

35.4 ± 3.9

2

30

7

2.7 ± 0.5

26.2 ± 1.2

3.01 ± 0.2

137.6 ± 27.7

34.6 ± 3.1

1

100

8

1.5 ± 0.5**

n.d.

2.21 ± 0.3**

68.3 ± 13.7**

24.3 ± 2.1**

8**

MPA ? MPAG Control

6

3.0 ± 0.0

25.3 ± 0.5

3.00 ± 0.1

140.1 ± 18.2

34.0 ± 3.9

0

0.03 ? 3

8

3.0 ± 0.0

25.0 ± 0.8

2.99 ± 0.2

127.6 ± 36.5

32.1 ± 3.1

1

0.1 ? 3

8

3.0 ± 0.0

25.7 ± 0.8

2.94 ± 0.2

158.5 ± 23.0

35.4 ± 2.1

1

0.25 ? 3

8

3.0 ± 0.0

25.1 ± 1.0

2.91 ± 0.2

144.0 ± 29.3

34.4 ± 1.2

1

Data are given as mean values ± standard deviation n.d. no data ** p \ 0.05 ** p \ 0.01 #

Embryos were used for whole immunostaining

large vesicles, probably apoptotic bodies, in aggregated cells (Fig. 3a–d). IMPDH activity The IMPDH activity can be calculated from the concentrations of AMP and XMP. The activity is denoted as the number of moles of XMP produced per second per mole of detected AMP (lM s-1 mol-1 AMP), and data are presented as mean values ± standard deviation (n = 3 per group). Embryos with yolk sac that matured in utero showed a slight increase in enzyme activity during the observation period, whereas the activity of cultured embryos decreased. Figure 4 illustrates the IMPDH activity of both groups, cultured embryos compared with embryos that developed in utero. The following activity was measured in four groups of embryos with yolk sac during in vitro and in utero development after a culture period of 12, 24, 36 and 48 h, respectively: in vitro development: 73 ± 41, 64 ± 6, 62 ± 30, 20 ± 11 lM s-1 mol-1 AMP and in utero development: 170 ± 16, 217 ± 31, 235 ± 27, 228 ± 31 lM s-1 mol-1 AMP. Embryos without yolk sac that matured in utero showed a pronounced increase in enzyme activity until the end of

the observation period. The IMPDH activity of embryos that matured in vitro slightly increased within the first 12 h and than decreased. The enzyme activity was measured in six groups of embryos without yolk sac during in vitro or in utero development, as illustrated in Fig. 5 (in vitro development: 105 ± 18, 152 ± 14, 73 ± 56 lM s-1 mol-1 AMP and in utero development: 56 ± 24, 134 ± 11, 267 ± 47 lM s-1 mol-1 AMP).

Discussion Mycophenolate is a widely used immunosuppressive agent in organ transplant patients. Routinely performed teratogenicity tests with pregnant rats and rabbits showed increased incidences of malformations, including hydrocephaly, anophthalmia, diaphragmatic hernia and abortions at doses, which lead to plasma concentrations in the therapeutic range (Tendron et al. 2002; Koren 2008). Besides these findings, some data indicate that mycophenolate is also a human teratogen. The National Transplantation Pregnancy Registry gathered pregnancy outcomes in female patients after organ transplantation and intake of MPA during gestation. Within this series of cases, 49 % of all examined pregnancy outcomes were spontaneous abortions.

123

366

Arch Toxicol (2013) 87:361–370

MPAG

Control

10 mg/l

30 mg/l

100 mg/l

MPA + MPAG

Control

0.03 + 3 mg/l

0.1 + 3 mg/l

0.25 + 3 mg/l

Fig. 1 Rat embryos prepared on day 9.5 of gestation after a 48-h culture period, exposure to mycophenolic acid glucuronide (MPAG) and a combination of mycophenolic acid (MPA) and the glucuronide. The embryos after exposure to 3 mg MPAG/l medium show no alteration compared with control embryos. Only slight changes in crown-rump length and protein content was visible in embryos exposed to concentrations of 10 mg MPAG/l and 30 mg MPAG/l.

Embryos that were exposed to the highest concentration showed malformations in head, truncated torso, deformed and fusioned branchial arches as well as missing eye and otic vesicles probably due to MPA present in the MPAG which was used in the experiments. The combination of mycophenolic acid at concentrations up to 0.25 mg/l plus its glucuronide revealed no enhanced effect compared with the parent compound alone

Table 2 Concentrations of MPAG in medium used for the whole-embryo culture Nominal concentration MPAG (mg/l)

Concentrations of MPAG measured by HPLC (mg/l) Start of culture

End of culture

0

–

–

3

2.9

2.8

10

11.4

10.4

30

35.7

34.0

100

n.d.

119.5a

– MPAG not detectable n.d. no data a

At this concentration, we also found mycophenolic acid at a concentration of 1 mg/l medium

Birth defects such as microtia and facial deformities were also observed (Nguyen et al. 2010). As a consequence of these observations, the FDA classified the drug in group ‘‘D’’. However, because the mothers were treated with mycophenolate in combination with other immunosuppressive drugs, the causal relationship remains obscure in the individual cases. Two mycophenolate drugs are available, enteric-coated mycophenolate sodium and mycophenolate mofetil, which

123

is hydrolyzed to MPA by esterases in the gut wall, blood, liver and other tissues. MPA is metabolized in the liver, gastrointestinal tract and kidney by uridine diphosphate gluconosyltransferases (UGTs). 7-O-MPA-glucuronide (MPAG) is the major metabolite of MPA, but at least three minor metabolites are also formed. MPAG is usually present in the plasma at 20- to 100-fold higher concentrations than MPA, but it exhibits no immunosuppressive activity (Bullingham et al. 1998; Staatz and Tett 2007).

Arch Toxicol (2013) 87:361–370

367

Trigeminal nerve

Control 0.1 % DMSO

}

Ganglion with ophthalmic branch and a few ventral nerves fibers

F ac ial nerve Ganglion and fibers, that have reached the second branchial arch

} } }

G los s opharyngeal nerve Ganglion was visible with dorsal and ventral fine rootlets

Vagus nerve A few thin fibers

0.25 mg MPA/l medium

0.75 mg MPA/l medium A

a

B

b

C

c

D

d

Fig. 2 Whole immunostaining with rat embryos after a 48-h culture period and exposure to the solvent or mycophenolic acid. The cranial nerves were stained with a specific antibody against 2H3 neurofilaments. At a concentration of 0.25 mg/l medium, development of nerves anlagen was impaired. Pronounced alterations in nerve development could be observed at 0.75 mg MPA/l medium. A No alterations in development of trigeminal nerve. B Developmental stage of facial nerve is reduced, ganglion is visible and only a few

ventral nerve fibers. C Developmental stage of glossopharyngeal nerve is reduced, only ganglion is visible. D The vagus nerve is undeveloped. a Pronounced alterations in development of trigeminal nerve, ganglion and a few thin fibers of ophthalmic nerve. b Pronounced alterations in development of facial nerve, only a few nerve fibers of ganglion are visible. c Pronounced alterations in development of glossopharyngeal nerve, only a few nerve fibers of ganglion are visible. d The vagus nerve is not developed

Nothing is known about a possible teratogenic potential of the glucuronide. We used the whole-embryo culture assay to study the effects of MPA and MPAG on rat embryos in vitro during the period of organogenesis and found that under these conditions, the glucuronide is much less potent than the parent compound. The results of our experiments with MPA using the routine approach for evaluation are in excellent agreement with results obtained earlier in our laboratory, underlining the reliability and reproducibility of the whole-embryo culture (Eckardt and Stahlmann 2010). In this new experimental series, MPA caused a strong dysmorphogenic development (ICMax) at a concentration of 0.75 mg/l, whereas MPAG did not induce such effects at concentrations below 100 mg/l medium. We observed similar alterations in embryos exposed to 1 mg MPA/l or 100 mg MPAG/l medium. In both treatment groups, embryos showed very similar dysmorphogenic changes in

the head area such as missing eye and otic vesicles. However, these findings with MPAG are probably caused by a small amount of MPA in the MPAG material used for the experiments. Analysis of chemical purity by the supplier shows that MPAG contains 0.6 % MPA; a concentration of 1.0 % was detected when we analyzed the medium at the end of the culture period with a nominal concentration of 100 mg MPAG/l medium. For a routine evaluation of cultured embryos, various score systems exist. Our laboratory uses the evaluation system that was developed by Klug and co-workers. This score system is used for the evaluation of growth, morphological differentiation and malformation of cultured embryos (Flick and Klug 2006). With this approach, it is possible to detect and evaluate gross morphological alterations in the embryos after a 2-day culture period. To obtain more detailed information on the effects of the test substances, we extended the methods of evaluation. The development of cranial nerves,

123

368

Arch Toxicol (2013) 87:361–370

Fig. 3 Rat embryonic cells of the head region after 48-h culture at developmental stage day 11.5. a Control group treated with 0.1 % DMSO. Loosely arranged embryonic cells with a huge nucleus, a thin stripe of cytoplasm perinuclear and a lot of pseudopodia. b Treated with 0.25 mg MPA/l medium for 48 h in culture. Cells exhibit degenerative changes that are particularly visible in irregular structure of cytoplasm and in increasing rate of inactive heterochromatin. c Treated with 0.5 mg MPA/l medium during 48-h culture. The

embryonic cells aggregated, and we partially observed large vesicles, which probably are apoptotic bodies. d Treated with 0.75 mg MPA/l medium during 48-h culture. At this concentration, the intensity of degenerative changes has increased significantly. Cy Cytoplasm, Ps Pseudopodia, N Nucleus, Nu Nucleolus, (Ellipse) irregular structure of cytoplasm, (Ellipse, dashed line) aggregated cells with large vesicles, (Star) nucleus with high content of euchromatin, (Asterisk) nucleus with increased content of heterochromatin

as well as the ultrastructural changes in MPA-exposed embryos, was examined. With the conventional dysmorphology score system by Klug and co-workers, we found that MPA induced slight embryotoxic effects already at a concentration of 0.25 mg/l medium. In addition, we used the immunostaining approach to study the development of the cranial nerve anlagen, but could not detect specific effects of MPA on the development of the nervous system at relevant concentrations. Embryos showed various degrees of impairment of the cranial nerve fibers, but we found no evidence that the effects on the nerval structures occur at lower concentrations than other dysmorphogenic effects. On an ultrastructural level, we observed degenerative cellular changes and signs of apoptosis, which were pronounced with increasing concentrations of MPA in culture medium. Previous studies suggested a potential of MPA to induce apoptosis in non-

embryonic cells. Besides inhibition of lymphocyte proliferation, MPA also has the potential to induce apoptosis in activated lymphocytes and insulin-producing cells (Kim et al. 2008; Park et al. 2010). Overall, with both additional methods, whole immunostaining and transmission electron microscopy, the LOAEC of 0.25 mg MPA/l has been confirmed. Thus, it was not possible to increase the sensitivity of this assay by these additional methods. A combination of MPA and the glucuronide at low concentrations did not increase the rate of dysmorphic embryos. Apparently, the metabolite does not displace enough MPA from the binding protein that it would influence the results of toxicity in this in vitro assay (Table 1). MPA is a potent, selective and reversible inhibitor of IMPDH, leading to eventual arrest of T- and B-lymphocyte proliferation. IMPDH catalyzes the oxidation of inosine-50 monophosphate (IMP) to XMP, a precursor of guanosine

123

Arch Toxicol (2013) 87:361–370

Fig. 4 Inosine-50 -monophosphate dehydrogenase (IMPDH) activity of embryos (broken line) and cultured embryos (continuous line) with yolk sac at developmental stages day 10, 10.5, 11, 11.5 post coitum (n = 3 per group). The activity of IMPDH in cultured embryos is lower than in embryos after development in utero. After 48-h culture, the embryos exhibited approximately 10 % of the enzyme activity of embryos that matured in utero

Fig. 5 Inosine-50 -monophosphate dehydrogenase (IMPDH) activity of embryos (broken line) after intrauterine growth and cultured embryos (continuous line) at developmental stages day 10.5, 11, 11.5 post coitum (n = 3 per group). The activity of IMPDH increased considerably in embryos grown in utero from 56 ± 24 to 267 ± 47 lM s-1 mol-1 AMP, whereas in cultured embryos, the activity showed only minor changes

monophosphate (GMP). A depletion of guanosine can minimize organ rejection after transplantation because lymphocyte proliferation depends on guanosine de novo synthesis (Winnicki et al. 2010). It is unknown by which mechanism MPA acts as a teratogen, because it has not been studied under this aspect. The immunosuppressive effect of MPA is mediated by inhibition of IMP dehydrogenase, which could be a target structure for MPA-induced dysmorphogenesis in embryos as well. In human tissues, two very closely related IMPDH isoenzymes exist, type I and type II. A comparison of gene expression of IMPDH types in various cell types shows that the expression of type II is

369

affected by cell proliferation and transformation and is increased in replicating and neoplastic cells (Gu et al. 2000). In five types of human fetal tissues (heart, brain, lung, liver and kidney), IMPDH activity was detected (Senda and Natsumeda 1994). It is possible that the development of embryonic tissues exhibiting a high proliferation rate is impaired if this enzyme is inhibited. A knockout mouse model reveals that a homozygous deletion of type II gene results in high embryonic lethality (Gu et al. 2000). We observed that the enzyme is active in the embryonic tissue over the entire culture period. The enzyme activity of embryos after in utero development was higher than in embryos, which developed in culture. It is known that IMPDH activity depends on AMP levels in cells. An increased enzyme activity in embryos that matured in utero is possibly based on optimal control in metabolism and evacuation of metabolites. On the basis of these results, further studies are reasonable to investigate whether a correlation exists between an inhibition of the IMPDH activity in the embryonic tissue and the induction of dysmorphogenic embryos. In addition, several other possible mechanisms of the embryotoxic effects have to be discussed. MPA can induce apoptosis of activated T lymphocytes by depleting guanosine nucleotides, and it suppresses glycosylation as well as the expression of some adhesion molecules. It has also been shown that by depleting guanosine nucleotides, MPA depletes tetrahydrobiopterin, a cofactor for iNOS, the inducible form of nitric oxide synthase (Allison and Eugui 2000). In summary, we have shown that 9.5-day-old rat embryos exposed for 48 h to MPA at concentrations that are considerably lower than those achieved in plasma of patients treated with this drug exhibit alterations that roughly correspond to teratogenic effects observed in animal studies and human newborns after in utero exposure. Using immunostaining of cranial nerves anlagen and electron microscopy, we obtained more detailed information on the effects on a microscopic and on an ultrastructural level. Activity of the target enzyme, IMPDH, was found in 10-day-old embryos, and it declined during the culture period. Adding the MPAG to MPA did not increase embryotoxicity under our experimental conditions. Acknowledgments We thank Irmela Baumann-Wilschke for her excellent technical assistance and Heidi Pretorius for her help with the preparation of the manuscript. Data presented in this publication are part of the Master Thesis of Flavia Schmidt, prepared during the Master Educational Program of Toxicology at Charite´—Universita¨tsmedizin Berlin.

References Allison AC, Eugui EM (2000) Mycophenolate mofetil and its mechanisms of action. Immunopharmacology 47:85–118

123

370 Anderka MT, Lin AE, Abuelo DN, Mitchell AA, Rasmussen SA (2009) Reviewing the evidence for mycophenolate mofetil as a new teratogen: case report and review of the literature. Am J Med Genet 149:1241–1248 Augustine-Rauch K, Zhang CX, Panzica-Kelly JM (2010) In vitro developmental toxicology assays: a review of the state of the science of rodent and zebrafish whole embryo culture and embryonic stem cell assays. Birth Defects Res C Embryo Today 90:87–98 Bullingham RE, Nicholls AJ, Kamm BR (1998) Clinical pharmacokinetics of mycophenolate mofetil. Clin Pharmacokinet 34: 429–455 Czock D, Rasche FM, Carius A, Glander P, Budde K, Bauer S, Keller F, von Mu¨ller L (2007) Pharmacokinetics and pharmacodynamics of mycophenolic acid after enteric-coated mycophenolate versus mycophenolate mofetil in patients with progressive IgA nephritis. J Clin Pharmacol 47:850–859 Eckardt K, Stahlmann R (2010) Use of two validated in vitro tests to assess the embryotoxic potential of mycophenolic acid. Arch Toxicol 84:37–43 Flick B, Klug S (2006) Whole embryo culture: an important tool in developmental toxicology today. Curr Pharm Des 12:1467–1488 Genschow E, Spielmann H, Scholz G, Seiler A, Brown N, Piersma A, Brady M, Clemann N, Huuskonen H, Paillard F, Bremer S, Becker K (2002) The ECVAM international validation study on in vitro embryotoxicity tests: results of the definitive phase and evaluation of prediction models. European Centre for the Validation of Alternative Methods. Altern Lab Anim 30:151– 176 Glander P, Budde K, van Gelder T, Hambach P, Liefelt L, Lorkowski C, Mai M, Neumayer HH, Vulto AG, Mathot RA (2009) Improved assay for the nonradioactive determination of inosine 50 -monophosphate dehydrogenase activity in peripheral blood mononuclear cells. Ther Drug Monit 31:351–359 Gu JJ, Stegmann S, Gathy K, Murray R, Laliberte J, Ayscue L, Mitchell BS (2000) Inhibition of T lymphocyte activation in mice heterozygous for loss of the IMPDH II gene. J Clin Invest 106:599–606 Kim JY, Huh KH, Park Y-J, Fang Y, Kang CM, Kim YS (2008) Molecular mechanisms of cell death of mycophenolic acid-treated primary isolated rat islets: implication of mitogen-activated protein kinase activation. Transplant Proc 40:2575–2577 Klug S, Lewandowski C, Neubert D (1985) Modification and standardization of the culture of early postimplantation embryos for toxicological studies. Arch Toxicol 58:84–88 Koren G (2008) Mycophenolate mofetil: emerging as a potential human teratogen. Can Fam Physician 54:1112–1113

123

Arch Toxicol (2013) 87:361–370 Lin EA, Singh EK, Strauss A, Nguyen S, Rawson K, Kimonis EV (2011) An additional patient with mycophenolate mofetil embryopathy: cardiac and facial analyses. Am J Med Genet 155:748–756 Mark M, Lufkin T, Vonesch JL, Ruberte E, Olivo JC, Dolle´ P, Gorry P, Lumsden A, Chambon P (1993) Two rhombomeres are altered in hoxa-1 mutant mice. Development 119:319–338 Merlob P, Stahl B, Klinger G (2009) Tetrada of the possible mycophenolate mofetil embryopathy: a review. Reprod Toxicol 28:105–108 Neumann I, Fuhrmann H, Kanzler M, Fang IF, Jaeger A, Graf H, Bayer P, Kovarik J (2008) Pharmacokinetics of enteric-coated mycophenolate sodium: comparative study in patients with autoimmune disease and renal allograft. Expert Opin Pharmacother 9:879–886 New DA (1976) Techniques for assessment of teratologic effects: embryo culture. Environ Health Perspect 18:105–110 Nguyen TDK, Carlin F, Coscia LA, Armenti D, Armenti VT (2010) Pregnancy outcomes with exposure to the mycophenolic acid products (MPA). Department of Surgery Faculty Papers & Presentations. Paper 27. (URL: http://jdc.jefferson.edu/surgeryfp/ 27, Date: 30.01.12) Park Y-J, Ahn HJ, Kim YS, Cho Y, Joo DJ, Ju MK (2010) Illuminamicroarray analysis of mycophenolic acid-induced cell death in an insulin-producing cell line and primary rat islet cells: new insights into apoptotic pathway involved. Cell Signal 22:1773– 1782 Perez-Aytes A, Ledo A, Boso V, Sa´enz P, Roma E, Poveda JL, Vento M (2008) In utero exposure to mycophenolate mofetil: a characteristic phenotype? Am J Med Genet A 146:1–7 Senda M, Natsumeda Y (1994) Tissue-differential expression of two distinct genes for human IMP dehydrogenase. Life Sci 54:1917– 1926 Sifontis NM, Coscia LA, Constantinescu S, Lavelanet AF, Moritz MJ, Armenti VT (2006) Pregnancy outcomes in solid organ transplant recipients with exposure to mycophenolate mofetil or sirolimus. Transplantation 82:1698–1702 Staatz CE, Tett SE (2007) Clinical pharmacokinetics and pharmacodynamics of mycophenolate in solid organtransplant recipients. Clin Pharmacokinet 46:13–58 Tendron A, Gouyon JB, Decramer S (2002) In utero exposure to immunosuppressive drugs: experimental and clinical studies. Pediatr Nephrol 17:121–130 Winnicki W, Weigel G, Sunder-Plassmann G, Bajari T, Winter B, Herkner H, Sengoelge G (2010) An inosine 50 -monophosphate dehydrogenase 2 single-nucleotide polymorphism impairs the effect of mycophenolic acid. Pharmacogenomics J 10:70–76