Localization of Monoamine Oxidase mRNA in Human Placenta

Abstract

Monoamine oxidase (MAO) oxidatively deaminates vasoactive and biogenic amines and exists in two distinct forms (A and B), coded for by separate genes, which exhibit distinct substrate specificities and inhibitor sensitivities. Using specific primers for MAO-A and MAO-B mRNA in a reverse transcription-polymerase chain reaction (RT-PCR) on RNA from human liver, the predicted products for both enzymes were detected. Furthermore, RT-PCR on RNA from human placenta, believed to contain predominantly (or only) MAO-A protein, also indicated the presence of both A and B gene transcripts. The cellular distribution of MAO mRNA in placental tissue was analyzed by in situ hybridization of MAO-A and MAO-B mRNA-specific cRNA probes on paraffin sections. MAO-A mRNA was mainly evident in the syncytiotrophoblastic layer. None was detected in the vascular endothelium/smooth muscles. Significantly, MAO-B mRNA signal was also evident in the placental villi, notably in the syncytiotrophoblasts, intermediate trophoblasts, cytotrophoblasts, and the vascular endothelium. To our knowledge, this is the first demonstration of the cellular distribution of MAO mRNA in human placenta via in situ hybridization. The expression of MAO-B in placental tissue rather than in blood elements within placenta is also unequivocally demonstrated. These highly specific cRNA probes can now be used to study the distribution of MAO-A and MAO-B expression in other tissues.

Monoamine oxidase (MAO; EC 1.4.3.4) is localized in the outer mitochondrial membrane and is a flavincontaining enzyme involved in the catalysis of oxidative deamination of several neuroactive, vasoactive, and other biogenic amines (von Kroff 1979; Chiba et al. 1984). MAO exists in two isoenzyme forms (MAOA and -B) which are distinguished by their substrate preference and inhibitor affinity (Fowler and Tipton 1984). MAO-A preferentially deaminates 5-hydroxytryptamine (serotonin), norepinephrine, and epinephrine, and is irreversibly inactivated by clorgyline (Johnston 1968). MAO-B preferentially deaminates β-phenylethylamine and benzylamine and is inhibited by deprenyl (Knoll and Magyar 1972). Monoamine oxidase-A and -B were proved to be independent entities when the genes encoding each of them were identified (Bach et al. 1988; Ito et al. 1988; Kuwahara et al. 1990).

MAO-A and -B are differentially expressed in a variety of tissues. Some, such as human liver and brain, contain both forms of MAO (Johnston 1968; Knoll and Magyar 1972), whereas others, such as human platelets (Bond and Cundall 1977) or placenta (Egashira and Yamanaka 1981; Weyler and Salach 1985), contain primarily one form (MAO-B and MAO-A, respectively). However, the expression of low levels of MAO-B in placenta has been reported, both as immunologically detectable protein and catalytic activity (Riley et al. 1989) and as mRNA (Bach et al. 1988). These observations are believed to be attributable to MAO-B expression in blood lymphocytes and platelets and in the vascular endothelium rather than the placental tissue itself (Riley et al. 1989). The presence of MAO-B mRNA in placental tissue is equivocal (Grimsby et al. 1990). In addition, the exact cellular localization of MAO mRNA is not known.

In this study we demonstrate the use of nonradioactively (digoxigenin, DIG)-labeled complementary RNA (cRNA) probes to specifically locate MAO-A and MAO-B mRNAs in placental sections via in situ hybridization histochemistry. The RNA probes were produced from cDNA clones encompassing the entire protein coding region (Bach et al. 1988), and their specificities were demonstrated by Northern blot hybridization. The MAO-A mRNA-specific probe reacts with MAO-A sense cRNA but not with MAO-B sense cRNA, and the MAO-B mRNA-specific probe reacts with only MAO-B sense cRNA. Thus, a clear picture of the cell types capable of expressing MAO in placenta has been obtained. Sequence-specific primers have also been used for detection of MAO-A and/or MAO-B expression by RT-PCR of RNA purified from human placenta and liver, to confirm the in situ hybridization findings.

Materials and Methods

Synthesis of the cRNA Probes

Plasmids (pSP65) carrying the appropriate cDNA (MAO-A, 2.5 kb for the sense and anti-sense; MAO-B, 2.5 kb and 2 kb for the sense and the anti-sense orientations, respectively) (Bach et al. 1988) were propagated in E. coli (JM83)-competent cells and purified by ethidium bromide-cesium chloride gradient centrifugation followed by phenol extraction and ethanol precipitation (Sambrook et al. 1989). Purified plasmids were then dissolved in 10 mM Tris-HCl, pH 7.6, 1 mM EDTA, and stored at -20C. Before transcription, plasmids carrying MAO-A and MAO-B cDNA were linearized using Sma I and Xba 1 (New England Biolabs; Beverly, MA), respectively. Transcription of the anti-sense and the sense (control) DIG-labeled cRNA probes was performed by incubating 1 μg of corresponding linearized plasmid in a 20-μl reaction mixture containing 40 mM Tris-HCl, 6 mM MgCl₂, 10 mM dithiothreitol, 2 mM spermidine, 10 mM NaCl, pH 8.0, 20 U RNasin ribonuclease inhibitor (Promega; Southampton, UK), 1 mM each of GTP, ATP, and CTP, 0.65 mM unlabeled UTP, 0.35 mM digoxigenin-labeled UTP (DIG-11-UTP; Boehringer Mannheim, Lewes, East Sussex, UK), and 40 U of SP6 RNA polymerase (Boehringer Mannheim) at 37C for 2 hr. The reaction was halted by digesting the template DNA with 10 U of RNase-free DNase (Promega) during further incubation for 15 min at 37C. Synthesized riboprobes were precipitated by the addition of LiCl and ice-cold ethanol (final concentrations of 100 mM and 75%, v/v, respectively), thorough mixing, and incubation for 1 hr at -70C. After centrifugation at 10,000 X g, 4C, pelleted RNA was washed twice with 75% ethanol, vacuum-dried, and dissolved in 100 μl DEPC (diethylpyrocarbonate)-treated ultrapure water. Unlabeled cRNAs were synthesized in a similar way, except that DIG-11-UTP was omitted and a total of 1 mM UTP was used. Probe length was verified by Northern blot analysis (Sambrook et al. 1989) and its specific labeling efficiency was detected using dot-blotting and immunological detection (see below).

Probes were subjected to limited alkaline hydrolysis in 0.2 M sodium carbonate-bicarbonate buffer (pH 10.2) and incubation at 60C for 20 min (Cox et al. 1984). The reaction was halted by lowering the pH using 3 M sodium acetate (pH 6) and 10% glacial acetic acid. The probe was precipitated with ethanol and resuspended in DEPC-treated water. Fragment length was checked as described above.

RNA Isolation

Placental tissues were obtained fresh from routine elective cesarean section deliveries performed at the Queens Medical Centre, Nottingham, UK, by Mr. G. M. Filshie and his team. The procedures used were in accordance with the ethical standards approved by the Ethics Committee, Queens Medical Centre. Total RNA was isolated using RNAStat RNA extraction solution (Biogenesis; Poole, UK). Briefly, tissues were homogenized in ice-cold RNAStat (1 ml/100 mg tissue). After the addition of chloroform to a concentration of 10% (v/v) and vigorous mixing, each sample was kept on ice for 5 min, followed by centrifugation (10,000 X g) for 20 min at 4C. The upper aqueous phase was then aspirated and mixed with an equal volume of isopropanol and placed on ice for 10 min. Total RNA was collected by centrifugation at 7500 X g for 20 min at 4C and the RNA pellet was vacuumdried after being washed twice with 75% ethanol-DEPC water. RNA was dissolved in DEPC-treated water containing RNasin (1 U/μl) and stored in liquid nitrogen.

Northern Blot Analysis

RNA samples were fractionated on 1% agarose-formaldehyde gels and capillary transferred onto nylon filters (Boehringer Mannheim) using 20 X SSC (saline-sodium citrate 1 X = 0.15 M NaCl, 0.015 M sodium citrate, pH 7.0). After washing with 6 X SSC, filters were allowed to air-dry. Then the transferred RNA was bound to the filter using a UV-translinker (Stratagene; Cambridge, UK). Marker lanes were removed and stained with 0.04% methylene blue. The rest of the filters were prehybridized in hybridization buffer (50% formamide, 4 X SSC, 7% SDS, 50 μg/ml denatured salmon sperm DNA) (Sigma; Poole, UK) at 55C for 4 hr. The positive control probe for Northern hybridization was DIG-labeled human β-actin anti-sense RNA probe (corresponding to bases 69-618 of β-actin) (Boehringer Mannheim), and the test probes were DIG-labeled MAO-A and MAO-B probes in the sense and anti-sense directions. The probe concentrations used were 80 ng/ml for the actin probe and 100 ng/ml for the MAO probes. Filters were then incubated in hybridization solution containing the designated probe at 55C overnight and washed as follows: 2 X SSC, 0.1% SDS for 15 min at 55C; 0.5 X SSC, 0.1% SDS for 15 min at 55C and 0.1 X SSC, 0.1% SDS for 30 min at 65C. Hybridization was detected immunologically. Filters were blocked by immersion in TBS (Tris-buffered saline: 100 mM Tris-HCl, 150 mM NaCl, pH 7.5) containing 2% (v/v) heatinactivated normal ovine serum and 0.3% (w/v) Triton X-100 for 30 min at room temperature (RT) and then incubated with alkaline phosphatase-conjugated sheep anti-digoxigenin (Fab fragment) (Boehringer Mannheim), diluted 1:1000 in the same buffer, for 30 min at RT. After two 15-min washes in TBS, the blots were equilibrated for 5 min in the detection buffer (100 mM Tris-HCl, 100 mM NaCl, pH 9.5). Alkaline phosphatase was visualized by the addition of 250 μM CSPD (disodium3-(4-methoxyspiro{1,2-dioxetane-3,2′-(5′-chloro)tricyclo{3.3.1.1^3,7]decan}-4yl) phenyl phosphate) chemiluminescent substrate (Perkin-Elmer; Warrington, UK) in detection buffer and the blots were incubated for 15 min at 37C. Signal was detected on Kodak Biomax ML film (Sigma).

In Situ Hybridization

Freshly prepared placental tissue sample blocks were fixed at RT in 4% formaldehyde in PBS for 24 hr, dehydrated, and embedded in paraffin. Sections (8 μm thick) were cut, mounted on silinated slides (Lewis and Wells 1992), dried at 45C for 1 hr, then at 37C overnight, and stored at RT. Sections were deparaffinized for in situ hybridization, rehydrated, and treated with proteinase K (10 μg/ml, 20 min at 37C), then acetylated in 0.25% acetic anhydride, 0.1 M triethanolamine-HCl, pH 8, for 10 min at RT.

Prehybridization involved incubation at 55C for 4 hr in hybridization solution (HS; 100 μl per section) containing 50% formamide, 4 X SSC, 1 X Denhardt solution (0.02% Ficoll, 0.02% polyvinyl pyrrolidone, 10 mg/ml RNase-free bovine serum albumin), 0.5 mg/ml sheared salmon sperm DNA, 0.25 mg/ml yeast tRNA, 5% dextran sulfate. After removal of the prehybridization solution, sections were hybridized by the addition of 30 μl of HS containing 0.05 μg heat-denatured probes (80C for 5 min, 4C for 10 min). Sections were covered with Parafilm, sealed with rubber cement, and incubated at 55C overnight. Hybridization fluid was then aspirated and slides washed in 2 X SSC for 5 min at RT, and 50% formamide, 2 X SSC for 30 min at 55C as the stringency wash. Filters were washed with 0.1 X SSC for 30 min at 55C to remove formamide. Signal detection was as described above, except that levamisole (0.24 mg/ml) was added to the substrate as an inhibitor of endogenous alkaline phosphatase. Color development was terminated by immersing slides in 50 mM Tris-HCl, pH 7.5, 1 mM EDTA buffer. Sections were mounted in gelatin-glycerol (Sigma), which was allowed to set at 4C.

Throughout the in situ hybridization procedure, temperature control was achieved using an Omnislide thermocycler (Hybaid).

Reverse Transcription-Polymerase Chain Reaction

Reverse transcription and PCR were performed to verify the presence or absence of MAO-A and MAO-B mRNA in hepatic and placental RNA. Sequence specific primers were selected from the full cDNA sequences of MAO-A and MAO-B (Bach et al. 1988). Primers for MAO-A were as follows: (5′ MAOA51; 3′ MAOA31), selected from exons 7 and 15, respectively. Primers for MAO-B were as follows: (5′ MAOBF1; 3′ MAOBR1), selected from exons 13 and 15 respectively, as follows:

Oligonucleotide	Orientation	Location	Sequence (5′-3′)
MAOA51	Sense	756-775	ACGGATAATGGACCTCCTCG
MAOBF1	Sense	1304-1323	ATATGGAAGGGTTCTACGCC
MAOA31	Anti-sense	151-1540	GGTGTGGGTGATTTCTACCG
MAOBR1	Anti-sense	1845-1864	AAACTGGTGAAACAGAACGC

The fragments flanked by these primers were 745 bp and 522 bp for MAO-A and MAO-B, respectively. Selection of primers was based on the search results of comparisons of the primer sequences with other human sequences using the GenBank data bases (NCBI, National Library of Medicine, National Institute of Health, Bethesda, MD), where only gene-specific sequences were selected. Reverse transcription was performed in 50 μl reaction buffer (50 mM Tris-HCl, pH 8.3, 75 mM KCl, 3 mM MgCl₂, 10 mM DTT) containing 10 pmol specific 3′ (anti-sense) primer (MAOA31, MAOBR1 for MAO-A and MAO-B, respectively); 5 μg of total RNA from either liver or placenta, 60 U RNasin ribonuclease inhibitor (Promega), 2 mM of each of dATP, dGTP, dCTP, and dTTP (Promega), and 400 U of Molony murine leukemia virus reverse transcriptase (M-MLV; Promega). The reaction mix was incubated for 1 hr at 37C and the reaction terminated by heating at 95C for 5 min. cDNA was used for PCR by adding 5 μl of RT samples to a reaction buffer (10 mM Tris-HCl, pH 9, 50 mM KCl, 0.1% Triton X-100, and 3 mM MgCl₂) containing 0.5 mM each dATP, dGTP, dCTP, and dTTP (Promega), 0.75 μM specific 3′ and 5′ primers (MAOA31, MAOA51; MAOBF1, MAOBR1 for MAO-A and MAO-B respectively) and 2.5 U Taq DNA polymerase (Promega) in a final volume of 50 μl. The reaction mix was overlaid with 25 μl mineral oil (Sigma) and heat-denatured for 7 min at 95C, followed by 30 cycles of annealing (56C, 1 min), elongation (72C, 2 min), and denaturation (95C, 45 sec), followed by a final elongation at 72C for 7 min. The PCR products were electrophoresed on 1% agarose in 89 mM Tris-borate, pH 8, 2 mM EDTA buffer, and stained with ethidium bromide (Sambrook et al. 1989).

Results

Northern Hybridization

Total RNA from human liver was used to confirm the specificities of the MAO-A and MAO-B probes. Single bands of 4.9 kb were evident when the anti-sense probes of MAO-A and MAO-B were used (Figure 1A, Lanes 2 and 6), whereas no bands were revealed with the sense probes (Figure 1A, Lanes 4 and 8). Using MAO-B sense cRNA as a template, the MAO-B anti-sense probe, but not the MAO-A probe, hybridized to a 2.0-kb transcript, which corresponds to the size of the cDNA insert that was used as template for the synthesis of MAO-B cRNA (Figure 1B, Lanes 2 and 4). Similarly, when MAO-A cRNA was used as a template, only the MAO-A anti-sense probe revealed a product of 2.4 kb (Figure 1B, Lanes 3 and 5). Therefore, the specificity of the probes appeared to be conclusive. Analysis of total RNA purified from fresh placental tissue also indicated hybridization with single bands of 4.9 kb with both MAO-A and MAO-B anti-sense probes (Figure 1A, Lanes 3 and 7), but no hybridization when the MAO-A or MAO-B sense probe was used (Figure 1A, Lanes 5 and 9). With both placental and liver RNA, the 4.9-kb band detected with the MAO-A anti-sense probe was weak and there was a strong signal at the top of the gel (Figure 1A, Lanes 2 and 3).

Figure 1

Northern blot analysis of (A) total RNA (10 μg/lane) purified from human liver (Lanes 2, 4, 6, 8, and 10) and placenta (Lanes 3, 5, 7, 9, and 11). MAO-A anti-sense probe was used in Lanes 2 and 3 and MAO-B anti-sense probe was used in Lanes 6 and 7. Lanes 4 and 5 and Lanes 8 and 9 represent MAO-A sense, and MAO-B sense, respectively. Human β-actin antisense RNA probe was used for hybridization on lanes 10 and 11. (B) MAO-B cRNA (0.5 μg, Lanes 2 and 4) and MAO-A cRNA (0.5 μg, Lanes 3 and 5) probed with anti-sense MAO-B (Lanes 2 and 3) and MAO-A (Lanes 4 and 5) probes. Lane 1 shows the RNA size marker (Promega; 6.58 kb, 4.98 kb, 3.64 kb, 2.6 kb, 1.9 kb, 1.4 kb, 0.955 kb, and 0.62 kb).

Similar signal patterns were obtained for MAO-A and MAO-B transcripts in placental and hepatic RNA when specific digoxigenin-labeled anti-sense oligonucleotide probes were used instead of RNA probes (data not shown).

As expected, the human β-actin anti-sense DIG-labeled RNA probe revealed a single band of approximately 1.8 kb for liver and placenta (Figure 1A, Lanes 10 and 11).

Reverse Transcription-Polymerase Chain Reaction

The RT-PCR revealed products from MAO-A and MAO-B genes in both liver and placenta. As predicted, a 745-bp product was generated for MAO-A from hepatic and placental RNA, similar to that produced by the unlabeled MAO-A (sense) cRNA when used as template for the RT-PCR (Figure 2, Lanes 2–4). Similarly, the products for MAO-B from hepatic and placental RNA, were 522 bp and, as expected, were similar to that produced when the unlabeled MAO-B (sense) cRNA was used as template (Figure 2, Lanes 5-7). It is recognized that quantitative conclusions are difficult to make using RT-PCR but, if one takes into account the fact that the efficiency of amplification of MAO-B cRNA is greater than that of MAO-A cRNA (equal amounts of the cRNAs having been used), it appears that the relative amounts of the A and B forms of mRNA are similar in liver. Furthermore, the amount of MAO-A mRNA is greater than MAO-B mRNA in placenta, but the latter is definitely present. These results also suggest that the signals observed with the MAO-A-specific RNA probes at the top of the gel in Figure 1A are real, i.e., for some reason MAO-A mRNA, but not MAO-B mRNA, is relatively insoluble.

Finally, these results give further confidence in the use of the RNA probes for in situ hybridization experiments.

In Situ Hybridization

The anti-sense probes for both MAO-A and MAO-B gave signals in paraffin-embedded human liver sections, with the MAO-B mRNA predominating, and the sense probes giving no signals (data not shown).

Figure 2

Sequence-specific MAO-A or MAO-B anti-sense primers were used for reverse transcription of human liver and placental (total) RNA. Sense cRNAs, previously transcribed from full-length MAO-A or MAO-B cDNAs (Bach et al. 1988), were used as positive controls. RT products were amplified by PCR and revealed on ethidium bromide-stained agarose gels after electrophoresis. Lanes 2—4 show the predicted 745-BP products when MAO-A-specific primers were used for liver RNA, placental RNA, and MAO-A cRNA, respectively. Lanes 5-7 show the predicted 522-BP products when MAO-Bspecific primers were used for liver RNA, placental RNA and MAO-B cRNA, respectively. Lane 1, DNA molecular weight markers (Promega; 1.35 kb, 1.078 kb, 0.872 kb, 0.603 kb, 0.310 kb, 0.281 kb, 0.271 kb, 0.234 kb, 0.194 kb and 0.118 kb).

In placental samples, both MAO-A mRNA and MAO-B mRNA were again detected. At low magnification, MAO-A mRNA was evident mainly on the outer (syncytiotrophoblastic) surface of villi (Figure 3A). In the stem villous trunks, a signal was observed only in the cytotrophoblastic cellular groups distributed in the stroma (Figure 3A). Very low levels of MAO-A mRNA were evident in the smooth musculature of placental vessels and the endothelial lining of the arteries and veins (Figure 4A). At higher magnification, the expression of MAO-A mRNA was mainly localized in the syncytiotrophoblastic and the cytotro-phoblastic cells of chorionic villi (Figure 4A, inset).

The use of MAO-B anti-sense probes indicated that the highest levels of MAO-B mRNA are in the syncytiotrophoblastic layer of the small and budding villi; (Figure 3C) and in the cytotrophoblastic cellular groups distributed in the villous trunk (Figures 3C and 4C). In the placental vasculature, expression of MAO-B mRNA was absent from the connective tissue surrounding blood vessels whereas, unlike that of MAO-A, there was strong expression in the smooth muscula-ture and the endothelial lining of the placental arteries and veins (Figure 4C). At higher magnifications, the expression of MAO-B mRNA in the villi was mainly localized in the syncytiotrophoblastic and cytotrophoblastic layers (Figure 4C, inset).

No hybridization was evident with the sense probes for MAO-A or -B (Figures 3B, 3D, 4B, and 4D). Posthybridization treatment with RNase also had no effect on the signal intensity or distribution; this was the case with both the sense and anti-sense probes.

Discussion

In this study we have clearly demonstrated the distribution of monoamine oxidase-A and -B mRNA in placental sections obtained from normal full-term deliveries, using in situ hybridization with DIG-labeled RNA probes. DIG-labeled cRNA probes were synthesized from MAO-A and MAO-B cDNA templates (Bach et al. 1988) in the sense (negative control) and anti-sense orientations and were partially hydrolyzed to around 400 nucleotides, using limited alkaline hydrolysis, to increase their intracellular permeation (Brahic and Haase 1978). Probe specificities were verified by Northern blot analysis. The use of unlabeled sense cRNA of either MAO-A or MAO-B as a template for Northern hybridization revealed the expected products only with their specific probes, and no hybridization signals were evident when the opposite probes were used. These findings prove that, despite the fact that the base sequences of MAO-A and MAO-B mRNAs are 50.5% homologous, (Bach et al. 1988), there was no cross-hybridization between the MAO-A probe and MAO-B mRNA or the MAO-B probe and MAO-A mRNA. Moreover, the studies on human liver RNA, which is known to contain both MAO transcripts (Grimsby et al. 1990), showed that the anti-sense probes hybridized to transcript sizes of approximately 4.9 kb for both MAO-A and MAO-B. Indeed, the products revealed when placental RNA was used for hybridization were also 4.9 kb for MAO-A and MAO-B, and the signal for the MAO-B transcript in the placenta was very strong.

Figure 3

Placental sections (8 mm) hybridized with MAO-A anti-sense (A), MAO-A sense (B), MAO-B anti-sense (C), and MAO-B sense (D) probes. (A) MAO-A mRNA signal is evident in all villi and villous buds and in the cytotrophoblast cell groups of the villous stem trunk (arrows). (C) MAO-B mRNA signal is strong in all villi and villous buds and in the cytotrophoblast cell groups of the villous stem trunk (arrows). Signal is also evident in the vessel walls (arrow). No signal is evident in sections hybridized with either the MAO-A sense or MAO-B sense probe. Bar = 0.2 μm.

Figure 4

Placental sections hybridized with anti-sense (A,C) and the sense (B,D) probes of MAO-A and MAO-B, respectively. (A) MAO-A mRNA is evident in the syncytiotrophoblastic layer of all villi (arrows). There is low signal in the smooth musculature and endothelial lining of blood vessels, and no signal in erythrocytes. Inset at higher magnification shows the presence of MAO-A mRNA in the syncytiotrophoblastic layer and in isolated cytotrophoblast cells. (C) MAO-B mRNA is strongly evident in the syncytiotrophoblastic layer of small villi and the endothelial lining of blood vessels (arrows). Signals are also evident in the vascular smooth musculature (s). There is no signal in the outer connective tissue layer (c) or in the erythrocytes. Inset at higher magnification shows the presence of MAO-B mRNA in the syncytiotrophoblastic layer and isolated cytotrophoblast cells. (B,D) Sense probes. No signal is evident in any part of these sections. Bars: A-D 5 = mm; insets = 25 μm.

Previous work using MAO cDNA as a probe has indicated that two MAO-A transcripts occur in placenta, one of between 4.4 and 5.4 kb and the other around 2 kb (Bach et al. 1988; Hsu et al. 1988; Grimsby et al. 1990). The smallest transcript is consistent with the full-length coding sequence, the variation in the larger transcript apparently being due to differences in polyadenylation sites (Grimsby et al. 1990). An MAO-B transcript of 3.1 kb has been reported by Bach et al. (1988) and Grimsby et al. (1990) in several tissues, but not one of around 4.9 kb as documented in the present study, and only a low level of an MAO-B transcript has been reported in placenta (Bach et al. 1988). It is not known why the results conflict, but it is clear that the RNA probes used by us are specific for transcripts for the two forms of MAO (Figure 1B). It is also clear that the transcript sizes appear to vary among different reports. This discrepancy may be partly attributed to the variation of integrity of RNA preparations used for Northern hybridization. Indeed, the hybridization products appeared sharper for RNA-extracted placenta (which was extracted fresh; Figure 1A, Lanes 3, 7, and 11) than for RNA extracted from liver at about 12-16 hr post mortem (Figure 1A, Lanes 2, 6, and 10). In this study, once the RNA was extracted, RNase inhibitor was always added. There are further uncertainties about the size and/or nature of MAO mRNA. For example, it has been reported that a significant lack of homology in the 3′-untranslated region of MAO-A cDNA was observed from two published sources (Hotamisligil and Breakfield 1991). What is clear from our Nothern blots, however, is that placental tissue extracts contain significant amounts of MAO-B mRNA as well as large amounts of MAO-A mRNA. To substantiate this claim we used RT-PCR on RNA purified from human placental and hepatic tissues as well as from some selected human cell lines, such as HepG2. We designed sequence-specific antisense and sense primers for MAO-A and others for MAO-B. In the RT step, specific anti-sense primers that flanked more than two exons for each gene were used (rather than random primers) to avoid the possibility of amplifying sequences similar to MAO. The products generated were exactly as predicted for each gene sequence (745 kb for MAO-B and 522 kb for MAO-A). These experiments proved conclusively that MAO-B mRNA occurs at quite high levels in placental samples. The significance of this is discussed below.

When the RNA probes were used for in situ hybridization on sections, MAO-A mRNA was easily detected in the trophoblast cells of the placental villi. This distribution agrees with earlier studies, based on enzyme catalytic activity (Yoshimoto et al. 1986) and immunohistochemical methods (Thorpe et al. 1987). Low levels of MAO-A mRNA appeared to be present also in the smooth musculature and the endothelial lining of placental arteries and veins. This is in agreement with the distribution of MAO-A protein detected using MAO-A-specific monoclonal antibodies in our laboratory (Church et al. 1994). MAO-B mRNA was seen to be highly concentrated in the trophoblast cells of the placental villi and also in the smooth musculature and the endothelial lining of placental arteries and veins. Because it is very difficult to make quantitative comparisons with different probes, the present data do not really enable one to compare the levels of MAO-A mRNA and MAO-B mRNA in placenta.

With histochemical techniques, MAO-B catalytic activity has not been detected in placental sections (Yoshimoto et al. 1986). However, low but significant MAO-B activity has been detected in placental extracts (e.g., Riley et al. 1989). In addition, using MAO-Bspecific monoclonal antibodies, Riley et al. (1989) have also reported the presence of MAO-B protein in these extracts at a level much greater than would be expected if it were derived solely from blood elements. Indeed, Riley et al. show that, under optimal conditions for their antibodies, placental mitochondria contain similar levels of MAO-A and MAO-B protein, a result considered surprising on the basis of the low level of MAO-B activity detected. However, using an MAO-B monoclonal antibody synthesized in our laboratory (Yeomanson and Billett 1992) to immunostain placental sections, we dectected only low levels of MAO-B protein (unpublished results), corroborating the results of Thorpe et al. (1987). On the other hand, our results directly show that MAO-B mRNA is indeed synthesized by cells intrinsic to the placenta (and in our sections is not present at detectable levels in the blood), supporting the suggestion that placental cells synthesize MAO-B in addition to MAO-A. Why MAO-B mRNA should be expressed at a higher level than would be anticipated from protein levels is not clear, but this may point to post-translational regulation of MAO-B expression.

Our findings clearly demonstrated that, using in situ hybridization with DIG-labeled RNA probes, it was possible to study, for the first time, the cellular distribution of MAO-A and MAO-B mRNAs in human placenta. Although in situ hybridization has recently been used to locate MAO mRNA in rat brain samples (Jahng et al. 1997), cDNA and oligonucleotide probes were used rather than the superior RNA probes employed here and, in addition, the probes were radiolabeled. Our work represents the first study using DIG-labeled RNA probes for MAO mRNA in human tissues. Our findings will be useful for further investigative studies on the expression of MAO in placenta and will, for example, allow an assessment of the role of placental MAO in conditions such as preeclamptic hypertension (Gujrati et al. 1985; Weiner 1987) and diabetes mellitus (Barnea et al. 1986). Our probes should also facilitate studies at the cellular level of the time course and regional differences in the pharmacological regulation of MAO gene expression.