Proteomic identification of membrane-associated placental protein 4 (MP4) as perlecan and characterization of its placental expression in normal and pathologic pregnancies

Abstract

Background: More than 50 human placental proteins were isolated and physico-chemically characterized in the 70-80s by Hans Bohn and co-workers. Many of these proteins turned to have important role in placental functions and diagnostic significance in pregnancy complications. Among these proteins was membrane-associated placental protein 4 (MP4), for which identity or function has not been identified yet. Our aim was to analyze the sequence and placental expression of this protein in normal and complicated pregnancies including miscarriage, preeclampsia and HELLP syndrome.

Methods: Lyophilized MP4 protein and frozen healthy placental tissue were analyzed using HPLC-MS/MS. Placental tissue samples were obtained from women with elective termination of pregnancy (first trimester controls, n = 31), early pregnancy loss (EPL) (n = 13), early preeclampsia without HELLP syndrome (n = 7) and with HELLP syndrome (n = 8), late preeclampsia (n = 8), third trimester early controls (n = 5) and third trimester late controls (n = 9). Tissue microarrays were constructed from paraffin-embedded placentas (n = 81). Slides were immunostained with monoclonal perlecan antibody and evaluated using light microscopy and virtual microscopy. Perlecan was also analyzed for its expression in placentas from normal pregnancies using microarray data.

Results: Mass spectrometry-based proteomics of MP4 resulted in the identification of basement membrane-specific heparan sulfate proteoglycan core protein also known as perlecan. Immunohistochemistry showed cytoplasmic perlecan localization in syncytiotrophoblast and cytotrophoblasts of the villi. Perlecan immunoscore decreased with gestational age in the placenta. Perlecan immunoscores were higher in EPL compared to controls. Perlecan immunoscores were higher in early preeclampsia without and with HELLP syndrome and lower in late preeclampsia than in respective controls. Among patients with preeclampsia, placental perlecan expression positively correlated with maternal vascular malperfusion and negatively correlated with placental weight.

Conclusion: Our findings suggest that an increased placental perlecan expression may be associated with hypoxic ischaemic injury of the placenta in miscarriages and in early preeclampsia with or without HELLP syndrome.

Keywords: Miscarriage; Placenta; Preeclampsia; Pregnancy; Proteoglycan.

Figure 1. Overview of the site-specific N-glycan distribution of perlecan.

The N-glycosylation microheterogeneity at the individual sites of the perlecan amino acid sequence is shown. Identified glycosylated residues with Asn-X-Ser/Thr motif are colored in blue, whereas non-identified residues are labeled in gray. Glycosaminoglycan chains are abbreviated as GAG chains.

Figure 2. Perlecan expression in first and third trimester placental tissues.

(A) Perlecan immunostaining was detected in the syncytiotrophoblast and cytotrophoblasts of first trimester placenta (GW9). There was a stronger staining in extravillous trophoblast (EVT) than in the syncytiotrophoblast (STB). (B and C) Representative images show stronger syncytiotrophoblastic perlecan expression in first trimester control (TOP) ((B) GW9) compared to third trimester late control ((C) GW40) placentas. Representative images, hematoxylin counterstain, 200× (A and B) and 400× (C) magnification, scale bar: 50 μm. (D) Box-plots represent gene expression levels of HSPG2 gene encoding perlecan in EVT and STB from first trimester placental tissue. Data was derived from GEO database (GSE9773). (E) Syncytiotrophoblastic perlecan immunoscore was higher in some selected placentas collected from first trimester controls (TOP: mean: 1.91 ± SE: 0.09, p = 0.003) than in third trimester late controls (mean: 0.5 ± SE: 0.21). (F) Box-plots represent gene expression levels in placentas from all trimesters. Data was derived from GEO database (GDS4037).

Figure 3. Placental perlecan immunostaining in first trimester miscarriages.

Syncytiotrophoblastic perlecan immunostaining was stronger in early pregnancy loss ((B) EPL, GW9) compared to first trimester controls ((A) TOP, GW9). (C) Perlecan immunoscores were higher in EPL (mean: 2.1 ± SE: 0.13, p = 0.040, n = 13) than in first trimester controls (TOP: 1.84 ± 0.06, n = 31). Representative images, hematoxylin counterstain, 200× magnifications, scale bar: 50 μm.

Figure 4. Placental perlecan expression in preeclampsia with or without HELLP syndrome.

(A–E) Perlecan immunostaining of the syncytiotrophoblast was stronger in early preeclampsia without HELLP syndrome ((B) GW34) and with HELLP syndrome ((C) GW34) than in third trimester early controls ((A) GW34). Weaker syncytiotrophoblastic perlecan immunostaining was detected in late preeclampsia ((E) GW 40) than in third trimester late controls ((D) GW39). (F) Syncytiotrophoblastic perlecan immunoscores increased with gestational age within controls (third trimester early controls mean: 1.37 ± SE: 0.09, n = 5 vs. third trimester late controls: 1.84 ± 0.07, p = 0.0001, n = 9). Syncytiotrophoblastic perlecan immunoscores were higher in early preeclampsia with and without HELLP syndrome (early PE: 1.69 ± 0.12, p = 0.040, n = 7; early PE + HELLP: 1.83 ± 0.06, p = 0.0001, n = 8) than in third trimester early controls (1.37 ± 0.09). Perlecan immunoscores were lower in late preeclampsia (1.47 ± 0.08, p = 0.0009, n = 8) than in third trimester late controls (1.84 ± 0.07). PE: preeclampsia; HELLP: HELLP syndrome; PE + HELLP: preeclampsia with HELLP syndrome, *p < 0.05, SE: standard error of mean.

Figure 5. Correlation of perlecan immunoscores with clinical parameters.

(A and B) Perlecan immunoscore positively correlated with maternal vascular malperfusion (MVMP) score of the placenta in preeclampsia ((B) R = 0.49, p = 0.017) but not in controls ((A) R = 0.11, p = 0.707). (C and D) Furthermore, perlecan immunoscore negatively correlated with placental weight among patients with preeclampsia ((D) R = −0.46, p = 0.038), but this was not significant in the control group ((C) R = 0.14, p = 0.653). (E and F) Perlecan immunoscore tended to negatively correlate with birthweight among patients with preeclampsia ((F) R = −0.31, p = 0.152), but this was not significant among controls ((E) R = 0.43, p = 0.12).