Biol Reprod Lalor Postdoctoral Fellowships -- Application Deadline January 15, 2009
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

BOR - Papers in Press, published online ahead of print October 26, 2005.
Biol Reprod 2005, 10.1095/biolreprod.105.047324
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
74/2/366    most recent
biolreprod.105.047324v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Nie, G.
Right arrow Articles by Salamonsen, L. A.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Nie, G.
Right arrow Articles by Salamonsen, L. A.
Agricola
Right arrow Articles by Nie, G.
Right arrow Articles by Salamonsen, L. A.
BIOLOGY OF REPRODUCTION 74, 366–374 (2006)
DOI: 10.1095/biolreprod.105.047324
© 2006 by the Society for the Study of Reproduction, Inc.

Research Article

Serine Peptidase HTRA3 Is Closely Associated with Human Placental Development and Is Elevated in Pregnancy Serum1

Guiying Nie 2 4, Ying Li 4, Kathryn Hale 4, Hidetaka Okada 3 4, Ursula Manuelpillai 5, Euan M. Wallace 5, and Lois A. Salamonsen 4

Prince Henry's Institute of Medical Research,4 Clayton, Victoria 3168, Australia Department of Obstetrics and Gynecology,5 Monash University, Monash Medical Centre, Clayton, Victoria 3168, Australia

ABSTRACT

HTRA3 is a newly identified serine peptidase of the mammalian HTRA (high-temperature requirement factor A) family, that is upregulated dramatically during mouse placental development. The current study determined whether HTRA3 was involved in human placentation. During the menstrual cycle, HTRA3 was expressed primarily in the endometrial glands, being significantly upregulated toward the mid- to late secretory phases; prominent expression in the stroma detected only in the decidual cells in the late secretory phase. Thus, overall endometrial HTRA3 expression was highest in the late secretory phase, when the endometrium is prepared for maternal-trophoblast interaction. During the first trimester of pregnancy, both glandular and decidual HTRA3 expression increased further with the decidual upregulation being highly significant. The strong link between HTRA3 expression and endometrial stromal cell decidualization was further established in an in vitro model using primary endometrial stromal cells. HTRA3 was also expressed by certain trophoblast subtypes in the first-trimester placenta: strongly in the villous syncytiotrophoblast, trophoblast shell, and endovascular trophoblast and weakly in the distal portion of the trophoblast cell columns but not in villous cytotrophoblast, the proximal region of the cell columns, or interstitial trophoblast. Upregulation of HTRA3 expression in association with placental development was revealed by a significant elevation of this protein in the maternal serum during the first trimester. We thus propose that HTRA3 is a previously unrecognized factor closely associated with and potentially important for human placentation. This study established crucial groundwork for future investigations toward establishing the physiological roles of HTRA3 in human placentation.

decidua, placenta, pregnancy, trophoblast

INTRODUCTION

In the human, establishment of pregnancy and successful placentation require active modulation of the maternal endometrium and proliferation, differentiation, migration, and invasion of trophoblast. In the endometrium, decidualization is the major change occurring during implantation and placentation and involves the transformation of stromal fibroblasts into decidual cells. This occurs even in the absence of a conceptus; it is initiated spontaneously around 9–10 days postovulation around the spiral arteries and then spreads and progresses further through the upper two-thirds of the endometrium during the late secretory phase of the menstrual cycle [1, 2]. If implantation occurs, decidual changes intensify and advance further to form the decidua of pregnancy. The proposed function of the decidua is to facilitate as well as to restrict trophoblast invasion [2, 3]; impaired decidualization is associated with failure of placentation and pregnancy [4, 5].

Rapid trophoblast proliferation and differentiation starts at implantation from cells surrounding the embryonic disc to form an outer layer of multinucleated previllous primitive syncytiotrophoblast and an inner layer of primitive mononuclear cytotrophoblast [2]. The primitive cytotrophoblast subsequently proliferates and differentiates along two main pathways, villous and extravillous, giving rise to different trophoblast subtypes of the first-trimester placenta containing 1) villous cytotrophoblast, providing precursors for all other trophoblast cells; 2) villous syncyctiotrophoblast, important for nutrient and gas exchange as well as hormone production; 3) cytotrophoblast cell columns and shell, facilitating the attachment of the placenta to the uterus and providing invasive extravillous trophoblasts; and 4) interstitial and endovascular trophoblast, essential for modulating the uterus, especially the spiral arteries, to ensure adequate blood supply to the intervillous space [2, 3].

Recently we identified and cloned a novel gene that is dramatically upregulated in the mouse uterus in association with placentation and predicted to encode a serine peptidase [6]. Subsequently, we cloned the full mRNA sequence of this gene in the human [7]. Further analysis revealed that this peptidase was structurally related to the previously identified mammalian HTRA (high-temperature requirement factor A) peptidases 1 and 2; thus, we have named this new gene HTRA3 [7]. HTRA3 and HTRA1 share very similar protein domain structures, whereas HTRA2 is distinct [7, 8].

In both the human and mouse, two alternatively spliced HTRA3 mRNAs (long and short forms), with two predicted protein isoforms of 49 and 38 kDa, respectively, have been cloned [6, 7]. The long form is predominant in the mouse uterus, whereas the expression of both forms is detected in the human placenta [6, 7]. HTRA3 is well conserved between the human and mouse with similarities of 87% at the mRNA and 95% at the protein level, respectively. An interesting feature of HTRA3 protein is that although it is a serine peptidase, it contains an insulin-like growth factor (IGF) binding domain at the N-terminal end immediately following the signal peptide, suggesting that it can be secreted and may be involved in the IGF system [6, 7].

In the mouse uterus, Htra3 mRNA expression in the endometrium is relatively low before pregnancy, but it increases as pregnancy is established, especially postimplantation and during placentation [6]. HTRA3 proteins are localized predominantly in the decidua basalis during mouse placental development [9].

In the present study, we investigated the expression and regulation of HTRA3 mRNA and protein during placental development in the human, specifically in the endometrium throughout the menstrual cycle and in the placenta and decidua in the first trimester of pregnancy. HTRA3 was strongly upregulated in the decidual cells and selectively expressed in certain trophoblast subtypes at the maternal-fetal interface. We thus further determined the regulation of HTRA3 mRNA and protein expression during in vitro decidualization of primary endometrial stromal cells. We also demonstrated that the association of HTRA3 with placental development was reflected by a significant elevation of this protein in the maternal circulation during early pregnancy.

MATERIALS AND METHODS

Tissues

Endometrial tissues were obtained at curettage from women with normal menstrual cycle and no apparent endometrial dysfunction, undergoing minor gynecological surgical procedures, such as laparoscopic sterilization or investigation of tubal patency. Endometrial samples were collected across different stages of the menstrual cycle: menstrual (Day [d] 1–4], proliferative (d5–14), early secretory (d15–19), midsecretory (d20–24), and late secretory (d25–28) phases. Stage of the cycle was determined from the patient's testimony and confirmed histologically by a qualified gynecological pathologist. First-trimester placental and decidual tissues were obtained from women undergoing elective termination of pregnancy (amenorrhea 8–12 wk). Serum samples were collected from first-trimester pregnant (8–12 wk) and nonpregnant (proliferative and secretory phases of the menstrual cycle) women. Written informed consent was obtained from all participants, and ethical approval was obtained from the Human Ethics Committee at Monash Medical Centre, Melbourne.

For Northern analysis, tissues were immersed in RNA-Later (Ambion, Austin, TX) immediately after collection and stored at –80°C for subsequent RNA extraction. For immunohistochemical analysis, tissues were fixed in buffered formalin (pH 7.4) at 4°C overnight, washed in Tris-buffered saline (TBS, pH 7.4). and processed to paraffin wax blocks. For cell culture experiments, biopsies were collected into DMEM (Trace Biosciences, Sydney, Australia).

RNA Extraction and Northern Blot Analysis

Total RNA was isolated and Northern analysis (15 µg/lane) performed as previously published [10]. Because only a small amount of high-quality RNA was obtained from each individual endometrial sample, RNA from three to four biopsies (from the same phase of the cycle) were pooled and used for Northern blotting. RNA extracted from first-trimester placenta (n = 4) and decidua (n = 4) were examined individually. A cDNA fragment of 457 bp representing the common region of human HTRA3 mRNAs, capable of detecting both HTRA3 transcripts, was used as a probe [7]. To evaluate the quality of RNA on the membrane, each blot was hybridized with a cDNA probe for GAPDH. The level of mRNA expression was analyzed densitometrically using the Gel Doc 2000 system (Bio-Rad, Hercules, CA) and expressed as a percentage of GAPDH.

Production of Antibody Against HTRA3 and Confirmation of Its Specificity

Anti-HTRA3 antibody was generated in sheep against a synthetic peptide ALQVSGTPVRQC corresponding to residues 116–126 of mouse HTRA3 protein (accession no. NP_084403). This peptide is highly specific to HTRA3, common to both isoforms, and highly conserved between the mouse and human HTRA3 proteins [6, 7]; the resultant antibody is thus predicted to recognize both isoforms of mouse and human HTRA3 proteins. A cystine residue was added to the C-terminal of the peptide, through which the peptide was coupled to an immunogenic carrier protein, bacterial diphtheria toxoid. The peptide was synthesized and purified by HPLC, and the identity was confirmed by ion spray mass spectrometry (Mimotopes Pty Ltd, Clayton, Australia). Sheep were immunized with Montanide/Marcol adjuvant according to a standard protocol [11] at the Prince Henry's Institute Animal Research Facility, Werribee, Victoria, Australia. Specific IgGs were affinity-purified using HiTrap NHS-activated HP affinity columns (Amersham Pharmacia, Piscataway, NJ) following the manufacturer's instructions.

The specificity of the antibody (#153) was validated by Western blotting. Total proteins (25 µg/ml) extracted from a nonpregnant uterus and d10.5 implantation site from the mouse and a human endometrial sample (cycle d25) were separated on 15% SDS-PAGE under reducing conditions and transferred to Hybond-P membranes (Amersham Life Science, Sydney, Australia). After blocking with a blocking buffer (5% [w/v] skimmed milk in TBS and 0.1% [v/v] Tween 20) overnight at 4°C, the membranes were incubated with HTRA3 antibody (35 µg/ml) in the blocking buffer for 1 h at room temperature, then with a HRP-conjugated donkey-anti sheep IgG (1:25000; Silenus Laboratories, Hawthorn, Australia) for 1 h at room temperature and developed by chemiluminescence (ECL Plus system; Amersham).

To further confirm the antibody specificity, the open reading frames of the long and short mRNA sequences of human HTRA3 were cloned into PCRT7-NT-TOPO vector (Invitrogen, Carlsbad, CA) using a PCR approach, and HTRA3 proteins were expressed in vitro using the Expressway Plus Expression System (Invitrogen) according to the manufacturer's instructions. The reaction mixture (5 µl) was precipitated with acetone, and the gene product was verified by Western analysis.

Immunohistochemistry

Five-micron sections were subjected to standard immunohistochemistry. Nonspecific binding was blocked by preincubation of tissue sections with a blocking buffer containing high-salt TBS (0.3 M NaCl in 50 mM Tris, pH 7.6), 0.1% Tween, 15% rabbit serum. and 2% horse serum for 20 min at room temperature. The primary antibody (sheep anti-HTRA3 antibody or preimmune sheep IgG as a negative control [both 0.5 µg/ml]) was incubated in the blocking buffer at 37°C for 1 h and washed with high-salt TBS plus 0.6% Tween. The secondary antibody (biotinylated rabbit anti-sheep IgG, 1:200; Vector Laboratories, Burlingame, CA) was applied in the blocking buffer for 30 min at room temperature. Positive immunostaining was revealed by incubating the sections with an avidin-biotin-complex conjugated to horseradish peroxidase (DakoCytomation, Botany, Australia) for 30 min at room temperature, followed by the application of the peroxidase substrate 3,3N-diaminobenzidine (DakoCytomation) leading to a brown precipitate for positive staining. The sections were counterstained with Harris hematoxylin. Microscopy was performed using an Olympus BH2 microscope fitted with a Fujix HC-2000 high-resolution digital camera (Fujix, Tokyo, Japan).

To determine the cellular location and relative intensity of HTRA3 protein across the menstrual cycle, tissue sections from different phases of the cycle (menstrual, n = 4; proliferative, n = 6; early secretory, n = 3; midsecretory, n = 4, late secretory, n = 6) were examined. The relative intensity of the immunostaining in separate cellular compartments was scored semiquantitatively by two independent observers; 0 represented no stain, and 4 represented maximal staining.

The expression and cellular localization of HTRA3 in pregnant tissues (decidua, n = 4; placenta, n = 4) were also examined by immunohistochemistry. Trophoblast cells were identified by immunostaining for cytokeratin using a mouse anti-human cytokeratin monoclonal antibody (7 µg/ml; BD Biosciences, San Jose, CA) as the primary and a biotinylated horse anti-mouse IgG (1:200; Vector Laboratories) as the secondary antibody, following antigen retrieval by microwaving the sections for 5 min in 0.01 M citric acid buffer (pH 6.0).

In Vitro Decidualization of Endometrial Stromal Cells

Endometrial stromal cells (ESCs) were isolated from endometrial biopsies and cultured as previously described [12]. The ESCs were seeded in six-well plates (1 x 106 cells/well), cultured to confluence, and decidualized by the treatment with estradiol 17 ß (E2, 10–8 M) plus medroxy-progesterone acetate (P, 10–7 M) (both E2 and P; Sigma, St. Louis, MO) for 12 days; cells treated with E2 alone served as the controls. The culture media were changed every 3 days and centrifuged supernatants assayed for the decidual cell marker, prolactin, by ELISA [12]. At the end of the experiment, the cells were harvested for RNA isolation or protein extraction. The experiment was repeated with three different cell preparations.

Total RNA was isolated from ESCs using an RNeasy Minikit (Qiagen GmbH, Hilden, Germany) according to the manufacturer's instructions. DNA-free RNA (1 µg) was reverse transcribed with random primers and quantitative real-time reverse transcription-polymerase chain reaction (RT-PCR) performed using a Roche LightCycler (Roche, Castle Hill, Australia) as published [12]. PCR analysis was performed using 4-µl diluted RT samples (1:5 dilution for HTRA3 and 1:500 for 18S) in glass capillaries in PCR mastermix (Roche) containing SYBR Green and supplemented with 3 mmol/L MgCl2 and specific primers (each 10 pmol) (Table 1) in a total volume of 20 µl. PCR reaction was initiated by a denaturing step of 10 min at 95°C followed by 26–44 cycles of 95°C for 15 sec, 62–64°C for 5 sec, and 72°C for 6–16 sec (specific parameters; Table 1). For each primer pair, serial dilution of a standard (known amount of purified PCR product derived from that primer pair) was included in each run together with all samples to be compared. Quantitation of mRNA level was performed when the amplification was in the log-linear phase and parallel to the standards. At the end of the program, melting curve analysis was carried out to ensure product specificity. The identity of each PCR product was also verified by agarose gel electrophoresis and DNA sequencing in initial experiments. The ratio of HTRA3 mRNA level to that of 18S was calculated for each sample, and the relative level of HTRA3 expression in decidualized ESCs was expressed as percentage of the controls.


View this table:
[in this window]
[in a new window]
  		TABLE 1. Primer details and amplification conditions for real-time PCR analysis

To extract total proteins from cultured ESCs, 100 µl of protein extraction buffer (6% SDS, 140 mM Tris-HCl, 22.4% glycerol) supplemented with 2 µl of a peptidase inhibitor cocktail (Calbiochem, San Diego, CA) were added to each well and the cells scraped with the tip of a 1-ml syringe. The total lysate was collected and centrifuged at 1400 rpm for 15 min; 30 µl of the resultant supernatant were subjected to standard 10% reducing SDS-PAGE and Western blotting. The relative level of HTRA3 protein was determined by densitometric analysis, and the level in decidualized cells was expressed as percentage of that in the controls.

Detection of HTRA3 Protein in Serum

The level of HTRA3 protein in serum from first-trimester pregnant (n = 8) and nonpregnant (n = 8, proliferative [n = 4] and secretory phases [n = 4] of the menstrual cycle, respectively) women were determined by Western blotting. Prior to analysis, major subclasses of human gamma globulin (IgG) in the serum was removed by incubating 10 µl neat serum with 20 µl protein G agarose (Roche) and 65 µl binding buffer (25 mM Tris, 25 mM NaCl, 0.01% sodium azide, pH 7.5) at 4°C for 12–16 h, followed by centrifugation for 20 sec at 12000 x g. The resultant supernatant (9 µl) was subjected to standard 15% nonreducing SDS-PAGE and Western blotting for HTRA3. The relative protein level was determined by densitometric analysis of band intensities.

Statistical Analysis

Data are expressed as mean ± SEM. Comparisons of HTRA3 mRNA expression levels of the two transcripts between the first-trimester placenta and decidua and comparisons of HTRA3 immnostaining intensities between multiple groups were performed using one-way analysis of variance, followed by the Tukey test (PRISM version 4 for Windows; GraphPad Software Inc, San Diego, CA). P < 0.05 was considered statistically significant. Comparisons between two parameters or two groups were performed using unpaired Student t-test; P < 0.05 (one-tail) was considered statistically significant.

RESULTS

Northern Analysis of HTRA3 mRNA Expression in Human Endometrium Throughout the Menstrual Cycle and in the Placenta and Decidua of First-Trimester Pregnancy

Two transcripts of ~2.8 and ~2.2 kb, representing the two alternatively spliced forms of HTRA3 mRNA (long and short, respectively), were detected in the total RNA isolated from endometrium throughout the menstrual cycle and tissues of early pregnancy (Fig. 1A), as in other human tissues such as the heart [7]. During the menstrual cycle, the long form was expressed predominantly at all stages, and the expression levels were highest in the mid- to late secretory phase for both transcripts (Fig. 1, A and B). During first-trimester pregnancy, both HTRA3 mRNA transcripts were detected in the decidua and the placenta (Fig. 1A). In the decidua, the two transcripts were at equivalent levels (Fig. 1B), while in the placenta, the 2.8-kb transcript was overall at a higher level than the short form, although the difference was not statistically significant (Fig. 1B).


Figure 1
View larger version (29K):
[in this window]
[in a new window]
  		FIG. 1. HTRA3 mRNA expression in human endometrium across the menstrual cycle and in first-trimester placenta and decidua. A) Northern analysis of total RNA (15 µg/lane) from phase-specific endometrium throughout the menstrual cycle (pooled RNA [n = 3–4 for each phase]) and from first-trimester placenta (n = 4) and decidua (n = 4). The top panel shows two bands at ~2.8 and ~2.2 kb representing the long and short HTRA3 mRNA transcripts, respectively, and the lower panel shows the GAPDH signal on the same membrane. B) Densitometric analysis of band intensities of HTRA3 mRNA (long and short, respectively) relative to that of GAPDH. Cycling Endo, Endometrial samples from different phases of the menstrual cycle; Mens, menstrual; Pro, proliferative; E-sec, early secretory; ML-sec, mid- to late secretory; 1st-trim Preg, first-trimester pregnancy; Pla, placenta; Deci, decidua

Production of a Polyclonal Antibody Against HTRA3 and Confirmation of Its Specificity

To analyze HTRA3 protein, a polyclonal antibody against a synthetic peptide common to both isoforms of mouse HTRA3 and highly homologous between the mouse and human was generated in the sheep and affinity-purified. Western blot using this antibody detected specific bands in mouse uterine tissues and in human endometrium (Fig. 2A). In the mouse, two clear bands around ~48 and ~41 KDa were detected, the level being much higher in the d10.5 implantation site than the nonpregnant uterus (Fig. 2A), in agreement with previous Northern analysis showing a dramatic upregulation of Htra3 mRNA during mouse placentation [6], and confirming the specificity of the antibody. As only the long-form Htra3 mRNA was well expressed in the mouse uterus and placenta [6], the two protein bands detected in the mouse tissues likely represent the native (~48-kDa band, predicted size 49 kDa) and processed (~41 kDa) HTRA3 long forms. Two clear bands of ~50 and ~41 kDa were also detected in the human endometrial samples (Fig. 2A), confirming that this antibody recognized human HTRA3 proteins as predicted.


Figure 2
View larger version (31K):
[in this window]
[in a new window]
  		FIG. 2. Western blot analysis to confirm the specificity of the sheep anti-HTRA3 antibody. A) Total proteins extracted from a nonpregnant mouse uterus (NP), a d10.5 mouse implantation site and a human endometrial sample (menstrual cycle d25, H-endo) were analyzed. B) HTRA3 protein expressed in an in vitro expression system was examined. Reaction mixture containing HTRA3 expression constructs (long and short, respectively) was compared to the control. A specific band was detected in each reaction containing HTRA3 construct, confirming the specificity of the antibody and illustrating the capacity of this antibody to recognize both isoforms of HTRA3 protein as predicted

The specificity of this antibody was further established using expressed human HTRA3 proteins. The two isoforms of human HTRA3 were expressed in an in vitro expression system and analyzed by Western blot using the HTRA3 sheep antibody. A single specific band at ~41 kDa for the long form and one band at ~30 kDa for the short form were detected only when the HTRA3 expression constructs were added to the reaction compared, to the control (Fig. 2B), further confirming the specificity of the antibody. Taken together, the data suggest that the two protein bands detected in the human endometrium were the native (~50 kDa) and processed (~41 kDa) forms of long HTRA3 protein, reflecting the early observation that the long-form HTRA3 mRNA was the predominant form expressed during the menstrual cycle.

Immunolocalization of HTRA3 Protein in Human Endometrium During the Menstrual Cycle and in Pregnant Tissues of First-Trimester Pregnancy

Immunoreactive HTRA3 was detected in the glandular epithelium at all the stages during the menstrual cycle (Fig. 3, A–D); stromal staining was seen only in the late secretory phase in the decidual cells (Fig. 3D). Glandular expression was low during the menstrual phase (Fig. 4A), increasing significantly in the proliferative phase but decreasing again in the early secretory phase (Fig. 4A). With the progression of the secretory phase, the glandular expression increased to reach a significantly higher level in the late secretory phase (Figs. 3, B–D, and 4A). In the late secretory phase, strong immunostaining was also detected in decidual cells (Figs. 3, D and F, and 4B). Thus, during the cycle, HTRA3 protein is maximal in the late secretory phase in both glandular and decidual cells.


Figure 3
View larger version (142K):
[in this window]
[in a new window]
  		FIG. 3. Immunolocalization of HTRA3 in human endometrium across the menstrual cycle (A–F) and in first-trimester pregnant tissues (G–M). Proliferative (A); early secretory (B); midsecretory (C); and late secretory phase (D). Insert in D, negative control. Strong staining in the late secretory phase glands and decidual cells is shown at high magnification in E and F, respectively. During first-trimester pregnancy, strong immunostaining was detected in maternal endometrial glands (G) and the decidual cells (H). Intense staining was also detected in syncytiotrophoblast in the floating villi (I, arrow), in trophoblast shell (J), and in endovascular trophoblast found in the blood vessel (K, arrowheads highlighting cells lining the vessel). Weak staining was seen in the distal region of the cell columns (L). Very low or no staining was detected in the cytotrophoblast in the floating villi (I, arrowhead), in the proximal region of the cell column (L), or in interstitial trophoblast (M). N–Q) serial sections of J–M, respectively, stained for cytokeratin. cc, Cell columns; deci, decidual cells; ge, glandular epithelium; str, stroma; ts, trophoblast shell. Bar = 20 µm


Figure 4
View larger version (12K):
[in this window]
[in a new window]
  		FIG. 4. Mean density (±SEM) of HTRA3 immunostaining in endometrial glands (A) and decidual cells (B) across the menstrual cycle and in early pregnancy. Sections from menstrual (Mens, n = 4), proliferative (Pro, n = 6), early secretory (E-sec, n = 3), midsecretory (M-sec, n = 4), and late secretory (L-sec, n = 6) phases of the cycle and first-trimester pregnancy decidua (1st-trim preg, n = 4) were analyzed. Glandular staining was significantly higher at Prof compared to Mens and E-sec and at L-sec compared to E-sec and Mens, respectively; decidual staining was significantly increased during early pregnancy. **P < 0.01; *P < 0.05

During first-trimester pregnancy, HTRA3 immunostaining was likewise detected in endometrial glands and decidual cells (Fig. 3, G and H), but the intensity in both cell types, especially in decidual cells, was higher than during the menstrual cycle (Figs. 3, F and H, and 4, A and B). HTRA3 was also detected in some but not all trophoblast. In the floating villi, strong staining was detected in the syncytiotrophoblast (Fig. 3I, arrow), with the adjacent cytotrophoblast showing very low or no staining (Fig. 3I, arrowhead). Intense staining was also detected in the trophoblast shell (Fig. 3J), which was identified by its positive staining for cytokeratin on a serial section (Fig. 3N). The endovascular trophoblast in the blood vessels, identified by positive staining for cytokeratin (Fig. 3O), showed strong immunostaining for HTRA3 (Fig. 3K); trophoblast both lining (Fig. 3K, arrowhead) and freely floating inside (Fig. 3K) the vessel was positive for HTRA3. In the trophoblast cell columns, most of the cells were negatively stained for HTRA3 (Fig. 3L), but those located in the distal portion of the columns appeared to be weakly positive (Fig. 3L), while all cells were positive for cytokeratin (Fig. 3P). Interstitial trophoblast infiltrated into the maternal decidua were also identified by their positivity for cytokeratin (Fig. 3Q) but were negative for HTRA3 (Fig. 3M). No HTRA3 immunostaining was detected in other cell types including the uterine natural killer cells, which were present in large numbers in the maternal decidua and positive for CD56 (data not shown).

HTRA3 Expression During In Vitro Decidualization of ESCs

To further establish that HTRA3 is upregulated in endometrial stromal cells during decidualization, human ESCs were isolated, decidualization was induced in vitro, and expression of HTRA3 mRNA and protein was determined. Successful decidualization was confirmed by an induction of prolactin secretion in cells cultured with E2 plus P compared to the control (E2 alone, data not shown). To determine HTRA3 mRNA expression, three primer pairs (Table 1) were designed to assess the long and short specific transcripts as well as the total HTRA3 mRNA (by amplification of a region common to the both transcripts). A single specific band corresponding to the expected size was obtained for all three types of amplifications (Fig. 5A). The PCR specificity was further confirmed by direct DNA sequencing of purified PCR products (data not shown). Real-time RT-PCR analysis demonstrated that overall, the decidualized ESCs showed higher levels of all forms of HTRA3 transcripts compared to nondecidualized control ESCs (Fig. 5B). However, the long form showed a significant (P < 0.01) induction, whereas the short form exhibited a trend of increase (P = 0.09) on decidualization (Fig. 5B). The level of total HTRA3 mRNA, determined by the amplification of a common region to the both transcripts, was also significantly higher (P < 0.05) in the decidualized compared to the control ESCs (Fig. 5B). The magnitude of increase in the total HTRA3 mRNA level was between that for each distinct transcript (Fig. 5B).


Figure 5
View larger version (20K):
[in this window]
[in a new window]
  		FIG. 5. Induction of HTRA3 mRNA (A and B) and protein (C and D) in endometrial stromal cells (ESCs) during in vitro decidualization. ESCs were cultured for 12 days with either E2 alone (control, Con) or E2 plus P (decidualization, Deci) and analyzed for HTRA3 mRNA by real-time RT-PCR or protein by Western blotting. For mRNA analysis, the levels of the long (L) and short (S) specific transcripts as well as total HTRA3 mRNA level (by amplification of a region common [Com] to both transcripts) were determined. A) Agarose gel electrophoresis of PCR products showing a pure single band in each amplification. B) real-time RT-PCR quantitation of the different HTRA3 transcripts in ESCs during decidualization (n = 3). The levels were adjusted to that of 18S, and results are expressed as percentage of the controls. The long transcript (**P < 0.01) and the common region (*P < 0.05) of HTRA3 mRNA are significantly higher in the decidualized compared to the control ESCs, whereas the short form showed a trend of increase (P = 0.09). C) Western blot analysis of HTRA3 proteins. One representative control and decidualized ESC sample, respectively, together with a human endometrial sample (Menstrual Cycle Day 25) used as an internal control are shown. Two specific bands (~41 and ~30 kDa) were detected in the cells with the intensities stronger in the decidualized ESCs. D) Densitometric analysis of the two HTRA3 protein bands in decidualized compared to the control ESCs (n = 3). The intensities of both bands are higher in decidualized ESCs (P = 0.07)

The change in HTRA3 protein during decidualization was analyzed by Western blotting. To confirm band specificity, a human endometrial sample (menstrual cycle d25) used in the previous experiments to establish antibody specificity was included as an internal control. Two specific bands at ~41 and ~30 kDa, corresponding to the processed long and short isoforms of HTRA3 protein, respectively, were detected in the ESCs, with the intensities stronger in the decidualized cells (Fig. 5C). The ~50-kDa band consistently seen in the endometrial tissues was not detected in the cells (Fig. 5C), reflecting differences in posttranslational modifications of proteins by whole tissues in vivo and by isolated stromal cells in culture. The processed short HTRA3 protein form, expected to be ~30 kDa, was not previously detected in endometrial tissue samples; it was now clearly seen in the decidualized ESCs when its mRNA was upregulated (Fig. 5C). Densitometric analysis showed that the levels of both HTRA3 protein isoforms were increased in the decidualized compared to the control ESCs (Fig. 5D), although the changes were not statistically significant (P = 0.07), reflecting the semiquantatitive nature of the analysis method.

Detection of HTRA3 in Maternal Serum

We next investigated whether HTRA3 was detectable in maternal serum of first-trimester pregnancy and in serum of nonpregnant women by Western blot analysis. Two specific bands clustered at ~39 kDa were detected in all serum samples examined (Fig. 6A, top panel). However, the overall level was very low in nonpregnancy sera, although relatively the bands were seen more clearly in the secretory phase than in the proliferative phase of the menstrual cycle (Fig. 6A, top panel). In contrast, much stronger bands were detected consistently in all first-trimester pregnancy sera (Fig. 6A, top panel). These bands were highly specific to HTRA3, as no signals were detected when identical blots were probed with preimmune sheep IgG instead of HTRA3 antibody (Fig. 6A, bottom panel). Densitometric analysis revealed that the HTRA3 band intensities were elevated 7-fold in the first trimester of pregnancy compared to the nonpregnancy sera (Fig. 6B).


Figure 6
View larger version (24K):
[in this window]
[in a new window]
  		FIG. 6. Detection of HTRA3 protein in human serum. Sera from nonpregnant (NP, n = 8, proliferative [n = 4] and secretory [n = 4] phases of the menstrual cycle, respectively) and first-trimester pregnant (1st-trim Preg, n = 8) women were analyzed by Western blot analysis. A) Top panel, a representative blot probed with HTRA3 antibody in sera from four NP and four pregnant women; bottom panel, a blot identical to the top panel probed with preimmune sheep IgG instead of HTRA3 antibody. B) Densitometric analysis of the HTRA3 protein bands detected in the serum, the intensity was 7-fold higher (**P < 0.01) in sera of pregnant compared to NP women

DISCUSSION

This study established a clear association between HTRA3 expression and placental development in the human. During the menstrual cycle, HTRA3 was expressed primarily in the endometrial glands, being significantly upregulated toward the mid- to late secretory phases; prominent stromal expression was detected in the decidual cells at the late secretory phase, when spontaneous decidualization occurred. Consequently, the overall endometrial HTRA3 expression is highest at the late secretory phase of the cycle, when the endometrium is prepared for maternal-trophoblast interaction should implantation be initiated. Both glandular and decidual expression of HTRA3 increased further during early pregnancy; in particular, decidual expression was significantly upregulated during first-trimester pregnancy when decidualization advanced. The strong link between HTRA3 expression (both its mRNA and protein) and decidualization was further established in an in vitro model using isolated primary endometrial stromal cells. HTRA3 was also detected in trophoblast, but the expression was restricted to certain subtypes: it was strongly expressed in the villous syncytiotrophoblast, trophoblast shell, and endovascular cytotrophoblast; weakly in the distal portion of the trophoblast cell columns; but not in the villous cytotrophoblast, proximal region of the cell columns, or interstitial trophoblast. The pattern implies that, apart from the interstitial trophoblast, upregulation of HTRA3 is associated with trophoblast differentiation. Except for decidual cells and trophoblast, other cell types in the utero-placental unit were negative for HTRA3. The strong association of HTRA3 expression with placental development was clearly reflected by a significant elevation of this protein in the maternal serum of first-trimester pregnancy.

Trophoblast proliferation, differentiation, and invasion are known to be stringently regulated at many levels by numerous growth factors, their binding proteins, adhesion molecules, and extracellular matrix (ECM) proteins produced by the decidua and trophoblast. Among these factors are the transforming growth factor-ßs (TGFBs), IGFs, and their binding proteins (IGFBPs) [13–20]. As HTRA3 is a newly identified protein [6, 7], its function is still not well understood. However, it has become evident that HTRA3 and its closely related family member HTRA1 are involved in regulating TGF-ß signaling [21, 22] and the IGF/IGFBP system [7, 23], suggesting that HTRA3 exerts its function by modulating these two pathways/regulators critical for placental development. In addition, HTRA3 and HTRA1 were shown to be able to degrade a number of ECM proteins [22, 24], implying that HTRA3 may also directly modulate the ECM microenvironment, especially in the decidua, to facilitate trophoblast invasion.

In mammals, three isoforms of TGF-ß proteins (TGFB1–3) have been identified that are pluripotent cytokines regulating cell proliferation and differentiation, positively or negatively, depending on the cell types [25]. TGFBs have been immunolocalized in villous syncytiotrophoblast, in cytotrophoblast shell, and strongly in the decidua throughout gestation [26]. In particular, intense immunostaining was detected in the decidual ECM during first-trimester pregnancy [26]. A number of functional studies demonstrated that TGFBs produced by the decidual cells (and to a minor extent by trophoblast) inhibit trophoblast invasion through paracrine and autocrine manners, respectively [26]. The inhibitory effect of TGFBs on trophoblast invasion was further confirmed by a very recent study showing that inhibition of endogenous and exogenous TGFBs increased the invasive capacity of extravillous trophoblast derived from placental explants in a dose-dependent manner [27].

HTRA3 and HTRA1 are expressed at sites where TGF-ß signaling plays important regulatory roles during mouse embryogenesis [21, 22] and are able to bind a broad range of TGFBs and to inhibit TGF-ß signaling in C2C12 myoblast cells [21, 22]. The inhibition was dependent on the proteolytic activities of the HTRA proteins, but TGFB molecules or their receptors were not degraded [21, 22]. Consequently, HTRA3 and HTRA1 were suggested to be novel inhibitors of the TGF-ß signaling pathway [21, 22]. In the current study, we established that the expression pattern and cellular localization of HTRA3 were also overall similar to that published for TGFBs during human placental development. Thus, it could be postulated that HTRA3 functions through its interaction with the TGFBs during placentation. As TGFBs are inhibitory for trophoblast invasion, it is possible that HTRA3 proteins, especially those produced by the decidual cells, inhibit TGF-ß signaling and facilitate trophoblast migration and invasion.

It is well established that the IGF/IGFBPs are important modulators in mediating the interaction between fetal and maternal tissues for the establishment and maintenance of pregnancy [16–18, 28]. All components of the IGF/IGFBP system are expressed in the endometrium/decidua or placenta at some stage of pregnancy in all species, including human and rodents [17]. In particular, it is proposed that trophoblast-derived IGF2 and decidual-derived IGFBPs provide autocrine/paracrine enhancement of trophoblast migration and invasion [14, 17, 28]. Convincing data supporting this notion is that when IGFBP1 was overexpressed in the decidua in mice [29] with resultant reduced IGF bioavailability in the decidual microenvironment, the development of the placenta was impaired and trophoblast invasion markedly reduced.

Proteolysis of IGFBPs is another level of modulation in the IGF/IGFBP system leading to increased IGF bioactivity [30, 31]. Activities of peptidases degrading IGFBPs have been demonstrated within the uterine/fetal axis during pregnancy [31, 32]. In particular, a large placental-derived glycoprotein named pregnancy-associated plasma protein A proteolyzes IGFBP4 in conditioned medium from cultured decidual cells [33]. Such proteolytic action at the maternal-fetal interface would facilitate IGF action on trophoblast migration [34].

Structural analysis revealed that HTRA3 and HTRA1 are serine peptidases possessing an IGF-binding domain [6, 10], suggesting that these proteins may interact with the IGF/IGFBP system. A recent study demonstrated that HTRA1 cleaves IGFBP5 and that the inhibition of its peptidase activity suppresses IGF1 action [23], providing evidence that HTRA1—and most likely HTRA3 too—could indeed modulate the IGF/IGFBP system. Thus, it is possible that HTRA3 may act as a specific peptidase regulating the IGF/IGFBP system during placentation, although further studies are needed to establish the exact molecular mechanism of its action.

Furthermore, proteolytic degradation and modification of ECM is a crucial component of placentation. Decorin is a member of a widely expressed family of proteoglycans and a modulator of ECM and is produced during decidualization and localized in decidual ECM during early pregnancy [35]. Decorin inhibits trophoblast migration and invasion independent of other growth factors such as TGFBs [36]. Recombinant HTRA3 and HTRA1 can degrade extracellular matrix proteins such as decorin [22, 24]. It is thus possible that the proteolytic activities of decidual-derived HTRA3 may also directly target decorin in the decidual ECM, thereby regulating the ECM microenvironment for trophoblast invasion and migration.

The two types of trophoblast invasion, endovascular and interstitial, are usually considered together. However, it is becoming evident that these two routes of invasion are different and due to distinct cell types. For example, E-cadherin is downregulated when the cytotrophoblast leave the villi and invade into the decidua as individual interstitial trophoblast cells, whereas endovascular trophoblast is positive for E-cadherin [37]. In the current study, HTRA3 was strongly expressed in the trophoblast shell, but it was downregulated in the interstitial trophoblast once it was in the decidual interstitum. In contrast, endovascular cytotrophoblast, which is thought to be the continuum of the trophoblast shell, retained HTRA3 expression. The significance of this differential expression of HTRA3 remains to be determined.

Taken together, we have demonstrated a clear association between HTRA3 expression and placental development. The elevation of HTRA3 protein in the maternal circulation during early pregnancy further substantiates this notion. Our results suggest that HTRA3 is likely to be a previously unrecognized important factor regulating the decidual remodeling of the endometrium, decidual-trophoblast interactions, and trophoblast differentiation and invasion during placental development. Future studies will establish the exact mechanism of action and physiological significance of HTRA3 in placental development.

ACKNOWLEDGMENTS

We thank Anne Hampton for technical assistance, S. Panckridge for help with the figures, and Dianne Arnold for assisting with online submission of the manuscript.

FOOTNOTES

1 Supported by the Rockefeller/World Health Organization (WHO) Initiative on Implantation (RF 99021#116) and the NH&MRC of Australia (#143798 and #241000). Back

2 Correspondence: Guiying Nie, Prince Henry's Institute of Medical Research, PO Box 5152, 246 Clayton Rd., Clayton, VIC 3168, Australia. FAX: 61 3 9594 6125; guiying.nie{at}princehenrys.org Back

3 Current address: Department of Obstetrics and Gynecology, Kansai Medical University, Moriguchi 570-8507, Japan. Back

Received: 7 September 2005.

First decision: 10 October 2005.

Accepted: 21 October 2005.

REFERENCES

  1. Wynn RM, Ultrastructural development of the human decidua. Am J Obstet Gynecol 1974 118:652-670[Medline]
  2. Loke YW, King A, Human Implantation: Cell Biology and Immunology New York The Press Syndicate of University of Cambridge 1995
  3. Aplin J, Maternal influences on placental development. Semin Cell Dev Biol 2000 11:115-125[CrossRef][Medline]
  4. Lim H, Paria BC, Das SK, Dinchuk JE, Langenbach R, Trzaskos JM, Dey SK, Multiple female reproductive failures in cyclooxygenase 2-deficient mice. Cell 1997 91:197-208[CrossRef][Medline]
  5. Gendron RL, Paradis H, Hsieh-Li HM, Lee DW, Potter SS, Markoff E, Abnormal uterine stromal and glandular function associated with maternal reproductive defects in Hoxa-11 null mice. Biol Reprod 1997 56:1097-1105[Abstract]
  6. Nie G-Y, Li Y, Minoura H, Batten L, Ooi GT, Findlay JK, Salamonsen LA, A novel serine protease of the mammalian HTRA family is up-regulated in mouse uterus coinciding with placentation. Mol Hum Reprod 2003 9:279-290[Abstract/Free Full Text]
  7. Nie GY, Hampton A, Li Y, Findlay JK, Salamonsen LA, Identification and cloning of two isoforms of human HTRA3, characterisation of its genomic structure and comparison of its tissue distribution with HTRA1 and HTRA2. Biochem J 2003 371:39-48[CrossRef][Medline]
  8. Clausen T, Southan C, Ehrmann M, The HTRA family of proteases: implications for protein composition and cell fate. Mol Cell 2002 10:443-455[CrossRef][Medline]
  9. Nie G, Li Y, He H, Findlay JK, Salamonsen LA, HTRA3, a serine protease possessing an IGF binding domain, is selectively expressed at the maternal-fetal interface during placentation in the mouse. Placenta published online ahead of press 10 June 2005
  10. Nie GY, Li Y, Minoura H, Findlay JK, Salamonsen LA, Specific and transient up-regulation of proprotein convertase 6 at the site of embryo implantation and identification of a unique transcript in mouse uterus during early pregnancy. Biol Reprod 2003 68:439-447[Abstract/Free Full Text]
  11. Baker RMKT, Maastricht S, Rose M, Smith H, Watson D, (eds.) The Use of Immuno-Adjuvants in Animals in Australia and New Zealand Adelaide Australian and New Zealand Council for the Care of Animals in Research and Teaching Limited 1998
  12. Okada H, Nie G, Salamonsen LA, Requirement for proprotein convertase 5/6 during decidualization of human endometrial stromal cells in vitro. J Clin Endocrinol Metab 2005 90:1028-1034[Abstract/Free Full Text]
  13. Irving JA, Lala PK, Functional role of cell surface integrins on human trophoblast cell migration: regulation by TGFB IGF2 and IGFBP1. Exp Cell Res 1995 217:419-427[CrossRef][Medline]
  14. Hamilton GS, Lysiak JJ, Han VKM, Lala PK, Autocrine-paracrine regulation of human trophoblast invasiveness by insulin-like growth factor (IGF2) and IGF-binding protein (IGFBP)-1. Exp Cell Res 1998 244:147-156[CrossRef][Medline]
  15. Chakraborty C, Gleeson LM, McKinnon T, Lala PK, Regulation of human trophoblast migration and invasiveness. Can J Physiol Pharmacol 2002 80:116-124[CrossRef][Medline]
  16. Constancia M, Hemberger M, Hughes J, Dean W, Ferguson-Smith A, Fundele R, Stewart F, Kelsey G, Fowden A, Sibley C, Reik W, Placental-specific IGF2 is a major modulator of placental and fetal growth. Nature 2002 417:945-948[CrossRef][Medline]
  17. Nayak NR, Giudice LC, Comparative biology of the IGF system in endometrium, decidua, and placenta and clinical implications for foetal growth and implantation disorders. Placenta 2003 24:281-296[CrossRef][Medline]
  18. Irwin JC, Suen L-F, Martina NA, Mark SP, Giudice LC, Role of the IGF system in trophoblast invasion and pre-eclampsia. Hum Reprod 1999 14:90-96[Abstract/Free Full Text]
  19. Carter AM, Hills F, O'Gorman DB, Roberts CT, Sooranna SR, Watson CS, Westwood M, The insulin-like growth factor system in mammalian pregnancy—a workshop report. Placenta 2004 25:S53-S56
  20. Pollheimer J, Knofler M, Signalling pathways regulating the invasive differentiation of human trophoblasts: a review. Placenta 2005 26:S21-S30
  21. Oka C, Tsujimoto R, Kajikawa M, Koshiba-Takeuchi K, Ina J, Yano M, Tsuchiya A, Ueta Y, Soma A, Kanda H, Matsumoto M, Kawaichi M, Htra1 serine protease inhibits signaling mediated by TGFB family proteins. Development 2004 131:1041-1053[Abstract/Free Full Text]
  22. Tocharus J, Tsuchiya A, Kajikawa M, Ueta Y, Oka C, Kawaichi M, Developmentally regulated expression of mouse HTRA3 and its role as an inhibitor of TGF-beta signaling. Dev Growth Differ 2004 46:257-274[CrossRef][Medline]
  23. Hou J, Clemmons DR, Smeekens S, Expression and characterization of a serine protease that preferentially cleaves insulin-like growth factor binding protein-5. J Cell Biochem 2005 94:470-484[CrossRef][Medline]
  24. Tsuchiya A, Yano M, Tocharus J, Kojima H, Fukumoto M, Kawaichi M, Oka C, Expression of mouse Htra1 serine protease in normal bone and cartilage and its upregulation in joint cartilage damaged by experimental arthritis. Bone 2005 37:323-336[Medline]
  25. Godkin J, Dore J, Transforming growth factor beta and the endometrium. Rev Reprod 1998 3:1-6[Abstract]
  26. Lala PK, Hamilton GS, Growth factors, proteases and protease inhibitors in the maternal–fetal dialogue. Placenta 1996 17:545-555[CrossRef][Medline]
  27. Lash GE, Otun HA, Innes BA, Bulmer JN, Searle RF, Robson SC, Inhibition of trophoblast cell invasion by TGFB1, 2, and 3 is associated with a decrease in active proteases. Biol Reprod 2005 73:374-381[Abstract/Free Full Text]
  28. Han VK, Bassett N, Walton J, Challis JR, The expression of insulin-like growth factor (IGF) and IGF-binding protein (IGFBP) genes in the human placenta and membranes: evidence for IGF-IGFBP interactions at the feto-maternal interface. J Clin Endocrinol Metab 1996 81:2680-2693[Abstract]
  29. Crossey PA, Pillai CC, Miell JP, Altered placental development and intrauterine growth restriction in IGF binding protein-1 transgenic mice. J Clin Invest 2002 110:411-418[CrossRef][Medline]
  30. Clemmons DR, Role of insulin-like growth factor binding proteins in controlling IGF actions. Mol Cell Endocrinol 1998 140:19-24[CrossRef][Medline]
  31. Gibson JM, Aplin JD, White A, Westwood M, Regulation of IGF bioavailability in pregnancy. Mol Hum Reprod 2001 7:79-87[Abstract/Free Full Text]
  32. Myers SE, Cheung PT, Handwerger S, Chernausek SD, Insulin-like growth factor-I (IGFI) enhanced proteolysis of IGF-binding protein-4 in conditioned medium from primary cultures of human decidua: independence from IGF receptor binding. Endocrinology 1993 133:1525-1531[Abstract]
  33. Sun IY, Overgaard MT, Oxvig C, Giudice LC, Pregnancy-associated plasma protein a proteolytic activity is associated with the human placental trophoblast cell membrane. J Clin Endocrinol Metab 2002 87:5235-5240[Abstract/Free Full Text]
  34. Coppock HA, White A, Aplin JD, Westwood M, Matrix metalloprotease-3 and –9 proteolyze insulin-like growth factor-binding protein-1. Biol Reprod 2004 71:438-443[Abstract/Free Full Text]
  35. Lysiak JJ, Hunt J, Pringle GA, Lala PK, Localization of transforming growth factor beta and its natural inhibitor decorin in the human placenta and decidua throughout gestation. Placenta 1995 16:221-231[CrossRef][Medline]
  36. Xu G, Guimond M-J, Chakraborty C, Lala PK, Control of proliferation, migration, and invasiveness of human extravillous trophoblast by decorin, a decidual product. Biol Reprod 2002 67:681-689[Abstract/Free Full Text]
  37. Floridon C, Nielsen O, Holund B, Sunde L, Westergaard JG, Thomsen SG, Teisner B, Localization of E-cadherin in villous, extravillous and vascular trophoblasts during intrauterine, ectopic and molar pregnancy. Mol Hum Reprod 2000 6:943-950[Abstract/Free Full Text]




This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
74/2/366    most recent
biolreprod.105.047324v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Nie, G.
Right arrow Articles by Salamonsen, L. A.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Nie, G.
Right arrow Articles by Salamonsen, L. A.
Agricola
Right arrow Articles by Nie, G.
Right arrow Articles by Salamonsen, L. A.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS