Introduction

Top Abstract Introduction Materials and Methods Results Discussion References

[D-Ala²]deltorphin I (DADTI) is an amidated opioid heptapeptide (Tyr-D-Ala-Phe-Asp-Val-Val-Gly.NH₂) isolated from the skin of the South American frog Phyllomedusa bicolor, which was demonstrated to have high affinity and selectivity for -opioid receptors (1). Prodeltorphin, identified by complementary DNA cloning from the same frog skin, was found to contain three DADTI sequences and one [D-Ala²]deltorphin II (DADTII) sequence (Tyr-D-Ala-Phe-Glu-Val-Val-Gly.NH₂) (2). A rabbit anti-DADTI antiserum raised to amidated DADTI (3) was used for immunostaining, with most initial studies focused on the central nervous system. DADTI-like immunoreactivity (DADTI-LI) has also been localized to bronchiolar epithelial cells in developing rat lung, in which the highest levels occur shortly before birth (4).

Endogenous neuropeptides can function to promote cell proliferation and differentiation in developing lung, as demonstrated for bombesin-like peptide, derived from fetal pulmonary neuroendocrine cells (5). Other neuropeptides present in these cells include calcitonin gene-related peptide, corticotropin, and leu-enkephalin, but the developmental role for these peptides was unexplained until recently (5). Whole-lung and many lung-cancer cell lines contain opioid immunoreactivity (-endorphin, enkephalin, and/or dynorphin) (6, 7), and messenger RNAs (mRNAs) encoding preprodynorphin and/or preproenkephalin (8, 9). Met-enkephalin immunoreactivity is present in pulmonary nerve fibers (10), which might contain prodermorphin (11).

Fetal cells with DADTI-LI in rat lung were of interest to us because of the nonclassical localization of this neuropeptide immunoreactivity (neither in neuroendocrine cells nor nerves). Further, it appeared likely that such DADTI-LI might have a function during lung development. DADTI or DADTII (the soluble equipotent analog of DADTI) injected into rat cerebral ventricles induces several central effects, including locomotor activity, grooming, facilitated social contacts, and memory consolidation (12). However, there were no reports of effects of deltorphins outside the central nervous system. Several studies analyzed effects of other opiate-like peptides on lung development (13). Taeusch and colleagues demonstrated that heroin given to pregnant rabbits induced fetal lung maturation together with decreased fetal body weight (13). Later, fetal rabbits treated with morphine were shown to have increased alveolar stability on deflation, whereas animals receiving the µ-blocker naloxone had decreased alveolar stability, suggesting a physiologic role for endogenous opioids (14). Central but not peripheral administration of -endorphin in developing rats enhanced lung maturation indirectly via glucocorticoid elevation (15). Finally, -, -, or µ-specific opioids can inhibit proliferation of lung cancer cell lines (7).

We have established a murine lung explant system for analyzing direct effects of bioactive peptides on lung development (5). In the present study, we localized DADTI-LI in developing mouse lung, then investigated the effects of DADTII versus dermorphin (µ-receptor-specific) in fetal lung explants. We assessed lung maturation using a combination of [³H]-choline incorporation and prosurfactant protein (proSP)-C immunostaining, and cell proliferation using [³H]-thymidine incorporation as well as proliferating cell nuclear antigen (PCNA) immunostaining. Finally, gene expression for the three major opiate receptor subclasses was determined during mouse lung ontogeny by semiquantitative reverse-transcribed polymerase chain reaction (RT-PCR) and immunostaining to investigate which receptors could be involved in these effects.

	Materials and Methods

Top Abstract Introduction Materials and Methods Results Discussion References

Antisera and Antibodies

Antiserum to DADTI was reported previously (3). Antiserum to the 10-kD murine Clara cell protein CC10 was a gift from Dr. G. Singh, University of Pittsburgh (Pittsburgh, PA). Antiserum to the type II cell marker proSP-C was provided by Dr. Jeff Whitsett (University of Cincinnati (Cincinnati, OH)). Antiserum to protein gene product 9.5 (PGP9.5) was purchased from Ultraclone, Ltd. (Isle of Wight, UK). Anti-rat calcitonin gene-related peptide (CGRP) was from Peninsula Laboratories (Belmont, CA). Monoclonal antibody to PCNA was purchased from Dako (Carpinteria, CA) as described previously (16). Antisera and the matching antigenic peptide to the murine µ-opiate receptor was obtained from Oncogene (Oncogene Research Products, Boston, MA). Antisera and corresponding antigens specific for the murine - and -opiate receptors were from Santa Cruz Biotechnology (Santa Cruz, CA).

Animals

Timed-pregnant and Swiss-Webster mice were obtained from Taconic Laboratories (Germantown, NY) at embryonic Day 7 (e7) and allowed to rest for at least 3 d before harvest. Lung tissue was removed from untreated fetal, newborn, and adult (6-wk old) mice between e10 and 4 mo of age for immunohistochemistry, lung organ culture, and RNA preparation as described (5). Fetal lungs were fixed by immersion in 4% paraformaldehyde, whereas postnatal and adult specimens were inflated via intratracheal instillation in situ, after opening the rib cage but before removing the lungs. The care of these mice was in accordance with National Institutes of Health and Harvard Medical School guidelines.

Immunoperoxidase Analyses

For DADTI, proSP-C, and PCNA immunostaining, fetal lungs were fixed overnight in 4% paraformaldehyde and processed into paraffin. For the opiate-receptor immunostaining, fetal lungs were flash-frozen; 5-µm frozen sections were cut and postfixed in 4% paraformaldehyde for 5 min. Thin (3-µm) sections were used for immunoperoxidase analyses as described (3), with two variations: for DADTI-LI, sections were incubated with 0.3% Triton-X 100 before primary antiserum (1:1,000 dilution); and the avidin-biotin complex immunoperoxidase was followed by biotinylated tyramide amplification (tyramide system amplification [TSA]) and diaminobenzidine as substrate. Immunostaining for proSP-C and PCNA were performed as described elsewhere (16).

DADTI antiserum specificity was thoroughly characterized previously and has negligible immunoreactivity toward the carrier poly-L-glutamate (3). In the current study, negative controls for immunohistochemistry included: (1) replacing primary antiserum with nonimmune rabbit serum; or (2) using antigen-absorbed DADTI antiserum (1:1,000 dilution of DADTI antiserum) incubated overnight at 4°C using 2 or 200 µM DADTI (1.8 or 180 µg/ml, respectively).

RNA Analyses

Total RNA was prepared and RT-PCR was performed as detailed previously (16). RT-PCR, required to detect mRNAs encoding -, -, and µ-receptors in small tissue samples, used 35 cycles of PCR for the receptors and 22 cycles for -actin, each cycle including denaturation (0.5 min, 93°C), annealing (1 min, 55°C), and extension (3 min, 72°C). Oligodeoxynucleotide pairs (from Oligos Etc., Inc., Wilsonville, OR) spanned  1 intron, corresponding to sequences of murine -opiate receptor (yielding a 417-base pair [bp] product) (17), -opiate receptor (540-bp product) (18), µ-opiate receptor (494-bp product) (19), and -actin (given elsewhere) (16). Adult mouse brain RNA was used as the positive control. PCR Southern blots were probed with specific end-labeled internal oligonucleotides. The primer sequences used were: for -receptor: 5': TTCCCCTTAAACGCCCCTC, 3': TTG CCTGAATGCCCTGTCC, probe: GCCAAACCAAATCATC GAC; for -receptor: 5': ACGGTGACTTGGGAAGGGAG, 3': GGCACACAGCAATGTAGCGG, probe: CCAGAGAATTGC CCACTAAG; and for µ-receptor: 5': GGCAACCAGTCCGA CCCATG; 3': GGGCAGACCAATGGCAGAAG, probe: TGA TGGTGATGGCTGTGAC.

Lung Organ Cultures and Determination of [³H]DNA, [³H]Saturated Phosphatidylcholine, DNA, and Protein Content

Fetal lung explants were cultured for 48 h total, with [³H]-thymidine (4 µCi/ml) or [³H] -choline (16 µCi/ml) added for the last 4 h, as detailed elsewhere (5). Acid-precipitable [³H]-labeled DNA and DNA content were determined as described (5). The rate of surfactant phospholipid synthesis was assayed using [³H]choline incorporation into saturated phosphatidylcholine (DSPC), normalized for protein content, as described (5). Experimental results were further normalized by expressing values as percentage changes above or below baseline (baseline is defined as the mean of the untreated control group).

Statistical Analyses

Numerical data were analyzed using the unpaired Student's t test with values expressed as means ± standard error (SE). Statistical significance was accepted at P < 0.05.

	Results

Top Abstract Introduction Materials and Methods Results Discussion References

Immunohistochemical Analyses

The distribution of DADTI-LI was mapped spatially and temporally in developing mouse lung between e12 and postnatal Day 14 (P14), and in adult mice at 6 wk and 4 mo of age. DADTI-LI is first detectable on e15 in occasional cuboidal-to-columnar epithelial cells of developing bronchioles close to the transition zone into the developing acinus (Figure 1A). Similar cells are present in lung sections from about half of the mice between e15 and e18 (three mice analyzed per day), with the greatest relative numbers present on e15 and the lowest number of these cells occurring on e18. The location of these cells with DADTI-LI suggests they could be developing Clara cells in the epithelium of conducting airways, but immunostaining for the Clara cell-specific antigen CC10 was undetectable at this early stage of fetal lung development (our unpublished data). Also, these cells do not immunostain for the neuroendocrine cell markers CGRP and PGP9.5.

View larger version (106K): [in this window] [in a new window]   View larger version (124K): [in this window] [in a new window]   Figure 1.   DADTI-LI in developing mouse lung. Thin sections were immunostained as detailed in MATERIALS AND METHODS using diaminobenzidine as substrate (dark brown) and methyl green as counterstain. The very dark brown (near-black) immunostaining at the center of DADTI-LI- positive cells represents high intensity of staining with diaminobenzidine, versus the moderate brown staining at the periphery of the cells corresponding to less intense staining. (A) Section of e15 mouse lung demonstrating DADTI-LI in a collection of columnar bronchiolar epithelial cells consistent with Clara cell precursors (original magnification: ×132). (B) Subpleural region of e18 fetal lung contains numerous DADTI-LI-positive cells projecting into primitive alveolar spaces. The few cells that appear to be located in the primitive alveolar septae were identified later as alveolar lining cells in adjacent sections (our unpublished data) (original magnification: ×160). (C) On e18, many future alveoli contain whorls of DADTI-LI-positive material similar to that observed by immunostaining for SPs, suggesting that this immunoreactive material is secreted into developing air spaces (original magnification: ×160). (D) After birth and up to P14, many DADTI-LI-positive cells were scattered throughout the developing alveoli, generally most concentrated in the subpleural region (original magnification: ×66). (E and F ) A higher-power view of P14 lung demonstrates more flattened DADTI-LI-positive alveolar lining cells, some with dendritic processes appearing to extend along the alveolar surface epithelium (arrows) (original magnification: ×160). (G, I, and K ). In e18 lung, DADTI-LI-positive cells are indicated by arrows, including one endothelial cell in the upper right-hand corner of (G) (original magnifications: G, ×132; I, ×100; K, ×160). (H ) A serial section adjacent to G run in parallel using preabsorbed antiserum (200 µM DADTI/ml) has no evidence of DADTI-LI (original magnification: ×132). (J ) Similarly, a section adjacent to I run in parallel with nonimmune rabbit serum replacing primary antiserum has no immunostaining (original magnification: ×100). (L) A serial section adjacent to K demonstrates proSP-C-immunopositive alveolar cells. Arrows indicate the locations of the cells with DADTI-LI in K and L (original magnification: ×160).

The most abundant DADTI-LI in developing mouse lung occurred in cells lining primitive air spaces in fetal mice (Figures 1B, 1C, 1G, 1I, and 1K) and postnatally in alveoli (Figures 1D, 1E, and 1F). These cells appear to be concentrated in the subpleural lung (Figures 1B, 1D, 1I, and 1K). During fetal gestation, the cells were generally cuboidal and protruded into the future air-space lumen (Figure 1B), although more elongated cells were occasionally observed (Figure 1E, arrow, upper right-hand corner). After birth, relative numbers of the alveolar cells with DADTI-LI declined, being infrequent by 14 d of age and undetectable by 6 wk and 4 mo of age (our unpublished data). The cells tended to be more elongated and did not generally protrude into the alveolar space in air-breathing postnatal lung (Figures 1D, 1E, and 1F). Occasional endothelial cells were also positive for DADTI-LI (Figure 1G, arrow, upper right-hand corner). A few cells had spindle-shaped extensions, possibly dendritic processes (Figures 1E and 1F, arrows). Intra-alveolar DADTI-LI-positive proteinaceous material appearing similar to surfactant proteins (SPs) was observed in immersion-fixed fetal lung specimens (Figure 1C), suggesting that this material with DADTI-LI might be secreted.

Antigen specificity of DADTI-LI was confirmed by absorption of the antiserum with DADTI peptide before immunoperoxidase analyses. DADTI-LI of cells and intra- alveolar material (Figures 1G and 1I) was abrogated in serial sections using 200 µM DADTI-preabsorbed antiserum (Figure 1H) or nonimmune rabbit serum (Figure 1J). The quantity of 2 µM DADTI was not sufficient to remove immunoreactivity (our unpublished data), consistent with prior observations (4).

To determine whether DADTI-LI-positive cells are classic type II pneumocytes, we stained thin sections of e18 mouse lung with anti-DADTI (Figure 1K) or anti-proSP-C (Figure 1L). The DADTI-LI-positive alveolar lining cells (Figure 1K) were consistently negative for proSP-C immunostaining (Figure 1L).

Fetal Lung Organ Cultures

Direct effects of DADTII and dermorphin on developing lung (Figure 2) were analyzed in fetal mouse lung explants, a system which is well established in our laboratory (5). To assess type II cell differentiation, we first analyzed incorporation of [³H]choline into DSPC, the rate-limiting step in surfactant phospholipid synthesis; as applied, this assay is type II cell-specific. Second, we evaluated proSP-C immunostaining, in terms of both the numbers of positive cells and the intensity of immunostaining. To assess cell proliferation, we first quantitated incorporation of [³H]thymidine into acid-precipitable counts, which we have shown to represent incorporation into nuclear DNA (5). Second, we assessed PCNA immunostaining to determine which cells had altered proliferation.

View larger version (17K): [in this window] [in a new window]   View larger version (17K): [in this window] [in a new window]   View larger version (68K): [in this window] [in a new window]   View larger version (15K): [in this window] [in a new window]   View larger version (16K): [in this window] [in a new window]   View larger version (84K): [in this window] [in a new window]   Figure 2.   Effects of DADTII versus dermorphin on type II cell differentiation and cell proliferation in fetal murine lung explants. (A and B) E17 lung explants were cultured for 48 h with or without 0.1, 1, or 10 nM of (A) DADTII or (B) dermorphin (Dermor). [3H]choline incorporation was determined and normalized for total protein content of each sample as detailed in MATERIALS AND METHODS. Data are expressed as means ± SEM. A: *P < 0.0002; B: P < 0.004. Mean baseline [3H]choline incorporation on e17 was 172 ± 18 dpm/mg protein. (C ) Immunostaining for proSP-C was carried out using tissue sections from lung explants cultured for 5 d. A representative section is given from the subpleural region. Similar results were obtained in three separate experiments. Note that treatment with 1 nM DADTII resulted in fewer proSP-C-positive cells with weaker immunostaining, whereas dermorphin led to both an increase in the number of SP-C-positive cells and stronger immunostaining. (original magnification: × 20). (D and E) e18 lung explants were cultured for 48 h with and without 0.1, 1, or 10 nM of (D) DADTII or (E ) Dermor. [3H]thymidine incorporation was determined and normalized for total DNA content of each sample as detailed in MATERIALS AND METHODS. D: P < 0.008; E: P < 0.0015. Mean baseline [3H]-thymidine incorporation on e18 was 1,367 ± 60 dpm/mg DNA. (F ) PCNA immunostaining was carried out using tissue sections from lung explants cultured for 5 d. A representative section is given including a large conducting airway and distal parenchyma. Similar results were obtained in three separate experiments. (original magnification: ×10). Note that treatment with 0.1 nM DADTII resulted in increased numbers of PCNA-positive cells in the conducting airways, whereas 0.1 nM dermorphin inhibited PCNA labeling in both airway epithelial cells and in the parenchyma. A, B, D, and E: Data shown represent the results of four experiments (each carried out with four samples per group), which were pooled after expressing the data from each experiment as the percent change over baseline values (that is, in the cultures without added peptides).

Choline incorporation was assessed on e17 because consistent, significant differences between experimental groups were observed on this day, corresponding to the time of increased production of surface-active material by fetal mouse lung (5). We used DADTII instead of DADTI for the functional studies because: (1) DADTII is soluble in media, whereas DADTI is not; (2) DADTII and DADTI are both derived from a single gene product by post-translational processing; and (3) DADTII has physiologic effects essentially identical to those of DADTI in experimental systems (1). The results of choline incorporation in e17 fetal mouse lung explants cultured for 48 h with 0.1 to 10 nM of DADTII or dermorphin are given in Figures 2A and 2B. There was significant inhibition of choline-incorporation in the presence of 1 nM DADTII (22% inhibition, P < 0.0002). This was a dose-dependent response in that 0.1 or 10 nM of DADTII had no significant effect, consistent with a receptor-mediated process. In contrast, 1 nM dermorphin increased choline incorporation by 22% (P < 0.004), again in a dose-dependent manner (Figure 2B). Although these effects on choline incorporation into DSPC were modest (~ 20% change compared with baseline), when four experiments were pooled the results were highly consistent from one experiment to another and were of high statistical significance.

To assess a second parameter of type II cell differentiation, we carried out immunostaining of lung explants for proSP-C. Representative results from one of four experiments are in Figure 2C. Compared with untreated controls (Figure 2C, left panel), treatment of lung explants with 1 nM DADTII (Figure 2C, middle panel) led to a decrease in proSP-C immunostaining, in terms of both the numbers of positive cells and the intensity of immunostaining. In contrast, lung explants cultured with 1 nM dermorphin demonstrated an increase in the number of proSP-C-positive cells and increased intensity of immunostaining.

Thymidine incorporation was evaluated on e18 because this was the time point for eliciting maximal differences between the experimental groups in the current study (Figure 2D and 2E). As shown, 0.1 nM DADTII increased thymidine incorporation in explants (50% increase, P < 0.008). This was a dose-dependent response in that 1 and 10 nM DADTII had no effect in these explants. In contrast, 0.1 nM dermorphin inhibited thymidine incorporation ~ 32% (P < 0.0015), also in a dose-dependent fashion in that 1 or 10 nM dermorphin had no effect.

To assess a second parameter of cell proliferation and to localize which cell populations were proliferating, we carried out immunostaining of lung explants for PCNA. Representative results from one of four experiments are given in Figure 2F. Compared with untreated controls (Figure 2F, left panel), lung explants treated with 0.1 nM DADTII (Figure 2F, middle panel) had increased numbers of PCNA-positive cells, which were predominantly epithelial cells of the conducting airways. Conversely, lung explants cultured with 0.1 nM dermorphin had fewer PCNA-positive cells both in the airway epithelium and in the distal lung parenchyma. Approximately half of these parenchymal PCNA-positive cells were mesenchymal (interstitial) in origin.

The next series of choline incorporation studies addressed the question of receptor specificity of these responses elicited by DADTII and dermorphin. Lung explants were treated with either 1 nM DADTII (Figure 3A) or 1 nM dermorphin (Figure 3B) ± 100 nM of blocking agents: binaltorphimine (-receptor-specific, abbreviated as Binal in Figure 3), naloxone (µ-receptor-specific), or naltrindole (-receptor-specific). Although 100 nM naltrindole alone did not alter choline incorporation, it did significantly reverse DADTII-mediated inhibition of choline incorporation (P < 0.02). Unexpectedly, although 100 nM binaltorphimine had no significant effect on choline incorporation, this -specific blocking agent was also able to reverse DADTII-mediated inhibition of choline incorporation, albeit at a marginally significant level (P < 0.04). Although 100 nM naloxone did elicit significant stimulation of choline incorporation (P < 0.005; Figure 4), naloxone did not have any significant effect on DADTII-induced choline incorporation. In a parallel series of experiments, the stimulation of choline incorporation by dermorphin was completely reversed by binaltorphimine or by naltrindole (P < 0.0001) and, less significantly, by naloxone (P < 0.01).

View larger version (19K): [in this window] [in a new window]   View larger version (19K): [in this window] [in a new window]   View larger version (18K): [in this window] [in a new window]   View larger version (18K): [in this window] [in a new window]   Figure 3.   Modulation of DADTII and dermorphin effects on fetal murine lung by -, -, and µ-opiate receptor blockers. (A and B) [3H]choline incorporation in e17 fetal lung and (C and D) [3H]thymidine incorporation in e18 fetal lung were determined after 48 h of culture with optimal doses of DADTI (A: 1 nM; C: 0.1 nM) or dermorphin (B: 1 nM; D: 0.1 nM) without or with 100 nM of naltrindole, binaltorphimine, or naloxone (which block -, -, and µ-opiate receptors). Results of controls using the blockers alone are also given. Results were pooled from four independent experiments; each experiment was assayed in quadruplicate. P values compare cultures with and without a specific blocker.

View larger version (31K): [in this window] [in a new window]   View larger version (30K): [in this window] [in a new window]   Figure 4.   Effects of -, -, and µ-opiate receptor blockers on choline and thymidine incorporation in fetal murine lung explants. e17 (A) and e18 (B) fetal lung explants were cultured for 48 h with and without 10 or 100 nM of naltrindole, binaltorphimine, or naloxone alone. Values are expressed as percent change over baseline cultures without added peptide. P values are comparing peptide-treated to untreated cultures.

The next series of thymidine-incorporation experiments addressed the question of opiate-receptor specificity of responses elicited by DADTII and dermorphin. Lung explants were treated with 0.1 nM DADTII (Figure 3C) or 0.1 nM dermorphin (Figure 3D) ± 100 nM of binaltorphimine, naloxone, or naltrindole. As shown in Figure 3C, although 100 nM naltrindole alone had no effect on baseline values, naltrindole increased DADTII-induced thymidine incorporation (P < 0.001). Moreover, although 100 nM naloxone augmented baseline thymidine incorporation (P < 0.03), the addition of naloxone to DADTII cultures reversed the stimulatory effect of DADTII on thymidine incorporation (P < 0.005). In parallel, the inhibition of thymidine incorporation by dermorphin was completely reversed by binaltorphimine (P < 0.016), but this antagonist had no effect on baseline thymidine incorporation. Surprisingly, there was no effect of naloxone on dermorphin inhibition of thymidine incorporation, although naloxone alone was stimulatory.

In view of these results suggesting that multiple opiate-receptor subclasses might regulate lung developmental processes, we carried out both e17 choline incorporation and e18 thymidine incorporation experiments with 10 versus 100 nM of binaltorphimine, naloxone, or naltrindole. The pooled results of four experiments are given in Figure 4. Naloxone stimulated choline incorporation at 100 (P < 0.005) and 10 nM (P < 0.02), and also increased thymidine incorporation ~ 20% at 100 (P < 0.03). Neither 10 nor 100 nM of naltrindole had any effect on choline incorporation, but 10 nM naltrindole stimulated thymidine incorporation by ~ 25% (P < 0.03). Binaltorphimine (10 nM) had only a marginal effect, increasing thymidine incorporation by ~ 11% (P < 0.05).

Analyses of Opiate Receptor Gene Expression in Developing Mouse Lung

To determine which of the known -, -, and µ-specific opiate receptor genes are expressed in mouse lung ontogeny, we used semiquantitative RT-PCR analyses of total lung RNA from e11 to P14, normalized for -actin. The results are shown in Figure 5. First, -receptor mRNAs were expressed at all time points in embryonic, fetal, postnatal, and adult mouse lung. Second, mRNAs encoding -receptors were expressed in a bimodal distribution from e11 to e16, and again at low levels on the day after birth. mRNAs encoding -receptors were also detected at e10 (our unpublished data). Finally, µ-receptor mRNAs were present from e11 to P14, with highest levels occurring at: e11 to e13, the embryonic period during which lung branching morphogenesis takes place; e15 to e16, the early canalicular phase during which type II cell differentiation begins; e18 to P1, the saccular period during which perinatal adaptation takes place; and P14, when most of alveolarization is complete. Interestingly, µ-receptors were not expressed during alveolarization at P2 to P7.

View larger version (53K): [in this window] [in a new window]   Figure 5.   Analyses of -, -, and µ-opiate receptor gene expression in developing mouse lung. (A) Total RNA was prepared and RT-PCR analyses carried out as described in MATERIALS AND METHODS, using conditions defined for detection of positive control samples in the midlinear range, with linear detection between 0.01 and 10 µg of RNA in parallel RT reactions. Comparison with -actin amplification in the same RT samples is shown. This RT-PCR analysis of the three opiate receptors was carried out twice with distinct pools of mouse fetal lungs, each time by a different trained individual, with similar results. Each lane represents pooled lungs from two to five litters (24 to 60 pups).

To analyze which cell types express -, -, and µ-opiate receptor proteins in fetal mouse lung, we carried out immunostaining for the three receptors in e18 mouse lung. We chose this time point to coincide with the timing of the lung organ culture experiments. Representative results are shown in Figure 6. In Figure 6A, the strongest immunostaining for -opiate receptors was demonstrated in alveolar epithelial cells (arrowheads) and interstitial mesenchymal cells (arrows). There was also weak to moderate immunostaining of bronchiolar epithelial cells and vascular endothelial cells. Immunostaining for -opiate receptors was localized to alveolar epithelial cells and interstitial mesenchymal cells (Figure 6C). The µ-opiate receptor immunostaining was also most prominent in alveolar epithelial cells (Figures 6E and 6F, arrowheads), and was also present at lower levels on interstitial mesenchymal cells, airway epithelial cells and vascular endothelial cells. All specific antigen-absorbed controls (Figures 6B, 6D, and 6F) were devoid of immunostaining.

View larger version (149K): [in this window] [in a new window]   Figure 6.   Immunostaining for -, -, and µ-opiate receptors in e18 murine lung. Frozen sections of e18 mouse lung were immunostained as detailed in MATERIALS AND METHODS. A, C, and E are immunostained for -, -, and µ-opiate receptors. B, D, and F are the corresponding antigen-absorbed control slides with a serial section of lung tissue. (A and B) The strongest immunostaining for -opiate receptors occurred in alveolar epithelial cells (arrowheads) and interstitial mesenchymal cells (arrows). There was also weak to moderate immunostaining of bronchiolar epithelial cells and vascular endothelial cells (b = bronchiole; v = blood vessel). (Original magnification ×20). (C and D) Immunostaining for -opiate receptors was localized to alveolar epithelial cells (arrowheads) and interstitial mesenchymal cells (arrows). There was no appreciable immunostaining of airway epithelial cells and vascular endothelial cells. (E and F ) Immunostaining for µ-opiate receptors was most prominent in alveolar epithelial cells (arrowheads), with moderate immunopositivity also present on interstitial mesenchymal cells (arrows). There was weak immunostaining of airway epithelial cells and vascular endothelial cells.

	Discussion

Top Abstract Introduction Materials and Methods Results Discussion References

The present study demonstrates a role for endogenous deltorphins in fetal lung development. DADTI-LI is present in a distinct population of alveolar lining epithelial cells, which appears negative for proSP-C immunostaining. It should be noted that these results are not absolutely quantitative, but that results of our immunostaining are consistent with alveolar lining cells with low levels of surfactant protein (SP)-C gene expression. There are reports of similar cells with low levels of SP-C gene expression but high levels of vascular endothelial growth factor gene expression (20). Further, these are not typical neuroendocrine cells because they are negative for the pan-neuroendocrine marker PGP9.5 (5). Our cumulative observations are most consistent with DADTI-LI-positive alveolar cells being undifferentiated type II cells with very low levels of proSP-C protein immunoreactivity. Alternatively, these cells could comprise a previously unrecognized epithelial cell subset. In contrast to our observations in the mouse, DADTI-LI-positive cells were localized to bronchiolar epithelial cells in fetal rat lung (4). Our findings in the mouse support and extend those earlier experiments because we did observe DADTI-LI-positive bronchiolar epithelial cells in fetal mouse lung. We can only visualize the DADTI-LI- positive alveolar cells using a newly developed immunoperoxidase amplification method, the TSA reagent. We did not observe any increase in numbers of proSP-C-positive cells using the TSA reagent. Further, differences in antigen distribution are not uncommon when two species are compared. For instance, bombesin-like immunoreactivity is visible in human, but not rodent, pulmonary neuroendocrine cells (21).

Functionally, we demonstrate a role for deltorphins in promoting cell proliferation (assessed using thymidine incorporation and PCNA immunostaining) and inhibiting type II cell differentiation (evaluated using choline incorporation and proSP-C immunostaining) in e17-e18 fetal mouse lung explants. All three major classes of opiate receptor subtypes are potentially involved in these responses because -, -, and µ-receptor mRNAs are present in lung throughout gestation and the corresponding proteins were detected on e18. Similar to our results, -receptor mRNA was the first receptor detected in developing mouse embryos (22). It is likely that at least part of the DADTII-induced inhibition of choline incorporation occurs via -receptors because this inhibition of choline incorporation was completely reversed by the -receptor blocking agent naltrindole, -receptor mRNAs were expressed in late-gestation fetal lung, and -receptor protein was localized to type II cells at high levels. Other investigators have reported that -receptors can transduce mitogenic responses in rat fibroblasts (17). In contrast, dermorphin promotes type II cell differentiation and inhibits cell proliferation, exactly opposite to the effects of DADTII. At least part of the dermorphin-induced stimulation of choline incorporation involves µ-receptors because this increased choline incorporation was completely reversed by the µ-receptor blocking agent naloxone, µ-receptor mRNAs were expressed in late-gestation fetal lung, and µ-receptor protein was present on type II cells.

We made several unexpected observations in the experiments using opiate receptor inhibitors:

(1) Our observation of increased thymidine incorporation in lung explants treated with DADTII appeared to be dependent on µ-receptor signaling because DADTII-mediated inhibition of thymidine incorporation was completely blocked by naloxone (Figure 3A) (although naloxone alone stimulated type II cell differentiation). Inhibition of DADTII-mediated growth stimulation by naloxone is likely to be due to the relatively broad opiate receptor specificity of this agent (23). It is notable that mice lacking µ-receptors have partially reduced activity of the -receptor, whereas -receptor signaling is unaffected (24).

(2) Reversal of the effect of DADTII on choline incorporation by the -blocker, binaltorphimine could reflect an additive effect of these two agents. Although binaltorphimine-stimulated choline incorporation was not statistically significant due to a wide standard error, this -blocker does cancel the DADTII effect on choline incorporation.

(3) Stimulation of DADTII-induced thymidine incorporation by 100 nM naltrindole could be a direct effect, because 10 nM naltrindole alone did stimulate thymidine incorporation, consistent with its function as an agonist, possibly at a novel receptor present in fetal lung.

(4) Both dermorphin and naloxone stimulated choline incorporation (Figure 3B). This could be due to agonist function of naloxone: naloxone has been shown to act as a partial agonist at both µ- and -receptors (23).

(5) The inhibitory effects of binaltorphimine and naltrindole on dermorphin-induced choline incorporation, and the reversal of dermorphin's inhibition of cell proliferation by binaltorphimine, are more difficult to understand. These unexpected observations, highly reproducible, could reflect the presence of receptor variants, novel opioid receptors in fetal lung, and/or immature opioid receptors:

(a) Receptor variants could occur via alternative splicing of opiate receptors (25, 26) through post-translational modification of receptors in fetal versus adult lung (such as glycosylation or proteolysis), or through alternative mechanisms of signal transduction such as receptor transmodulation (24) or modulatory effects of cytokines, arachidonic acid metabolites, or other factors present at high levels in fetal tissues (27).

(b) Opiate receptor signaling is a complex process involving multiple ligands and receptors (27, 28). Novel orphan opiates and opiate receptors have been identified (29), some of which could potentially mediate DADTII- or dermorphin-induced effects in fetal mouse lung, especially in view of the observation of nonclassical opioid receptors on small-cell lung carcinoma cell lines (7).

(c) Immature opiate receptors could be expressed in fetal lung. Although we demonstrate mRNAs and protein immunoreactivity for all three classes of opiate receptors, it is not known whether these receptors are completely processed, expressed on the cell surface, and capable of signal transduction. For instance, receptor internalization is required for opioid stimulation of mitogen-activated protein kinase (30). Responses of immature cells can differ from those of the same cell type after maturation, as shown for a µ-receptor agonist in cultured oligodendrocytes (31). Such differences could be related to receptor modulation by factors differentially expressed in fetal lung, such as cytokines, which can directly bind to opiate receptors and alter their function (32).

We do not know whether effects of DADTII or dermorphin are direct or mediated via the release of other factors. Morphine administration can induce pulmonary vasoconstriction due to histamine release from the lung via nonconventional opioid receptors located on mast cells (33). Interestingly, in the present study all three opiate receptor subtypes were expressed both on type II cells and in the alveolar mesenchyme (Figure 6). These observations suggest that opioid peptides may have both direct effects on type II cell differentiation and indirect effects mediated via putative mesenchymal-derived mediators (5). In contrast, immunoreactivities only for - and µ-opiate receptors occurred on airway epithelial cells.

Our observations of opposite responses elicited by deltorphins and dermorphins are reminiscent of results from other laboratories comparing multiple opiate receptors. In one study, intracerebrally infused DADTII evoked motor stimulation, whereas a µ-receptor agonist, PL017, elicited motor inhibition (34). Elsewhere, PL017 increased growth of oligodendrocytes in culture, whereas the -blocker binaltorphimine increased myelin production but had no effect on growth responses (31). When levels of µ- and - receptors were reduced using antisense oligodeoxynucleotides in mice, animals with decreased -receptors had blunted responses to DADTII but not to morphine. Also, reduced morphine withdrawal signs precipitated by naloxone were present in morphine-dependent mice, suggesting functional interdependence of µ- and -receptors (35).

It was previously found that exogenous opioids can inhibit cell proliferation and induce type II cell differentiation in developing lung in rabbits in vivo (13, 14) or in vitro (36). However, the doses of heroin, morphine, and/or methadone required to elicit responses in those studies of rabbits and rats were supraphysiologic, in the micromolar range, which could stimulate multiple opiate receptor subtypes. In contrast, in the present study we observed significant effects in the nanomolar/subnanomolar dose range. The low doses of DADTII eliciting biologic responses further support the concept that endogenous deltorphins are likely to play a role in modulating fetal lung development in vivo.

In summary, our observations of opposing effects of DADTII and dermorphin in fetal lung suggest that endogenous opioid-like peptides might play important roles in vivo, regulating cell proliferation and differentiation, with the net outcome dependent on relative ratios of these and other signaling molecules.