Hormonal Regulation of Intercellular Communication: Parathyroid Hormone Increases Connexin 43 Gene Expression and Gap-Junctional Communication in Osteoblastic Cells
Paul C. Schiller, Guy A. Howard
Parmender
P. Mehta,
Bernard
A. Roos, and
Geriatric Research, Education, and Clinical Center Veterans Affairs Medical Center (P.C.S., B.A.R., G.A.H.) and Departments of Medicine (P.C.S., B.A.R., G.A.H.) Biochemistry and Molecular Biology (P.C.S., G.AJ -1.) Physiology and Biophysics (P.P.M.) and Neurology (B.A.R.) University of Miami School of Medicine Miami, Florida 33101
The presence of gap junctions between osteoblastic cells has been previously reported. For this study we used the rat osteosarcoma cell line UMR 106, which expresses the osteoblastic phenotype, as a model to characterize further the nature, physiology, and regulation of gap junctions. Northern blot analysis identified a 3.0-kilobase RNA species corresponding to the gap junction protein connexin 43. The presence of two other connexin RNA species (26 and 32) could not be detected by this method in these cells. The identified connexin RNA was amplified by reverse transcription coupled to polymerase chain reaction; the sequence of the amplified product appears identical to the sequence of a cloned rat heart connexin 43 gene. After treatment with PTH, forskolin, and 6-Br-CAMP (a CAMP analog), the levels of connexin 43 RNA in UMR 106 cells increased. Further evidence for the role of PTH and CAMP in the physiology of gap junctions in these cells was obtained with Lucifer yellow dye transfer experiments. Gap-junctional intercellular communication increased in response to PTH and forskolin (an inducer of adenylate cyclase activity). Expression of connexin 43 RNA increased severalfold in response to PTH in a concentrationand time-dependent fashion. Connexin 43 RNA and its PTH-mediated stimulation were also observed in several other osteoblastic cell lines. The roles of PTH and forskolin in regulating the physiological state of gap junctions were confirmed in primary cultures of rat calvaria osteoblasts. (Molecular Endocrinology 6: 14331440, 1992) om-&309/92/1433-1440$03.00/0 Molecular Endocrinology CopyrIght D 1992 by The Endocme
INTRODUCTION Bone tissue consists of a mineralized matrix and a heterogeneous population of cells which participate in bone development and remodeling. Four major cell types line and penetrate bone. Osteoblasts (bone-forming cells), osteoclasts (bone-resorbing cells), and lining cells are present on the surface of the bone, while osteocytes permeate the interior of the bone. Even though systemic hormones influence the overall processes of bone remodeling and calcium homeostasis, it is becoming more likely that these cells produce paracrine and autocrine factors that regulate these events (see Ref. 1 for review). In this context, osteoblasts appear to play a pivotal role, being required for both bone formation and resorption (2). Mature osteoclasts apparently lack receptors for bone-resorbing factors such as PTH and 1,25dihydroxyvitamin D3 (2). Accumulating evidence suggests that osteoblasts regulate bone resorption by stimulating PTH-dependent osteoclast activity (2, 3) and by mediating the proliferation and maturation of precursor cells into osteoclasts (4). In addition, mature osteoblasts seem to be necessary for the PTH-mediated stimulation of osteoprogenitor cell proliferation and bone formation (5). Several models have been proposed for studying the origin and development of osteoclasts. It is generally accepted that osteoclasts are formed from hemopoietic stem cells, with the participation of different cytokines, 1,2!5dihydroxyvitamin Ds, and PTH. In addition, some models require not just the presence of osteoblastic cells in this process, but direct cell-to-cell contact between osteoclast progenitors and osteoblasts for osteoclast development to occur (6). Interestingly, in the context of osteoblast development, osteoblasts have
Society
1433 The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 21 October 2015. at 05:43 For personal use only. No other uses without permission. . All rights reserved.
MOL 1434
ENDO.
Vol6
been reported to mediate the PTH-stimulated proliferation of osteoprogenitor cells, where a direct cell-to-cell contact is a prerequisite (5). Morphological and structural analyses in bone tissue of the cellular processes that connect osteocytes with each other and with osteoblasts have demonstrated the presence of gap junctions between these cells (7). These observations support an earlier finding that dyes injected into a single osteoblast on an intact rat calvarium are transmitted to numerous surrounding cells, thus suggesting that gap junctions are involved in intercellular communication in bone (8). Involvement of junctional cell-to-cell communication in bone-cell physiology is supported by studies with the protooncogene c-src, a membrane-bound protein tyrosine kinase that regulates gap-junctional communication (9). In experiments with chimeric mice, targeted disruption of the c-src protooncogene in these animals causes a deficiency in bone remodeling leading to osteopetrosis (10). Gap junctions are aggregations of individual cell-tocell channels that allow the exchange of small molecules between the cytoplasmic interiors of adjacent cells (11). Each channel is formed by a pair of connexons (one per cell) tightly connected through the intercellular space. Each connexon comprises six symmetrically associated individual proteins termed connexins (12). Several different connexins have been characterized, generally identified according to their derived molecular mass. Only ions and small polar molecules including water, amino acids, cyclic nucleotides, sugars, and small peptides of molecular mass up to 1000-2000 Daltons can pass through gap junctions (13-15). Increasing evidence suggests that gap junctions are involved in the control of cell growth, tissue development, and cellular differentiation by allowing cell-to-cell transfer of signal molecules and metabolites (16, 17). A number of factors are reported to regulate channels and their gating (opening or closing), including intracytoplasmic signals such as Ca”+ (18, 19), pH (20-22) CAMP (23, 24) and protein phosphorylation (25). Previously reported hormonal (TSH) stimulation of gap junction-mediated cell-to-cell communication in correlation with intracellular CAMP levels (26) has recently been described in thyroid cells (27). In the present work we have used the well characterized rat osteosarcoma cell lines, which express osteoblastic phenotypes, and normal rat calvarial osteoblasts to identify which of the best characterized connexin genes are expressed in these cells. Since PTH regulates osteoblasts and works at least in part via increased CAMP, we have also sought evidence for PTH up-regulation of gap-junctional communication (23, 24).
RESULTS Expression of Connexin Osteosarcoma Cells
43 (Cx43) in UMR 106 Rat
Northern blot analysis was used to identify the type(s) of connexin genes expressed in UMR 106 osteosar-
No. 9
coma cells (see Materials and Methods). Total RNA was prepared from confluent cells, separated in an agaroseformaldehyde gel, and transferred to a nylon membrane. The blotted RNA was probed with 32P-labeled DNA fragments corresponding to Cx26 (28) Cx32 (29) and Cx43 (30). A 3.0-kilobase mRNA species was detected when the membrane was hybridized to the Cx43 probe (Fig. 1A). No signal was detected with either the Cx26 or the Cx32 probe (not shown). The Common Identity of Bone Connexin the Cloned Rat Heart Cx43 Gene
mRNA and
In order to characterize further the nature of the Cx43 mRNA detected by Northern blot, we obtained partial nucleotide sequence information from a region of the gene that, from sequence comparison of cloned connexin genes, appears to be variable among connexins. The total RNA isolated from UMR 106 cells was the starting material for reverse transcription, with oligodT2, as primer. The product of the reverse transcription reaction was amplified by the polymerase chain reaction (PCR) method. Since we were uncertain of the nature of the Cx43 identified, the selection of primers for PCR was based on regions in Cx43 that appear conserved, based on amino acid sequence comparisons, with other cloned connexin genes. The selected regions correspond to the first and second extracellular domains (El and E2), which are highly conserved among Cx26, Cx32, Cx38 (31) and Cx43. The nucleotide sequence of the primers corresponds to the DNA sequence of Cx43 selected from the amino acid sequence comparison (Fig. 1 B). The amplified product would expand the regions corresponding to the second transmembrane (TM2), second cytoplasmic (C2), and third transmembrane (TM3) domains, where C2 diverges considerably among the connexin genes compared. The size of the amplified product for Cx43 would be 461 base pairs, which approximately corresponds to the size of the fragment obtained (Fig. 1C). The product of the amplification reaction was cloned into the vector pGEM5Zf(+) (Promega Biotech, Madison, WI); sequencing of the inserted DNA fragment showed it to be identical to its corresponding region in the cloned rat heart Cx43 gene. PTH and CAMP Stimulation of Cx43 mRNA Expression in UMR 106 Cells In bone and kidney cells PTH mediates its action through a GTP-binding protein-coupled receptor, thereby stimulating adenylate cyclase and phospholipase C (2). As mentioned above, CAMP is one of the intracytoplasmic signaling systems involved in gapjunctional permeability. To investigate the effect of PTH on Cx43 at the level of gene expression, UMR 106 cells were grown to confluence and treated with rat PTH (l34, 5 nM) for different periods of time. Total RNA was extracted from untreated and treated cells, and the stimulation of Cx43 gene expression was assessed by
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 21 October 2015. at 05:43 For personal use only. No other uses without permission. . All rights reserved.
PTH-Mediated
Expression
of Cx43
1435
A: Detection 0
of Cx43 RNA
lhr
B: Priers
2hr
C: PCR-Amplifiition
for PCR
ptict
El domains
Pmer:TCT MC
ACT
‘X4 CL4 CCT ‘XX
E2 domains Cx26: cJ2: Cx38: cx43:
-GAPDH
SER s* sm SER
AR0 AR0 AR0 Alto
PRO THR PRO TNR PRO MEr PRO THR
OL” OL” OL” OL”
LYS THR LYS 3-m LYS TtrR l.YS THR
461
nuc-
m43: TCA COT ccc AC0 0.40 AM ACC Pmer:ACT WA Ccc K% CTC T2-T KC
1. Characterization of the Connexin RNA Expressed in UMR 106 Ceils A, Northern blot analysis of total RNA from untreated UMR 106 cells and from cells treated with 5 nM PTH for 1 and 2 h. The membrane was probed with radioactively labeled DNA fragments corresponding to Cx43 and glyceraldehyde phosphate dehydrogenase. Hybridization to both probes is shown. The same membrane was stripped of radioactivity and probed again with DNA fragments for Cx26 and Cx32; no signal could be detected for the last 2 connexin probes. B, Selection of the regions of Cx43 that appear highly conserved based on amino acid sequence comparisons. These regions correspond to the first (El) and second (E2) Cx43 extracellular domains. Primers for PCR-mediated amplification of reverse-transcribed total RNA from UMR 106 cells were designed based on the nucleotide sequence (Nu43) of the rat heart Cx43 cDNA. C, Separation of the amplification products on an agarose gel and staining with ethidium bromide. Left lane, negative control; two right lanes, duplicate runs of product using RNA from UMR 106 cells as starting material.
Fig.
Northern blot analysis. The level of expression of Cx43 mRNA in confluent untreated UMR 106 cells is very low (Fig. 2) and as shown later, the coupling between cells is minimal. After PTH addition, a slight increase in Cx43 expression can be detected within 60 min; at 120 min, Cx43 mRNA expression levels increase about 5- to 1Ofold depending on the experiment (Fig. 2) and decrease thereafter. The level of expression of Cx43 mRNA was normalized to the ethidium bromide-stained 28s rRNA band observed in the nylon membrane after total RNA transfer. It is clear that PTH stimulates Cx43 mRNA expression in a time-dependent fashion in UMR 106 cells.
Primary < Cultures
88r-CAMP 120
160
>
5nM CO
PTH x min
120
Paul C. Schiller, Guy A. Howard
Parmender
P. Mehta,
Bernard
A. Roos, and
Geriatric Research, Education, and Clinical Center Veterans Affairs Medical Center (P.C.S., B.A.R., G.A.H.) and Departments of Medicine (P.C.S., B.A.R., G.A.H.) Biochemistry and Molecular Biology (P.C.S., G.AJ -1.) Physiology and Biophysics (P.P.M.) and Neurology (B.A.R.) University of Miami School of Medicine Miami, Florida 33101
The presence of gap junctions between osteoblastic cells has been previously reported. For this study we used the rat osteosarcoma cell line UMR 106, which expresses the osteoblastic phenotype, as a model to characterize further the nature, physiology, and regulation of gap junctions. Northern blot analysis identified a 3.0-kilobase RNA species corresponding to the gap junction protein connexin 43. The presence of two other connexin RNA species (26 and 32) could not be detected by this method in these cells. The identified connexin RNA was amplified by reverse transcription coupled to polymerase chain reaction; the sequence of the amplified product appears identical to the sequence of a cloned rat heart connexin 43 gene. After treatment with PTH, forskolin, and 6-Br-CAMP (a CAMP analog), the levels of connexin 43 RNA in UMR 106 cells increased. Further evidence for the role of PTH and CAMP in the physiology of gap junctions in these cells was obtained with Lucifer yellow dye transfer experiments. Gap-junctional intercellular communication increased in response to PTH and forskolin (an inducer of adenylate cyclase activity). Expression of connexin 43 RNA increased severalfold in response to PTH in a concentrationand time-dependent fashion. Connexin 43 RNA and its PTH-mediated stimulation were also observed in several other osteoblastic cell lines. The roles of PTH and forskolin in regulating the physiological state of gap junctions were confirmed in primary cultures of rat calvaria osteoblasts. (Molecular Endocrinology 6: 14331440, 1992) om-&309/92/1433-1440$03.00/0 Molecular Endocrinology CopyrIght D 1992 by The Endocme
INTRODUCTION Bone tissue consists of a mineralized matrix and a heterogeneous population of cells which participate in bone development and remodeling. Four major cell types line and penetrate bone. Osteoblasts (bone-forming cells), osteoclasts (bone-resorbing cells), and lining cells are present on the surface of the bone, while osteocytes permeate the interior of the bone. Even though systemic hormones influence the overall processes of bone remodeling and calcium homeostasis, it is becoming more likely that these cells produce paracrine and autocrine factors that regulate these events (see Ref. 1 for review). In this context, osteoblasts appear to play a pivotal role, being required for both bone formation and resorption (2). Mature osteoclasts apparently lack receptors for bone-resorbing factors such as PTH and 1,25dihydroxyvitamin D3 (2). Accumulating evidence suggests that osteoblasts regulate bone resorption by stimulating PTH-dependent osteoclast activity (2, 3) and by mediating the proliferation and maturation of precursor cells into osteoclasts (4). In addition, mature osteoblasts seem to be necessary for the PTH-mediated stimulation of osteoprogenitor cell proliferation and bone formation (5). Several models have been proposed for studying the origin and development of osteoclasts. It is generally accepted that osteoclasts are formed from hemopoietic stem cells, with the participation of different cytokines, 1,2!5dihydroxyvitamin Ds, and PTH. In addition, some models require not just the presence of osteoblastic cells in this process, but direct cell-to-cell contact between osteoclast progenitors and osteoblasts for osteoclast development to occur (6). Interestingly, in the context of osteoblast development, osteoblasts have
Society
1433 The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 21 October 2015. at 05:43 For personal use only. No other uses without permission. . All rights reserved.
MOL 1434
ENDO.
Vol6
been reported to mediate the PTH-stimulated proliferation of osteoprogenitor cells, where a direct cell-to-cell contact is a prerequisite (5). Morphological and structural analyses in bone tissue of the cellular processes that connect osteocytes with each other and with osteoblasts have demonstrated the presence of gap junctions between these cells (7). These observations support an earlier finding that dyes injected into a single osteoblast on an intact rat calvarium are transmitted to numerous surrounding cells, thus suggesting that gap junctions are involved in intercellular communication in bone (8). Involvement of junctional cell-to-cell communication in bone-cell physiology is supported by studies with the protooncogene c-src, a membrane-bound protein tyrosine kinase that regulates gap-junctional communication (9). In experiments with chimeric mice, targeted disruption of the c-src protooncogene in these animals causes a deficiency in bone remodeling leading to osteopetrosis (10). Gap junctions are aggregations of individual cell-tocell channels that allow the exchange of small molecules between the cytoplasmic interiors of adjacent cells (11). Each channel is formed by a pair of connexons (one per cell) tightly connected through the intercellular space. Each connexon comprises six symmetrically associated individual proteins termed connexins (12). Several different connexins have been characterized, generally identified according to their derived molecular mass. Only ions and small polar molecules including water, amino acids, cyclic nucleotides, sugars, and small peptides of molecular mass up to 1000-2000 Daltons can pass through gap junctions (13-15). Increasing evidence suggests that gap junctions are involved in the control of cell growth, tissue development, and cellular differentiation by allowing cell-to-cell transfer of signal molecules and metabolites (16, 17). A number of factors are reported to regulate channels and their gating (opening or closing), including intracytoplasmic signals such as Ca”+ (18, 19), pH (20-22) CAMP (23, 24) and protein phosphorylation (25). Previously reported hormonal (TSH) stimulation of gap junction-mediated cell-to-cell communication in correlation with intracellular CAMP levels (26) has recently been described in thyroid cells (27). In the present work we have used the well characterized rat osteosarcoma cell lines, which express osteoblastic phenotypes, and normal rat calvarial osteoblasts to identify which of the best characterized connexin genes are expressed in these cells. Since PTH regulates osteoblasts and works at least in part via increased CAMP, we have also sought evidence for PTH up-regulation of gap-junctional communication (23, 24).
RESULTS Expression of Connexin Osteosarcoma Cells
43 (Cx43) in UMR 106 Rat
Northern blot analysis was used to identify the type(s) of connexin genes expressed in UMR 106 osteosar-
No. 9
coma cells (see Materials and Methods). Total RNA was prepared from confluent cells, separated in an agaroseformaldehyde gel, and transferred to a nylon membrane. The blotted RNA was probed with 32P-labeled DNA fragments corresponding to Cx26 (28) Cx32 (29) and Cx43 (30). A 3.0-kilobase mRNA species was detected when the membrane was hybridized to the Cx43 probe (Fig. 1A). No signal was detected with either the Cx26 or the Cx32 probe (not shown). The Common Identity of Bone Connexin the Cloned Rat Heart Cx43 Gene
mRNA and
In order to characterize further the nature of the Cx43 mRNA detected by Northern blot, we obtained partial nucleotide sequence information from a region of the gene that, from sequence comparison of cloned connexin genes, appears to be variable among connexins. The total RNA isolated from UMR 106 cells was the starting material for reverse transcription, with oligodT2, as primer. The product of the reverse transcription reaction was amplified by the polymerase chain reaction (PCR) method. Since we were uncertain of the nature of the Cx43 identified, the selection of primers for PCR was based on regions in Cx43 that appear conserved, based on amino acid sequence comparisons, with other cloned connexin genes. The selected regions correspond to the first and second extracellular domains (El and E2), which are highly conserved among Cx26, Cx32, Cx38 (31) and Cx43. The nucleotide sequence of the primers corresponds to the DNA sequence of Cx43 selected from the amino acid sequence comparison (Fig. 1 B). The amplified product would expand the regions corresponding to the second transmembrane (TM2), second cytoplasmic (C2), and third transmembrane (TM3) domains, where C2 diverges considerably among the connexin genes compared. The size of the amplified product for Cx43 would be 461 base pairs, which approximately corresponds to the size of the fragment obtained (Fig. 1C). The product of the amplification reaction was cloned into the vector pGEM5Zf(+) (Promega Biotech, Madison, WI); sequencing of the inserted DNA fragment showed it to be identical to its corresponding region in the cloned rat heart Cx43 gene. PTH and CAMP Stimulation of Cx43 mRNA Expression in UMR 106 Cells In bone and kidney cells PTH mediates its action through a GTP-binding protein-coupled receptor, thereby stimulating adenylate cyclase and phospholipase C (2). As mentioned above, CAMP is one of the intracytoplasmic signaling systems involved in gapjunctional permeability. To investigate the effect of PTH on Cx43 at the level of gene expression, UMR 106 cells were grown to confluence and treated with rat PTH (l34, 5 nM) for different periods of time. Total RNA was extracted from untreated and treated cells, and the stimulation of Cx43 gene expression was assessed by
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 21 October 2015. at 05:43 For personal use only. No other uses without permission. . All rights reserved.
PTH-Mediated
Expression
of Cx43
1435
A: Detection 0
of Cx43 RNA
lhr
B: Priers
2hr
C: PCR-Amplifiition
for PCR
ptict
El domains
Pmer:TCT MC
ACT
‘X4 CL4 CCT ‘XX
E2 domains Cx26: cJ2: Cx38: cx43:
-GAPDH
SER s* sm SER
AR0 AR0 AR0 Alto
PRO THR PRO TNR PRO MEr PRO THR
OL” OL” OL” OL”
LYS THR LYS 3-m LYS TtrR l.YS THR
461
nuc-
m43: TCA COT ccc AC0 0.40 AM ACC Pmer:ACT WA Ccc K% CTC T2-T KC
1. Characterization of the Connexin RNA Expressed in UMR 106 Ceils A, Northern blot analysis of total RNA from untreated UMR 106 cells and from cells treated with 5 nM PTH for 1 and 2 h. The membrane was probed with radioactively labeled DNA fragments corresponding to Cx43 and glyceraldehyde phosphate dehydrogenase. Hybridization to both probes is shown. The same membrane was stripped of radioactivity and probed again with DNA fragments for Cx26 and Cx32; no signal could be detected for the last 2 connexin probes. B, Selection of the regions of Cx43 that appear highly conserved based on amino acid sequence comparisons. These regions correspond to the first (El) and second (E2) Cx43 extracellular domains. Primers for PCR-mediated amplification of reverse-transcribed total RNA from UMR 106 cells were designed based on the nucleotide sequence (Nu43) of the rat heart Cx43 cDNA. C, Separation of the amplification products on an agarose gel and staining with ethidium bromide. Left lane, negative control; two right lanes, duplicate runs of product using RNA from UMR 106 cells as starting material.
Fig.
Northern blot analysis. The level of expression of Cx43 mRNA in confluent untreated UMR 106 cells is very low (Fig. 2) and as shown later, the coupling between cells is minimal. After PTH addition, a slight increase in Cx43 expression can be detected within 60 min; at 120 min, Cx43 mRNA expression levels increase about 5- to 1Ofold depending on the experiment (Fig. 2) and decrease thereafter. The level of expression of Cx43 mRNA was normalized to the ethidium bromide-stained 28s rRNA band observed in the nylon membrane after total RNA transfer. It is clear that PTH stimulates Cx43 mRNA expression in a time-dependent fashion in UMR 106 cells.
Primary < Cultures
88r-CAMP 120
160
>
5nM CO
PTH x min
120
