The serum amyloid A (SAA) protein has been implicated in the pathogenesis of several chronic inflammatory diseases. Its induction mechanism in response to a chronic inflammatory condition was investigated in rabbits following multiple s.c. injections of AgNO3 over a period of 35 days. During unremitting exposure to inflammatory stimulus, a persistently higher than normal level of SAA2 expression was seen in multiple tissues. Induction of SAA was correlated with higher levels of several transcription factor activities. Increased SAA-activating factor (SAF) activity was detected in the liver, lung, and brain tissues under both acute and chronic inflammatory conditions. In the heart, kidney, and skeletal muscle tissues, this activity remained virtually constant. In contrast, CCAAT enhancer binding protein (C/EBP) DNA-binding activity was transiently induced in selective tissues. Higher than normal NF-κB DNA-binding activity was detected in the lung and to a lesser extent in the liver and kidney tissues under both acute and chronic conditions. This result suggested that C/EBP, SAF, and NF-κB are required for transient acute phase induction of SAA whereas SAF and NF-κB activities are necessary for persistent SAA expression during chronic inflammatory conditions.

During inflammation, serum amyloid A (SAA)3 proteins, a member of acute phase protein family, are found in the circulation associated with high-density lipoprotein (1). Under normal conditions, a trace amount of SAA is present, but in response to various injuries including trauma or infection, its level is increased 100- to 1000-fold (2, 3). Such an increase of SAA concentration is primarily caused by a transcriptional induction event mediated by several cytokines, including IL-1, IL-6, and TNF-α, released during inflammation (3, 4). SAA is induced transiently and rapidly returns to normal low basal level within 72–96 h of the initial inflammatory stimulus. Strong phylogenic conservation of SAA together with its dramatic up-regulation during altered physiological conditions suggest that this protein may have beneficial role in maintenance of cellular homeostasis. Although exact functions of SAA are still unclear, its association with high-density lipoprotein suggests a role in lipid metabolism (5) or lipid transport during the host response to injury. SAA may have some role in the suppression of immune responses (6), inhibition of platelet aggregation (7), and neutrophil oxidative burst (8). The SAA superfamily is composed of a number of genes and proteins. The human SAA gene family is comprised of the highly homologous SAA1 and SAA2, and a less related SAA4 and a nonexpressing SAA3 gene (9, 10, 11, 12). In mice, among the four SAA isoforms, SAA1 and SAA2 are 95% homologous, whereas SAA3 and SAA4 are distinct (12, 13). In other species, the nomenclature of the members of this family has been the result of comparisons with previously reported gene sequences in mice and humans. In the rabbit, six cDNA clones have been reported (14, 15, 16, 17). The sequence of these clones are slightly different from each other, probably due to allelic variations. It is unclear whether all of these isoforms are involved in amyloid formation. To date, only one rabbit cDNA clone (16), designated as SAA2, matches exactly with the published amino acid sequences of rabbit protein AA (18) isolated from amyloid fibrils. The upstream promoter region of this isoform is isolated (19) and used in the present investigation.

Although transient induction of SAA is not harmful, prolonged expression of SAA under chronic inflammatory conditions results in amyloid deposition in humans (20) and mice (21). A higher than normal serum level of SAA is seen in patients with rheumatoid arthritis (22, 23, 24). SAA can induce collagenase, a key enzyme involved in tissue destruction that occurs in inflammatory and proliferative rheumatoid arthritis and osteoarthritis (17, 25). Recent studies suggest a role of SAA in the development of atherosclerosis (26). SAA is also highly expressed in mice fed atherogenic diet (27) and in monocyte cells exposed to minimally modified low density lipoprotein, a major risk factor of atherosclerosis (28). Owing to the pathophysiological effects of abnormal SAA expression, efforts have been directed toward understanding the transcriptional induction process that leads to SAA synthesis. Induction level of SAA is variable and highly dependent upon the nature of inflammatory stimulus. To date, at least two distinctly different pathways of SAA induction have been reported (29). For example, inflammation induced by turpentine injection in rabbit predominantly activate CCAAT enhancer binding protein (C/EBP) family members (30), whereas LPS injection results in the activation of both NF-κB and C/EBP (31, 32). In the latter condition, a combined action of NF-κB and C/EBP family members was found to be responsible for the SAA induction (32). Studies on mouse, rat, and human SAA gene expression also indicated the involvement of NF-κB and C/EBP (33, 34, 35, 36). More recently, a novel cytokine-inducible transcription factor designated as SAA-activating factor (SAF) was shown to be involved in SAA gene expression (37, 38, 39, 40). However, all of these studies have focused on the mechanism of transient up-regulation of SAA associated with acute inflammatory conditions. As it is becoming increasingly clear that persistent expression of SAA during chronic inflammation is responsible for all SAA-linked pathogenesis, we aimed this study toward understanding the mechanism of induction of SAA in multiple tissues under both chronic and acute inflammatory conditions. We show that C/EBP, SAF, and NF-κB are required for transient acute phase induction of SAA, whereas SAF and NF-κB activities are necessary for persistent SAA expression during chronic inflammatory conditions.

Twelve New Zealand White rabbits were divided into four groups. One group was injected s.c. once with 2% (w/v) AgNO3 solution (1 ml/kg body weight). The second group received the same amount of s.c. AgNO3 injection three times a week for 18 days (a total of 9 injections), and the third group received the same regimen for 35 days (a total of 16 injections). The fourth group received sterile water injections and was used as a control. The animals were euthanized by CO2 asphyxiation at the end of treatments. Tissues were collected and stored in liquid nitrogen until further use.

Total RNA was extracted from multiple tissues by guanidinium thiocyanate lysis method (41). Poly(A)+ RNA was isolated from the total RNA by using oligo(dT)-Sepharose. Single-stranded complimentary antisense RNase protection probe was produced by transcription with T7 RNA polymerase of the linearized pTZ19U plasmid template containing rabbit SAA2 cDNA sequences from 180 to 280 nt (16). In the RPA, this probe generates a 100-base protected fragment only from the rabbit SAA2 transcript. Other SAA isoforms of rabbit have several mismatch sequence in the region between nt 180 and 280 that would result in fragmentation of the 100-base-long probe in the RPA. This probe is thus highly specific for rabbit SAA2. The RPAs were performed with the RPAII kit (Ambion, Austin, TX) following manufacturer’s protocol. One microgram of poly(A)+ RNA was used for each sample. To evaluate the quality and quantity of each RNA sample, β-actin riboprobe cRNA was used as an internal control. Protected RNA fragments were electrophoretically separated in a 7% polyacrylamide-8 M urea gel and visualized by autoradiography.

Nuclear extracts were prepared from control and AgNO3-induced rabbit tissues essentially following the method described earlier (30, 37). Protein concentrations were measured by the method of Bradford (42). Binding assays for the DNA-protein interaction was performed following a standard protocol described earlier (37) with different 32P-labeled dsDNA probes as described. The DNA-protein complexes were detected by mobility shift assay in a 6% native polyacrylamide gel. The labeling of DNA was performed by filling in the overhangs at the termini with Klenow fragment of DNA polymerase and incorporating [α-32P]dATP or [α-32P]dCTP. Some DNA-protein binding assay mixture contained antisera to C/EBP, Sp1, or SAF proteins. Antisera to C/EBP-α, C/EBP-β, C/EBP-δ, and Sp1 were obtained from Santa Cruz Biotechnology (Santa Cruz, CA). Anti-SAF Ab was prepared as described (38). Nuclear extracts (10 μg protein) were preincubated with the 1 μl of antisera per 10 μl of binding assay mixture for 30 min on ice and subsequently used in the DNA-binding assays.

The DNAs used as probes for the binding assays consisted of the sequences described below and derived from rabbit SAA2 promoter region (30, 31, 39). For annealing, equal amounts of complementary strands of these oligonucleotides were heated at 95°C for 2 min in 50 mM Tris (pH 7.4), 60 mM NaCl, and 1 mM EDTA and allowed to cool slowly to room temperature in 2–3 h. The oligonucleotides are as follows: NF-κB, 5′-−112CCTAGGGGAAATGACCTGAGGGGCTTTCCAGGCA−79-3′ (31); SAF, 5′-−254CCCTTCCTCTCCACCCACAGCCCCCATGG−226-3′ (39); and C/EBP, 5′-−193GGCCTCCACAGGTTGCACAACTGGGGACGG GATCTGCGGATGAAGAAACCATGCATGT−136-3′ (30).

The C/EBP DNA-binding element was isolated by restriction enzyme cleavage of SAA promoter DNA which contained the above sequences.

SAA expression pattern under chronic inflammatory conditions was assessed by RPA using poly(A)+ RNA. A group of rabbits was s.c. injected with AgNO3 for a period up to 35 days. Induction of SAA2 during continuous exposure to AgNO3-mediated inflammation was measured in different tissues of the inflamed rabbits. As seen in Fig. 1,A, SAA expression was highly induced at 24 h following a single AgNO3 stimulation. The level of induction was most significant in the liver. SAA was also induced in the kidney and lung tissues, albeit at a much lower level. Brain and muscle tissues exhibited detectable level of induction, whereas rather insignificant level of SAA transcript was present in the heart tissue. During chronic inflammatory condition in rabbits injected with AgNO3 for 18 and 35 days, tissue-specific SAA expression remained noticeably higher compared with that in the untreated normal rabbit tissues. However, the expression level was significantly lower than that seen in 24-h samples. A persistently higher than normal level was quite evident in the liver, kidney, lung, brain, and skeletal muscle tissues. Fig. 1 B summarizes the results of quantitative densitometric measurements of protected band intensities from RPA analyses of three independent experiments. The data confirmed that SAA expression during chronic inflammatory condition is persistent in multiple tissues at a significantly higher level than that was seen in control untreated animals.

FIGURE 1.

A. SAA mRNA level in different tissues of control and chronically inflamed rabbits. RPA was performed with 1 μg of poly(A)+ RNA isolated from multiple tissues of control and AgNO3-treated rabbits, as indicated. Lanes 1, 5, 9, 13, 17, and 21 contain RNA isolated from control, untreated rabbit; lanes 2, 6, 10, 14, 18, and 22 contain RNA isolated from AgNO3/1-day treated rabbit; lanes 3, 7, 11, 15, 19, and 23 contain RNA isolated from AgNO3/18-day treated rabbit; lanes 4, 8, 12, 16, 20, and 24 contain RNA isolated from AgNO3/35-day treated rabbit. Lanes P, indicated untreated probe. RNase protected bands were fractionated by polyacrylamide gel electrophoresis and visualized by autoradiography. Protected bands were somewhat smaller than untreated probe due to the removal of extra polylinker sequences generated during T7 RNA polymerase driven in vitro transcription. B, Quantification of SAA mRNA level during various inflammatory conditions. RNase protected SAA mRNA fragments, shown in A, were analyzed by measuring radioactivity in the bands. Relative mRNA level was determined by comparing the radioactivity with that of control tissues. Data presented here represents the mean values of three independent experiments.

FIGURE 1.

A. SAA mRNA level in different tissues of control and chronically inflamed rabbits. RPA was performed with 1 μg of poly(A)+ RNA isolated from multiple tissues of control and AgNO3-treated rabbits, as indicated. Lanes 1, 5, 9, 13, 17, and 21 contain RNA isolated from control, untreated rabbit; lanes 2, 6, 10, 14, 18, and 22 contain RNA isolated from AgNO3/1-day treated rabbit; lanes 3, 7, 11, 15, 19, and 23 contain RNA isolated from AgNO3/18-day treated rabbit; lanes 4, 8, 12, 16, 20, and 24 contain RNA isolated from AgNO3/35-day treated rabbit. Lanes P, indicated untreated probe. RNase protected bands were fractionated by polyacrylamide gel electrophoresis and visualized by autoradiography. Protected bands were somewhat smaller than untreated probe due to the removal of extra polylinker sequences generated during T7 RNA polymerase driven in vitro transcription. B, Quantification of SAA mRNA level during various inflammatory conditions. RNase protected SAA mRNA fragments, shown in A, were analyzed by measuring radioactivity in the bands. Relative mRNA level was determined by comparing the radioactivity with that of control tissues. Data presented here represents the mean values of three independent experiments.

Close modal

Recent studies on the acute phase induction of SAA, which occurs within 24 h of exposure to inflammatory agent, have indicated that transcriptional induction of rabbit SAA2 gene is primarily regulated by three positive regulatory elements and their cognate binding factors, C/EBP, NF-κB, and SAF. In the rabbit SAA2 promoter, NF-κB DNA-binding element is present between −96 and −83 bp, two adjacent C/EBP DNA-binding elements are present between −191 and −134 bp, and a SAF DNA-binding element is present between −254 and −226 bp. To evaluate the mechanism of transcriptional induction during chronic inflammatory condition, we have investigated whether these three transcription factors are activated. Promoter binding activity was determined by EMSAs with specific DNA-binding probes.

Nuclear extracts were prepared from normal and chronically inflamed rabbit tissues and incubated with the radiolabeled SAF DNA-binding element (−254/−226) of SAA promoter. Results presented in Fig. 2 showed the presence of two very weak DNA-protein complexes formed with untreated rabbit liver nuclear extract (lane 1). Upon incubation with the same protein amount of nuclear extracts from the liver tissue of AgNO3-injected rabbits, a steep increase in the levels of these two complexes were seen (complexes c and d, lanes 2–4). These results indicated induction of SAF like DNA-binding activity in the liver within 24 h of AgNO3 injection. This activity remained at a persistently high level during chronic inflammatory condition. Nuclear extract prepared from untreated lung tissue formed two DNA-protein complexes (complexes d and e, lane 5). However, this pattern was dramatically changed when AgNO3-stimulated lung nuclear extracts were used (lanes 6–8). Two additional slower mobility complexes (complexes a and b) were detected. These complexes were present throughout the chronic inflammatory condition, although their levels declined considerably. Normal brain tissue also exhibited a constitutive SAF-like DNA-binding activity (complexes c and d, lane 17). A slower mobility complex was detected predominantly in the AgNO3/18-day sample (complex a, lane 19), which suggested a delayed induction of SAF in these tissues. Nuclear extracts prepared from untreated heart, kidney, and skeletal muscle tissues formed one major complex (complex d, lanes 9–12, 13–16, and 21–24, respectively). There was virtually no change in their levels following AgNO3 injection (lanes 9–16 and 21–24). We used anti-SAF Ab to characterize these DNA-protein complexes (Fig. 3). Inhibition of both constitutive and induced DNA-protein complexes by anti-SAF Ab indicated the involvement of SAF in their formation. Only one of these complexes (complex a in the lung and brain nuclear extracts, described in Fig. 2) was partly inhibited by both Sp1 and SAF Abs (Fig. 3, B and E, lanes 5 and 6), suggesting that this complex is formed by the interaction of these two transcription factors. Similar phenomenon of combinatorial action of SAF and Sp1 has been observed in recent studies (39). The current study provides further evidence of their participation in SAA promoter activation under a pathophysiological condition.

FIGURE 2.

Induction of SAF activity in different tissues of control, acute, and chronic inflamed rabbits. SAF activity was determined by incubating the nuclear extracts with 32P-labeled SAA promoter DNA fragment containing SAF-binding element (−254 to −226) described in Materials and Methods. DNA-protein complexes were resolved in a 6% nondenaturing polyacrylamide gel and visualized by autoradiography. Lanes 1, 5, 9, 13, 17, and 21 contain nuclear extract isolated from different tissues of control, untreated rabbit; lanes 2, 6, 10, 14, 18, and 22 contain nuclear extract isolated from different tissues of AgNO3/1-day treated rabbit; lanes 3, 7, 11, 15, 19, and 23 contain nuclear extract isolated from different tissues of AgNO3/18-day treated rabbit; lanes 4, 8, 12, 16, 20, and 24 contain nuclear extract isolated from different tissues of AgNO3/35-day treated rabbit.

FIGURE 2.

Induction of SAF activity in different tissues of control, acute, and chronic inflamed rabbits. SAF activity was determined by incubating the nuclear extracts with 32P-labeled SAA promoter DNA fragment containing SAF-binding element (−254 to −226) described in Materials and Methods. DNA-protein complexes were resolved in a 6% nondenaturing polyacrylamide gel and visualized by autoradiography. Lanes 1, 5, 9, 13, 17, and 21 contain nuclear extract isolated from different tissues of control, untreated rabbit; lanes 2, 6, 10, 14, 18, and 22 contain nuclear extract isolated from different tissues of AgNO3/1-day treated rabbit; lanes 3, 7, 11, 15, 19, and 23 contain nuclear extract isolated from different tissues of AgNO3/18-day treated rabbit; lanes 4, 8, 12, 16, 20, and 24 contain nuclear extract isolated from different tissues of AgNO3/35-day treated rabbit.

Close modal
FIGURE 3.

Characterization of SAF activity in different tissues of control, acute and chronic inflamed rabbits. SAF activity was determined following the method described in Fig. 2. To characterize the DNA-protein complexes, nuclear extracts were incubated with anti-SAF and anti-Sp1 Abs before their interaction with the radiolabeled probe. Because of the presence of a Sp1-binding site embedded within the SAF-binding site and because of the ability of SAF to form heteromer with Sp1 (39 ), anti-Sp1 Ab was used to identify any potential heteromeric complexes. Nuclear extracts from different tissues were used in different panels: A, liver; B, lung; C, heart; D, kidney; E, brain; and F, skeletal muscle. In each panel, lanes 1–3 contain nuclear extract from control rabbit tissue; lanes 4–6 contain nuclear extract from AgNO3/1-day treated rabbit tissue; lanes 7–9 contain nuclear extract from AgNO3/18-day treated rabbit tissue; and lanes 10–12 contain nuclear extract from AgNO3/35-day treated rabbit tissue. In each panel, −, SAF, and Sp1 indicate preincubation of nuclear extracts with no, anti-SAF, or anti-Sp1 Abs respectively. The arrow indicates the position of supershifted DNA-protein complex.

FIGURE 3.

Characterization of SAF activity in different tissues of control, acute and chronic inflamed rabbits. SAF activity was determined following the method described in Fig. 2. To characterize the DNA-protein complexes, nuclear extracts were incubated with anti-SAF and anti-Sp1 Abs before their interaction with the radiolabeled probe. Because of the presence of a Sp1-binding site embedded within the SAF-binding site and because of the ability of SAF to form heteromer with Sp1 (39 ), anti-Sp1 Ab was used to identify any potential heteromeric complexes. Nuclear extracts from different tissues were used in different panels: A, liver; B, lung; C, heart; D, kidney; E, brain; and F, skeletal muscle. In each panel, lanes 1–3 contain nuclear extract from control rabbit tissue; lanes 4–6 contain nuclear extract from AgNO3/1-day treated rabbit tissue; lanes 7–9 contain nuclear extract from AgNO3/18-day treated rabbit tissue; and lanes 10–12 contain nuclear extract from AgNO3/35-day treated rabbit tissue. In each panel, −, SAF, and Sp1 indicate preincubation of nuclear extracts with no, anti-SAF, or anti-Sp1 Abs respectively. The arrow indicates the position of supershifted DNA-protein complex.

Close modal

The role of C/EBP on SAA gene expression under chronic condition was examined by gel mobility shift assay. Nuclear extracts prepared from normal and chronically inflamed rabbit tissues were incubated with the radiolabeled C/EBP DNA-binding element of SAA2 promoter. Results presented in Fig. 4 showed that 24 h after AgNO3 injection, all tissues except heart contain an increased level of C/EBP activity compared with that of untreated normal tissues (compare lanes 1, 5, 13, 17, and 21 with lanes 2, 6, 14, 18, and 22). During chronic inflammatory condition at days 18 and 35, C/EBP DNA-binding activity declined considerably in the lung (lanes 5–8), kidney (lanes 13–16), and skeletal muscle (lanes 21–24) tissues but remained at a low but detectable level in the liver and brain. It is noteworthy that the lung tissue appears to contain a considerably higher level of C/EBP DNA-binding activity than any other tissues. Further characterization of these DNA-protein complexes were performed using Ab specific to α, β, and δ isoforms of C/EBP, and the data obtained from the untreated control, and 1-day AgNO3-treated rabbit tissue nuclear extracts are shown in Fig. 5. Both 18-day and 35-day AgNO3-treated rabbit tissue nuclear extracts exhibited complexes similar to those in Fig. 5 and therefore, for brevity, are not shown. Because the heart tissue exhibited no detectable C/EBP activity, characterization studies were performed with nuclear extracts from liver (Fig. 5,A), lung (Fig. 5,B), kidney (Fig. 5,C), brain (Fig. 5,D), and skeletal muscle (Fig. 5,E) tissues only. Ablation of DNA-protein complex of normal rabbit liver nuclear extract by anti-C/EBP-α (Fig. 5,A, lane 2) indicated the presence of mostly C/EBP-α in normal rabbit liver tissue. DNA-protein complex formed by nuclear extract of 24-h AgNO3-treated rabbit liver (Fig. 5,A, lane 5) was inhibited by both anti-C/EBP-α and anti-C/EBP-β Abs (Fig. 5,A, lanes 6 and 7) but not by anti-C/EBP-δ. This finding indicated that C/EBP-α together with C/EBP-β plays a significant role in mediating SAA expression via the C/EBP element. Interestingly, in sharp contrast to the present finding, C/EBP-δ is immensely activated in rabbit liver by turpentine, a potent inflammatory agent (30). DNA-protein complexes formed by the lung nuclear extract of normal rabbit (Fig, 5B, lane 1) were inhibited by anti-C/EBP-α and supershifted by anti-C/EBP-β Ab (Fig. 5,B, lanes 2 and 3), indicating the presence of C/EBP-α and -β in normal rabbit lung. The level of these activities is increased in 1-day AgNO3-activated lung (Fig. 5,B, lanes 6 and 7). In addition, appearance of C/EBP-δ activity is also quite evident (Fig. 5,B, lane 8). Taken together, these results suggest that the induction of C/EBP-β and C/EBP-δ in lung is transient, and these proteins are therefore involved primarily in the acute inflammatory response. In the kidney, higher level C/EBP-specific DNA-binding activity was noted primarily in the 1-day AgNO3-activated rabbits (Fig. 4, lanes 13–16). Further characterization using anti C/EBP Abs revealed a slight but nonspecific inhibition of the DNA-protein complexes (Fig. 5,C, lanes 5–8). In the brain also, higher levels of C/EBP-specific DNA-binding activity were seen in 1-day AgNO3-activated rabbits (Fig. 4, lanes 17–20). These complexes were not inhibited by α-, β-, or δ-specific C/EBP Abs (Fig. 5,D, lanes 5–8). Taken together, these data suggest that the DNA binding activities in kidney and brain are unrelated to C/EBP-α, -β, or -δ but may be due to other, yet unidentified members of this family of transcription factors. Further studies along this line is necessary to identify these apparent C/EBP-related proteins. Similarly, skeletal muscle tissue exhibited high but transient induction (1-day AgNO3-treated) of C/EBP activity, which was characterized as those composed of C/EBP-α and -β and to a lesser extent C/EBP-δ (Fig. 5,E, lanes 1–4). C/EBP activity in normal and chronic inflamed skeletal muscle tissue nuclear extract preparations was almost negligible (Fig. 4, lanes 21, 23, and 24) and therefore was not characterized.

FIGURE 4.

Induction of C/EBP activity in acute and chronic inflamed rabbit tissues. C/EBP activity was determined by incubating nuclear extracts with 32P-labeled SAA promoter DNA fragment containing the C/EBP-binding element (−193 to −136) described in Materials and Methods. The resulting DNA-protein complexes were resolved in a 6% nondenaturing polyacrylamide gel. Lanes 1, 5, 9, 13, 17, and 21 contain nuclear extract isolated from different tissues of control, untreated rabbit; lanes 2, 6, 10, 14, 18, and 22 contain nuclear extract isolated from different tissues of AgNO3/1-day treated rabbit; lanes 3, 7, 11, 15, 19, and 23 contain nuclear extract isolated from different tissues of AgNO3/18-day treated rabbit; lanes 4, 8, 12, 16, 20, and 24 contain nuclear extract isolated from different tissues of AgNO3/35-day treated rabbit. The vertical line represents the region of C/EBP-specific complexes and was determined by the competition of their DNA-binding ability by C/EBP-specific oligonucleotides.

FIGURE 4.

Induction of C/EBP activity in acute and chronic inflamed rabbit tissues. C/EBP activity was determined by incubating nuclear extracts with 32P-labeled SAA promoter DNA fragment containing the C/EBP-binding element (−193 to −136) described in Materials and Methods. The resulting DNA-protein complexes were resolved in a 6% nondenaturing polyacrylamide gel. Lanes 1, 5, 9, 13, 17, and 21 contain nuclear extract isolated from different tissues of control, untreated rabbit; lanes 2, 6, 10, 14, 18, and 22 contain nuclear extract isolated from different tissues of AgNO3/1-day treated rabbit; lanes 3, 7, 11, 15, 19, and 23 contain nuclear extract isolated from different tissues of AgNO3/18-day treated rabbit; lanes 4, 8, 12, 16, 20, and 24 contain nuclear extract isolated from different tissues of AgNO3/35-day treated rabbit. The vertical line represents the region of C/EBP-specific complexes and was determined by the competition of their DNA-binding ability by C/EBP-specific oligonucleotides.

Close modal
FIGURE 5.

Characterization of activated C/EBP in multiple tissues of acute and chronic inflamed rabbit. Mobility shift assay was performed as described in Fig. 4, with nuclear extracts from different tissues and shown in different panels: A, liver; B, lung; C, kidney; D, brain; and E, muscle. In A–D,lanes 1–4 contain nuclear extract from control rabbit tissue; lanes 4–8 contain nuclear extract from AgNO3/1-day treated rabbit tissues. In E, lanes 1–4 contain nuclear extract from AgNO3/1-day treated rabbit tissue. In each panel, −, α, β, and δ indicate preincubation of nuclear extracts with no, anti-C/EBP-α, anti-C/EBP-β, or anti-C/EBP-δ Abs, respectively. The arrow indicates the position of supershifted DNA-protein complex.

FIGURE 5.

Characterization of activated C/EBP in multiple tissues of acute and chronic inflamed rabbit. Mobility shift assay was performed as described in Fig. 4, with nuclear extracts from different tissues and shown in different panels: A, liver; B, lung; C, kidney; D, brain; and E, muscle. In A–D,lanes 1–4 contain nuclear extract from control rabbit tissue; lanes 4–8 contain nuclear extract from AgNO3/1-day treated rabbit tissues. In E, lanes 1–4 contain nuclear extract from AgNO3/1-day treated rabbit tissue. In each panel, −, α, β, and δ indicate preincubation of nuclear extracts with no, anti-C/EBP-α, anti-C/EBP-β, or anti-C/EBP-δ Abs, respectively. The arrow indicates the position of supershifted DNA-protein complex.

Close modal

Analysis of NF-κB DNA-binding activity showed induction of this transcription factor in the liver, lung, and kidney tissues following AgNO3 stimulation, which remained at a persistently higher level during chronic conditions (Fig. 6). No appreciable NF-κB activity was seen in the heart, brain, and muscle tissues. It should be noted that this autoradiogram was exposed for a longer period of time (5-fold higher) than the other two autoradiograms showing the DNA-binding activities of SAF and C/EBP (Figs. 2 and 4). Nonetheless, this result indicated a persistent low level activation of NF-κB under chronic inflammatory conditions.

FIGURE 6.

Induction of NF-κB activity in acute and chronic inflamed rabbit tissues. NF-κB activity was determined by incubating nuclear extracts with 32P-labeled SAA promoter DNA fragment containing the NF-κB-binding element (−112 to −79) described in Materials and Methods. The resulting DNA-protein complexes were resolved in a 6% nondenaturing polyacrylamide gel. Lanes 1, 5, 9, 13, 17, and 21 contain nuclear extract isolated from different tissues of control, untreated rabbit; lanes 2, 6, 10, 14, 18, and 22 contain nuclear extract isolated from different tissues of AgNO3/1-day treated rabbit; lanes 3, 7, 11, 15, 19, and 23 contain nuclear extract isolated from different tissues of AgNO3/18-day treated rabbit; lanes 4, 8, 12, 16, 20, and 24 contain nuclear extract isolated from different tissues of AgNO3/35-day treated rabbit. The vertical line represents the region of NF-κB-specific complexes and was determined by the competition of their DNA-binding ability by NF-κB-specific oligonucleotides.

FIGURE 6.

Induction of NF-κB activity in acute and chronic inflamed rabbit tissues. NF-κB activity was determined by incubating nuclear extracts with 32P-labeled SAA promoter DNA fragment containing the NF-κB-binding element (−112 to −79) described in Materials and Methods. The resulting DNA-protein complexes were resolved in a 6% nondenaturing polyacrylamide gel. Lanes 1, 5, 9, 13, 17, and 21 contain nuclear extract isolated from different tissues of control, untreated rabbit; lanes 2, 6, 10, 14, 18, and 22 contain nuclear extract isolated from different tissues of AgNO3/1-day treated rabbit; lanes 3, 7, 11, 15, 19, and 23 contain nuclear extract isolated from different tissues of AgNO3/18-day treated rabbit; lanes 4, 8, 12, 16, 20, and 24 contain nuclear extract isolated from different tissues of AgNO3/35-day treated rabbit. The vertical line represents the region of NF-κB-specific complexes and was determined by the competition of their DNA-binding ability by NF-κB-specific oligonucleotides.

Close modal

The NF-κB activity was also characterized by using specific Abs to p50 and p65, two major isoforms of this transcription factor. In all preparations, wherever activity was detected, both p50 and p65 were present (data not shown). Thus, a persistent inflammatory stimulus brought on by AgNO3 continuously maintain an activated status of low level of NF-κB.

The present study provides a first report on the mechanism of SAA induction under chronic inflammatory condition and determines the role of regulatory factors involved in this process. An increased level of plasma SAA is seen and linked to the pathophysiology of many chronic inflammatory diseases, but expression patterns of SAA under chronic conditions, especially in different tissues, and the mechanisms of such induction remained elusive. In this paper we have shown that chronic inflammation induces SAA in multiple tissues. It is interesting to note that chronic inflammation resulted in a tolerance to further stimulation of SAA expression in response to continued exposure to inflammatory agents and thus produced a much lower level of SAA mRNA during chronic condition. Such anomaly in the acute-phase protein response has been noted in various diseases. Studies on systemic lupus erythromatosus, mixed connective disease, ulcerative colitis, and chronic active hepatitis have shown low acute phase protein responses for the amount of inflammatory activity present (43). Although these observations are of interest, the mechanisms involved are likely to be complex. To delineate, we have examined the roles of the regulatory transcription factors, including SAF, C/EBP, and NF-κB, that are known to be involved in SAA gene induction in response to inflammatory signals.

Studies on SAF, C/EBP, and NF-κB DNA-binding activities show a steep stimulation of these three transcription factor activities within 24 h of AgNO3 stimulation which correlated very well with the steep rise in SAA mRNA expression at this time. However, under chronic conditions, these factors appeared to be differentially regulated. Relative levels of the DNA-binding activity of the three transcription factors in different tissues of rabbit under both acute and chronic inflammatory conditions are summarized in Table I. C/EBP DNA-binding activity, which was elevated sharply in multiple tissues following initial stimulation, declined sharply despite the presence of continuous inflammatory stimuli. At day 35, in most tissues a very low level of C/EBP activity was seen. In contrast, activities of both SAF and NF-κB declined at a lower rate. At day 35, considerably high levels of SAF and NF-κB DNA-binding activities were present. These results implied that SAF and NF-κB are major regulators of SAA expression under chronic condition whereas C/EBP, SAF, and NF-κB are involved in the transient induction of SAA under acute condition. It should be noted that all tissues did not induce or express C/EBP, SAF, or NF-κB proteins at the same rate. No induction of NF-κB DNA-binding activity was seen in heart, brain, and muscle tissues (Fig. 6). Similarly, no induction of C/EBP activity was seen in the heart tissue (Fig. 4). A constitutive high level of SAF DNA-binding activity was seen in lung, heart, kidney, brain, and skeletal muscle tissues (Fig. 2). The exact role of these constitutively present SAF isoforms is still unclear. One interesting observation of this study is the procreation of tolerance against repeated stimuli that resulted in a lower level of SAA synthesis in all tissues. Such low levels of SAA induction was due to the lower level induction of regulatory transcription factors SAF and NF-κB. Why are lesser amounts of inflammation responsive transcription factors activated in response to a continuous stream of inflammatory stimuli? It is possible that low levels of transcription factor activity arise due to 1) low levels of functionally active proteins, or 2) activation of negative regulatory proteins that may act as counterbalancers. NF-κB is known to be negatively regulated by IκB group of proteins, including IκB-α, IκB-β, (44) and bcl-3 (45). These proteins, by interacting with NF-κB, inactivate the transactivation potential of NF-κB. Recent studies suggested that during persistent viral infection at a later stage, synthesis of IκB-α and IκB-β takes place which inactivate NF-κB activity (46). For the C/EBP group of transcription factors, three proteins, Ig/EBP (47), LIP (48), and CHOP (49), act as negative regulators. These proteins have strong sequence similarities to the C/EBP group of proteins within the b-ZIP region corresponding to the DNA-binding domain. Bacterially expressed CHOP can inhibit the DNA-binding ability of C/EBP-β, whereas Ig/EBP can inhibit the transcriptional ability of C/EBP-β and C/EBP-α. CHOP is shown to be expressed under nutritionally deprived conditions (50). Feedback control mechanisms may activate such inhibitory molecules to reduce the transcription factor activity during the chronic phase of the inflammatory process. Existence of such regulators for SAF has not yet been reported. However, it is possible that one of the multiple SAF family members may possess inhibitory activity (our unpublished observation). Increased levels of such a molecule appearing at the chronic inflammatory phase would reduce transcriptional up-regulatory activity of SAF. Also, phosphorylation of SAF is necessary for its activity (40). It is possible that during chronic inflammatory condition, the kinase activity that modulates SAF activity is under tight control and present only at a low level. Further studies along these lines would shed some light on the mode of regulation of SAF activity during chronic inflammatory condition. As this process becomes fully understood, it will be possible to manipulate the activity of the inflammatory transcription factors and consequently prevent persistent SAA expression for a therapeutic measure against SAA-linked pathogenesis.

Table I.

Relative abundance of SAF, C/EBP, and NF-κB during acute and chronic inflammationa

TissueAgNO3 Treatment (days)Transcription Factors
SAFC/EBPNF-κB
Liver ± − 
 ++ ++ ± 
 18 ++ ++ 
 35 ++ ± 
     
Lung ± 
 +++ +++ ++ 
 18 ++ ++ 
 35 
     
Heart ± ± 
 ± ± 
 18 ± ± 
 35 ± ± 
     
Kidney ± ± 
 ++ ++ ++ 
 18 ++ ++ 
 35 ++ ± 
     
Brain ± − 
 ++ − 
 18 ++ ± 
 35 ++ ± 
     
Muscle − − 
 ++ − 
 18 − − 
 35 − ± 
TissueAgNO3 Treatment (days)Transcription Factors
SAFC/EBPNF-κB
Liver ± − 
 ++ ++ ± 
 18 ++ ++ 
 35 ++ ± 
     
Lung ± 
 +++ +++ ++ 
 18 ++ ++ 
 35 
     
Heart ± ± 
 ± ± 
 18 ± ± 
 35 ± ± 
     
Kidney ± ± 
 ++ ++ ++ 
 18 ++ ++ 
 35 ++ ± 
     
Brain ± − 
 ++ − 
 18 ++ ± 
 35 ++ ± 
     
Muscle − − 
 ++ − 
 18 − − 
 35 − ± 
a

−, ±, +, ++, and +++ denote different levels, from absence to very high levels, of the three transcription factors in six tissues as judged by DNA-binding assays described in Figs. 2, 4, and 6. The relative amounts of DNA-protein complexes with each transcription factor were determined by densitometric scanning of the bands.

1

This work was supported in part by U.S. Public Health Service Grant DK49205 and funds from the College of Veterinary Medicine, University of Missouri.

3

Abbreviations used in this paper: SAA, serum amyloid A; SAF, SAA-activating factor; C/EBP, CCAAT enhancer binding protein; RPA, RNase protection assay.

1
Benditt, E. P., N. Eriksen.
1977
. Amyloid protein SAA is associated with high density lipoprotein from human serum.
Proc. Natl. Acad. Sci. USA
74
:
4025
2
Kushner, I..
1982
. The phenomenon of acute phase response.
Ann. NY Acad. Sci.
389
:
39
3
Sipe, J. D..
1994
. Amyloidosis.
Crit. Rev. Clin. Lab. Sci.
31
:
325
4
Mackiewicz, A., M. K. Ganapathi, D. Scultz, D. Samols, J. Reese, I. Kushner.
1988
. Regulation of rabbit acute phase protein biosynthesis by monokines.
Biochem J.
253
:
851
5
Kisilevsky, R..
1992
. Serum amyloid A changes high density lipoprotein’s cellular affinity: a clue to serum amyloid A’s principal function.
Lab. Invest.
66
:
778
6
Benson, M. D., M. Aldo-Benson.
1979
. Effect of purified protein SAA on immune responses in vitro: mechanisms of suppression.
J. Immunol.
122
:
2077
7
Zimlichman, S., A. Danon, I. Nathan, G. Mozes, R. Shainkin-Kestenbaum.
1990
. Serum amyloid A, an acute phase protein, inhibits platelet activation.
J. Lab. Clin. Med.
116
:
180
8
Linke, R. P., V. Bock, G. Valet, G. Roth.
1991
. Inhibition of oxidative burst response of N-formyl peptide-stimulated neutrophils by serum amyloid A protein.
Biochem. Biophys. Res. Commun.
176
:
1100
9
Betts, J. C., M. R. Edbrooke, R. V. Thakker, P. Woo.
1991
. The human acute-phase protein serum amyloid A gene family: structure, evolution, and expression in hepatoma cells.
Scand. J. Immunol.
34
:
221
10
Whitehead, A. S., M. C. deBeer, D. M. Steel, M. Rits, J. M. Lelias, W. S. Lane, F. C. deBeer.
1992
. Identification of novel members of the serum amyloid A protein super family as constitutive apolipoproteins of high density lipoprotein.
J. Biol. Chem.
267
:
3862
11
Kluve-Beckerman, B., M. L. Drumm, M. D. Benson.
1991
. Non-expression of the human serum amyloid A three (SAA3) gene.
DNA Cell Biol.
10
:
651
12
Sipe, J. D..
1999
. Revised nomenclature for serum amyloid A (SAA).
Amyloid Int. J. Exp. Clin. Invest.
6
:
67
13
Lowell, C. A., D. A. Potter, R. S. Stearman, J. F. Morrow.
1986
. Structure of the murine serum amyloid A gene family.
J. Biol. Chem.
261
:
8442
14
Rygg, M., G. Marhaug, G. Husby, S. B. Dowton.
1991
. Rabbit serum amyloid protein A: expression and primary structure deduced from cDNA sequence.
Scand. J. Immunol.
34
:
727
15
Tatum, F., J. Alam, A. Smith, W. T. Morgan.
1990
. Molecular cloning, nucleotide sequence heterozygosity and regulation of serum amyloid A cDNA.
Nucleic Acids Res.
18
:
7447
16
Ray, B.K., A. Ray.
1991
. Complementary DNA cloning and nucleotide sequence of rabbit serum amyloid A protein.
Biochem. Biophys. Res. Commun.
178
:
68
17
Mitchell, T. I., C. I. Coon, C. E. Brinckerhoff.
1991
. Serum amyloid A (SAA3) produced by rabbit synovial fibroblasts treated with phorbol esters or interleukin 1 induces synthesis of collagenase and is neutralized with specific antiserum.
J. Clin. Invest.
87
:
1177
18
Sano, K..
1988
. Experimental amyloidosis induced by saponin.
Acta Pathol. Jpn.
38
:
1241
19
Ray, B. K., A. Ray.
1991
. Rabbit serum amyloid A gene: cloning, characterization, and sequence analysis.
Biochem. Biophys. Res. Commun.
180
:
1258
20
Glenner, G. G..
1980
. Amyloid deposits and amyloidosis: the β fibrillosis.
N. Engl. J. Med.
302
:
1283
21
Benson, M. D., M. A. Scheinberg, T. Shirahama, E. S. Cathcart, M. Skinner.
1977
. Kinetics of serum amyloid A in casein-induced murine amyloidosis.
J. Clin. Invest.
59
:
412
22
Benson, M. D., A. S. Cohen.
1979
. Serum amyloid A protein in amyloidosis, rheumatic, and neoplastic diseases.
Arthritis Rheum.
22
:
36
23
deBeer, F. C., R. K. Mallya, E. A. Fagan, J. G. Lanham, G. R. V. Hughs, M. B. Pepys.
1982
. Serum amyloid A protein concentration in inflammatory diseases and its relationship to the incidence of reactive systemic amyloidosis.
Lancet
2
:
231
24
Loose, L. D., B. H. Littman, J. D. Sipe.
1990
. Inhibition of acute phase proteins by tenidap.
Clin. Res.
38
:
A579
25
Brinckerhoff, C. E., T. I. Mitchell, M. J. Karmilowicz, B. Kluve-Beckerman, M. D. Benson.
1989
. Autocrine induction of collagenase by serum amyloid A-like and β2-macroglobulin-like proteins.
Science
243
:
655
26
Meek, R.L., S. Urieli-Shovel, E. P. Benditt.
1994
. Expression of apolipoprotein serum amyloid A mRNA in human atherosclerotic lesions and cultured vascular cells: implications for serum amyloid A function.
Proc. Natl. Acad. Sci. USA
91
:
3186
27
Liao, F., A. Andalibi, J. H. Qiao, H. Allayee, A. M. Fogelman, A. J. Lusis.
1994
. Genetic evidence for a common pathway mediating oxidative stress, inflammatory gene induction, and aortic fatty streak formation in mice.
J. Clin. Invest.
94
:
877
28
Ray, B. K., S. Chatterjee, A. Ray.
1999
. Mechanism of MM-LDL mediated induction of serum amyloid A gene in monocyte/macrophage cells.
DNA Cell Biol.
18
:
65
29
Ray, A., B. K. Ray.
1997
. Serum amyloid A gene expression level in liver in response to different inflammatory agents is dependent upon the nature of activated transcription factors.
DNA Cell Biol.
16
:
1
30
Ray, A., B. K. Ray.
1994
. Serum amyloid A gene expression under acute-phase condition involves participation of inducible C/EBP-β and C/EBP-δ and their activation by protein phosphorylation.
Mol. Cell. Biol.
14
:
4324
31
Ray, B. K., A. Ray.
1993
. Functional NF-κB element in rabbit serum amyloid A gene and its role in acute phase induction.
Biochem. Biophys. Res. Commun.
193
:
1159
32
Ray, A., M. Hannink, B. K. Ray.
1995
. Concerted participation of NF-κB and C/EBP heteromer in lipopolysaccharide induction of serum amyloid A gene expression in liver.
J. Biol. Chem.
270
:
7365
33
Huang, J. H., W. S. L. Liao.
1994
. Induction of mouse serum amyloid A3 gene by cytokines require both C/EBP family proteins and a novel constitutive nuclear factor.
Mol. Cell. Biol.
14
:
4475
34
Betts, J. C., J. K. Cheshire, S. Akira, T. Kishimoto, P. Woo.
1993
. The role of NF-κB and NF-IL6 transactivating factors in the synergistic activation of human serum amyloid A gene expression by interleukin-1 and interleukin-6.
J. Biol. Chem.
268
:
25624
35
Li, X., W. S. L. Liao.
1991
. Expression of rat serum amyloid A1 gene involves both C/EBP-like and NF-κB-like transcription factors.
J. Biol. Chem.
266
:
15192
36
Edbrooke, M. R., D. W. Burt, J. K. Cheshire, P. Woo.
1989
. Identification of cis-acting sequences responsible for phorbol ester induction of human serum amyloid A gene expression via a nuclear factor κB-like transcription factor.
Mol. Cell. Biol.
9
:
1908
37
Ray, A., B. K. Ray.
1996
. A novel cis-acting element is essential for cytokine mediated transcriptional induction of the serum amyloid A gene in nonhepatic cells.
Mol. Cell. Biol.
16
:
1584
38
Ray, B. K., A. Ray.
1997
. Involvement of an SAF-like transcription factor in the activation of serum amyloid A gene in monocyte/macrophage cells by lipopolysaccharide.
Biochemistry
36
:
4662
39
Ray, B. K., A. Ray.
1997
. Induction of serum amyloid A (SAA) gene by SAA-activating sequence-binding factor (SAF) in monocyte/macrophage cells. Evidence for a functional synergy between SAF and Sp1.
J. Biol. Chem.
272
:
28948
40
Ray, A., B. K. Ray.
1998
. Isolation and functional characterization of cDNA of serum amyloid A-activating factor that binds to the serum amyloid A promoter.
Mol. Cell. Biol.
18
:
7327
41
Chomczynski, P., N. Sacchi.
1987
. Single step method of RNA isolation by acidic guanidinium thiocyanate-phenol-chloroform extraction.
Anal. Biochem.
162
:
156
42
Bradford, M. M..
1976
. A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding.
Anal. Biochem.
72
:
248
43
Pepys, M. B..
1981
. Serum C-reactive protein, serum amyloid P component and serum amyloid A protein in autoimmune disease.
Clin. Immunol. Allergy
1981
:
77
44
Zabel, U., P. A. Baeuerle.
1990
. Purified human IκB can rapidly dissociate the complex of the NF-κB transcription factor with its cognate DNA.
Cell
65
:
225
45
Wulczyn, F. G., M. Neumann, C. Scheidereit.
1992
. Candidate proto-oncogene bcl-3 encodes a subunit specific inhibitor of transcription factor NF-κB.
Nature
358
:
597
46
Bitko, V., S. Barik.
1998
. Persistent activation of RelA by respiratory syncytial virus involves protein kinase C, underphosphorylated IκBβ, and sequestration of protein phosphatase 2A by the viral phosphoprotein.
J. Virol.
72
:
5610
47
Cooper, C., A. Henderson, S. Artandi, N. Avitahl, K. Calame.
1995
. Ig/EBP (C/EBP-γ) is a transdominant negative inhibitor of C/EBP family transcriptional activators.
Nucleic Acids Res.
23
:
4371
48
Decombes, P., U. Scheibler.
1991
. A liver-enriched transcriptional activator protein, LAP, and a transcriptional inhibitory protein, LIP, are translated from the same mRNA.
Cell
67
:
569
49
Ron, D., J. F. Habener.
1992
. CHOP, a novel developmentally regulated nuclear protein that dimerizes with transcription factors C/EBP and LAP and functions as a dominant negative inhibitor of gene transcription.
Genes Dev.
6
:
439
50
Flemming, J. V., S. M. Hay, D. N. Harries, W. D. Rees.
1998
. Effects of nutrient deprivation and differentiation on the expression of growth arrest genes (gas and gadd) in F9 embryonal carcinoma cells.
Biochem J.
330
:
573