Luminal polyamines stimulate repair of gastric mucosal stress ulcers JIAN-YING

WANG

AND

LEONARD

R. JOHNSON

Department of Physiology and Biophysics, University Memphis, Tennessee 38163

WANG,

polyamines

JIAN-YING,

stimulate

AND LEONARD R. JOHNSON. Luminczl repair of gastric mucosal stress ulcers. Am.

J. Physiol. 259 (Gastrointest. Liver Physiol. 22): G584-G592, 1990.-The purpose of this study was to examine whether luminal polyamines can substitute for tissue polyamines in the healing process of gastric mucosal stress ulcers. Rats were fasted 22 h, placed in restraint cages, and immersed in water to the xiphoid process for 6 h. Animals were killed either immediately or at 4,12, or 24 h after the period of stress. Stress significantly increased ornithine decarboxylase (ODC) activity and tissue polyamine content. Mucosal polyamine levels peaked 4 h after stress and remained significantly elevated for 12 h. The healing process, which was significant by 12 h, was inhibited by DL-a-difluoromethylornithine (DFMO), a specific inhibitor of ODC. DFMO totally prevented the marked increases in ODC and polyamine levels that usually followed stress. Oral administration of polyamines, putrescine, cadaverine, spermidine, or spermine, immediately after stress increased the normal rate of healing and prevented the inhibition of repair caused by DFMO. Spermidine or spermine accelerated healing better than putrescine or cadaverine. The delayed recovery of mucosal DNA, RNA, and protein content after stress in the DFMO-treated rats was also significantly prevented by exogenous polyamines. The reduced amounts of gastric mucosal spermidine and spermine in rats treated with DFMO returned toward control levels after administration of exogenous spermidine (100 mg/kg). These results show that 1) increased levels of polyamines provided by ODC are absolutely required for normal healing of gastric mucosal stress ulcers, 2) the polyamines are active from the luminal side, and 3) polyamines accelerate healing at least partly through a mechanism involving cell renewal. damage; healing; stomach; deoxyribonucleic acid; ribonucleic acid; rat; DL-a-difluoromethylornithine GASTRIC MUCOSAL STRESS ULCERS heal rapidly in rats and are almost totally repaired 24 h after the period of stress (37). Human stress ulcerations usually disappear within days (14). Full repair of the damaged mucosa consists of at least two different mechanisms. One, the rapid process of mucosal restitution, occurs by sloughing off the damaged epithelial cells and migration of remaining viable cells from areas adjacent to, or just beneath, the injured surface to cover the denuded areas (22, 26). The other is the replacement of lost cells by the process of cell division and is much slower (15). Recent studies (16, 20, 25) have shown that the polyamines, spermidine and spermine, and their precursor putrescine are intimately involved in, and required for, G584

of Tennessee Medical School,

cell growth and differentiation, and their concentrations within cells are highly regulated. Intracellular polyamine levels are highly dependent on the activity of ornithine decarboxylase (ODC). The enzyme is the initial ratelimiting step in polyamine biosynthesis and is induced by a number of hormones and growth factors (28, 39). Although the exact mechanism of action polyamines is uncertain, numerous data suggest that these substances regulate cellular functions by virtue of their efforts on enzyme activities, nucleic acid synthesis and stabilization, protein synthesis, and membrane structure (19,20). We recently investigated the relationship between ODC activity and gastric lesion formation and repair in a stress-ulcer model (27, 38). Those studies indicated that 6 h of stress increased ODC activity significantly and that enzyme activity remained markedly elevated over that of the corresponding controls for 12 h after the period of stress. Peak enzyme activity occurred 4 h after the stress period. Inhibition of ODC with DL-a-difluoromethylornithine (DFMO) did not alter the degree of damage when assays were run immediately after stress, but DFMO significantly delayed recovery following stress. We suggested that endogenous mucosal polyamines may be important in the mucosal repair process. In support of this hypothesis, intragastric administration of polyamines has been shown to prevent mucosal lesions caused by 1 h exposure to acidified ethanol (17, 18). Using immunocytochemical techniques we have shown that ODC and presumably polyamines are released from granules of the mucous neck cells during stress damage of the rat stomach (unpublished data). This finding suggests that luminal polyamines may be important in the mucosal repair process. The current study was designed to answer several questions regarding the mechanism of the involvement of increased ODC .activity in the healing process. First, can exogenous polyamines substitute in the healing process for the endogenous polyamines produced by ODC? This question was examined by providing polyamines to rats treated with DFMO. Second, are polyamines effective when present on the luminal side of the stomach? Third, can luminal polyamines accelerate the normal rate of healing? MATERIALS AND METHODS Materials and experimental design. L- [ l-‘4C]ornithine (sp act 51.6 mCi/nmol) was obtained from New England Nuclear, Boston, MA. DFMO was the kind gift of Merrell

0193-1857/90 $1.50 Copyright 0 1990 the American Physiological

Society

Downloaded from www.physiology.org/journal/ajpgi at Glasgow Univ Lib (130.209.006.061) on February 14, 2019.

POLYAMINES

AND

Dow Research Institute, Cincinnati, OH. Putrescine, cadaverine, spermidine, and spermine as hydrochloride salts were purchased from Sigma Chemical, St. Louis, MO, and suspended in 0.9% normal saline immediately before use. Male Sprague-Dawley rats weighing between 125-150 g were housed in wire-bottomed raised cages and given water and standard laboratory rat food ad libitum. All animals were obtained from Timco Breeding Laboratories, Houston, TX. The animal quarters were maintained at a temperature of 22 t 1°C with a 12-h light-dark cycle. Animals were deprived of food but allowed free access to tap water for 22 h before the experiments. Each study was carried out using five to eight rats per group. In the first study, changes of endogenous polyamine levels and the mucosal repair process were examined in stressed rats with or without treatment with DFMO. Animals were placed in restraint cages identical to ones described in detail elsewhere (29). The rats were then immersed vertically to the level of the xiphoid process in a water bath (23°C) for 6 h. DFMO was dissolved in distilled water and administered intraperitoneally (500 mg/kg) 10 min before the period of stress and repeated at 8-h intervals in the same dose for up to 24 h after stress. The dose of DFMO used in the present study extensively delayed the normal repair of gastric mucosal stress ulcers (38). Control rats were injected with saline. The animals were killed by ether anesthesia and exsanguination immediately or at 4, 12, and 24 h after the period of stress. The stomachs were removed, opened along the greater curvature, and rinsed in ice-cold saline. The stomachs were laid flat on a Petri dish inverted over ice and inspected for gross damage. Takagi and Okabe’s method (29) as modified by Thirumalai et al. (33) was used to determine the severity of lesions. The incidence of lesions was noted, and the lesions were traced onto waxed paper. The length of visible lesions was measured, and the ulcer index was reported as total lesion length per tissue. The antrum was removed, and the oxyntic gland mucosa was scraped from the underlying smooth muscle with a glass slide. The mucosal scrapings were weighed and divided into two portions. One was assayed to determine polyamine levels. The other was used for measurement of ODC activity. In the second study, the effects of exogenous polyamines, putrescine, cadaverine, spermidine, and spermine, on the repair of gastric mucosal stress ulcers in DFMOtreated rats or in normal rats were determined. Mucosal repair was delayed by administering DFMO interperitoneally as described above, and polyamines were administered intragastrically in doses of 50 or 100 mg/kg immediately after the period of stress and repeated at 8-h intervals for up to 24 h thereafter. In animals allowed to recover normally (without DFMO), polyamines in doses of 100 mg/kg were administered intragastrically immediately after and repeated at 6-h intervals after stress. Control animals received the vehicle alone. Animals were killed either 24 h (the DFMO-treated rats) or 12 h (normal rats) after the period of stress, and stomachs were examined for ulcers macroscopically and microscopically.

MUCOSAL

REPAIR

G585

In the third study, the influences of exogenous spermidine on mucosal nucleic acid, ODC, and polyamine levels were examined in stressed rats treated with DFMO. To determine the time course of the effect of spermidine on ODC activity and nucleic acid content, DFMO (500 mg/kg) was administered intraperitoneally 10 min before stress and at 8-h intervals for up to 24 h after the period of stress. Spermidine (100 mg/kg) was given intragastrically immediately after stress and repeated at 8-h intervals for up to 24 h thereafter. Animals were killed either immediately or 4, 12, or 24 h after the period of stress. In studies of mucosal polyamine levels, DFMO (500 mg/kg) was administered intraperitoneally 10 min before stress. Spermidine in doses of 50 or 100 mg/kg was given intragastrically immediately after the period of stress. Control rats were treated with saline. Animals were killed 1 h after administration of spermidine. ODC assays. The activity of the enzyme ornithine decarboxylase was assayed by a radiometric technique in which the amount of 14C02 liberated from DL-[~-~~C]ornithine was estimated (21). Tissue samples were collected as above and placed in 1.0 ml (pH 7.4) 0.067 M sodium potassium phosphate, 10 PM EDTA, and 2 mM dithiothreitol (DTT). ODC activity is dependent on the presence of pyridoxal phosphate, and DTT stimulates and helps stabilize the enzymatic activity (21). The mucosa was homogenized, sonicated, and then centrifuged at 30,000 g at 4°C for 30 min. An aliquot from the 30,000g supernatant was incubated in stoppered vials in the presence of 2 pmol of [14C]ornithine for 15 min at 37°C. The 14COZ liberated by the decarboxylation of ornithine was trapped on a piece of filter paper impregnated with 20 ~1 of 2 N NaOH, which was suspended in a center well above the reaction mixture. The reaction was stopped by the addition of trichloroacetic acid to a final concentration of 10%. The 14C02 trapped in the filter paper was measured by liquid scintillation spectroscopy at a counting efficiency of 99%. Aliquots of the 30,000-g supernatant were assayed for total protein, using the method described by Bradford (3). Enzymatic activity was expressed as picomoles of COa per milligram of protein per hour. Measurement of nucleic acid and protein. RNA content of the gastric mucosa was measured by the orcinol reaction (5). DNA was dissolved in 10% perchloric acid at 70°C and measured by a modification (10) of the Burton procedure (4). Protein finally was dissolved in 1 M NaOH and measured by the Bradford technique (3). The amount of DNA, RNA, and protein was expressed as the total amount per entire oxyntic gland mucosa. High-performance liquid chromatographic (HPLC) analysis of mucosal polyamines. The mucosal polyamine

content was determined by the method of Tsai and Lin (34). Mucosal scrapings were weighed and collected in glass tubes and frozen in -85°C until analyzed. The samples were placed in 0.4 N perchloric acid, homogenized for 20 s, ultrasonicated for 15 s, and centrifuged at 1,600 g for 10 min. This supernatant was collected, neutralized to pH 7.0 with 6 N KOH, and centrifuged to remove the precipitate. A 0.2-ml aliquot of the solution

Downloaded from www.physiology.org/journal/ajpgi at Glasgow Univ Lib (130.209.006.061) on February 14, 2019.

G586

POLYAMINES

AND

was delivered to a 0.35-ml vial equipped with a Teflonlined screw cap. After addition of 100 ~1 saturated Na2C03 and 200 ~1 dansyl chloride solution (10 mg/ml acetone), reaction was allowed to proceed by heating at 70°C for 30 min and then added to 0.5 ml glass distilled water and 2.5 ml toluene. After mixing and centrifugation, the organic portion was collected and evaporated to dryness by a nitrogen stream. To the residue, 500 ~1 acetone were added and filtered, and an aliquot of 200 ~1 was used for HPLC analysis. A Waters HPLC system (Milford, MA), which included a 710 Wisp automatic sample injector, two 6000 A solvent delivery units, a 680 gradient controller, and a Novapack Cl8 column in a radial compression module, was used. The fluorescence detector was a monochromator. Solvents A and B were composed of acetonitrile, water, glacial acetic acid, and triethylamine in the proportions of 40:60:0.02:0.001 and 95:5:0.02:0.005, respectively. The mobile phases used in this separation consisted of a program of a linear gradient starting at 68% solvent A and 32% solvent B and increasing in solvent B linearly to 100%. Each sample was run for 13 min, and the equilibration delay between injections was 2 min. Enough of the mobile phases (A and B) was prepared fresh before starting the automatic injector. Measurements of putrescine, spermidine, and spermine were made by comparing ratios of polyamines to l,lOdiaminodecane peak heights with a standard curve. Samples for a calibration curve were obtained by adding known amounts of standards in 1 ml of glass distilled water followed by extraction and dansylation as previously described. The amount of polyamines was expressed as micromoles per 100 mg of mucosal weight. Histology. The oxyntic gland areas were collected immediately or 4, 12, or 24 h after 6 h of stress. Animals were quickly anesthetized with ether and opened via a midline laparotomy. After gastric contents were expressed into the duodenum by gentle pressure on the stomach, 2 ml of 4% Formalin were injected into the stomach, which was then placed into 4% Formalin for 10 min. Subsequently, the stomach was incised along the greater curvature and immersed in 10% Formalin. Samples of stomach were excised from the gastric glandular epithelium along the greater curvature at a region located 2 mm below the limiting ridge, which separates the forestomach from the glandular epithelium. Blocks were removed at right angles to the long axis of the stomach. A light microscopic sample measuring 1 X 10 mm was removed. The same procedure was performed on a second sample that was removed 0.5 cm distally from the first sampling area (37). Thus two light-microscopic blocks were removed and processed from each tissue. Four rats were examined from each group. Routine paraffin-embedded hematoxylinand eosin-stained sections were used for light-microscopic evaluation. Surface mucous cells were considered injured if one or more of the following criteria were present: marked cytoplasmic vacuolization, cytoplasmic swelling, nuclear pyknosis, or nuclear swelling with chromatin margination. Statistics. Data are presented as means t SE from five to eight rats per group. Statistical analysis was performed

9IUCOSAL

REPAIR

using a two-tailed Dunnett’s multiple comparison test (13) for unpaired variates, and values of P c 0.05 were regarded as significant. RESULTS

Effects of stress on mucosal polyamine levels. Gastric mucosal polyamine levels increased significantly following stress and remained significantly elevated over those of the corresponding controls for 12 h (Fig. 1). The maximum increases in putrescine, spermidine, and spermine occurred 4 h after the period of stress and were ~6.0, 2.1, and 1.3 times the control prestress levels, respectively. By 24 h mucosal putrescine and spermine levels had returned to near normal, whereas spermidine levels remained increased significantly. The administration of 500 mg/kg of DFMO, which totally inhibited ODC activity and extensively delayed the normal repair of gastric mucosa after stress (38), completely prevented the increased accumulation of polyamines (Fig. 1). The only significant decrease below basal levels in the measured parameters in the animals 2.0

&dTht

-m-w-

--

--

+

0-J

// I

-6

/I

// I

0

I

1

a Hours after Stress 4

1

12

II

1

24

FIG. 1. Polyamine levels of oxyntic gland mucosa in rats stressed for 6 h and those stressed and treated with DL-cu-difluoromethylornithine (DFMO). DFMO was administered intraperitoneally 10 min before stress and at 8-h intervals for up to 24 h after stress. Measurements were taken immediately and 4, 12, and 24 h after stress. Values represent means t SE from 5 or 6 rats/group. *t Significantly different from control and stress groups, respectively, at P < 0.05.

Downloaded from www.physiology.org/journal/ajpgi at Glasgow Univ Lib (130.209.006.061) on February 14, 2019.

POLYAMINES

stressed and treated with DFMO mucosal spermine (Fig. 1, bottom).

AND

was a decrease in

Effects of polyamine administration on delayed and normal repair of stress-induced ulcers. Stress for 6 h

consistently induced visible lesions in the oxyntic gland mucosa (Fig. 2). These lesions appeared as elongated bands ranging from 2 to 10 mm in length and from 1 to 3 mm in width, generally parallel to the long axis of the stomach. Histologically, the gastric mucosa after 6 h of stress showed a discontinuous surface with many cells sloughed off into the lumen. The gastric pits were greatly shortened, and some were eliminated entirely. As shown in Fig. 2, the ulcer index, when it was measured immediately after stress, was not affected by DFMO. Exogenous administration of polyamines, putrescine, cadaverine, spermidine, and spermine, also failed to prevent the formation of gastric mucosal stress ulcers (Fig. 2). The ulcer index in rats pretreated with polyamines was indistinguishable from that of the stress alone group. However, following stress the mucosa recovered quickly, as evidenced by a decrease in the ulcer index. By 24 h after stress most visible erosions had disappeared (Fig. 3). When DFMO was administered intraperitoneally, however, repair was extensively delayed compared to animals treated with the vehicle alone (Fig. 3). There was no significant decline in the ulcer index after 24 h in the animals treated with DFMO. The mucosal surface remained discontinuous, showing sloughed cells and debris (Fig. 4A). Damage still extended into the gland, and the mucosal thickness was one-half to two-thirds of normal. Spermidine in a dose of 100 mg/kg or in doses of 50 and 100 mg/kg significantly prevented the DFMO induced delay in healing at 12 and 24 h, respectively. Histological studies showed that the gastric mucosa from rats given polyamines along with DFMO was nearly normal. Gastric pits and glands were evident and most of the mucosal surface had a complete epithelial lining. The effects of 100 mg/kg of spermidine administered for 24 h to DFMO-treated rats are shown in Fig. 4B. Figure 5 shows that 24 h after stress lOO-mg/kg doses

3530. PS E. v 20u = ii -0

15-

=

10.

5O50

loo Dose

loo

50

loo

50

loo

OWW

FIG. 2. Effects of a-difluoromethylornithine (DFMO) and polyamines on formation of stress-induced ulcers in gastric mucosa of rats. DFMO was administered intraperitoneally 10 min before stress, and polyamines were given intragastrically 30 min before period of stress. Values represent means + SE of 6 observations.

MUCOSAL 30

G587

REPAIR

12 hour *

1

24 hour

Control

100 m@g

Control I

I DFMO (WOmg/kg)

?prmidine I

I DFMO (bOOmg/kg)

FIG. 3. Effects of cu-difluoromethylornithine (DFMO) and spermidine given together with DFMO on the recovery of stress-induced gastric damage in rats. DFMO was administered intraperitoneally 10 min before stress and repeated at 8-h intervals after stress. Spermidine was given intragastrically immediately after period of stress and repeated at 8-h intervals for up to 12 or 24 h thereafter. Values represent means + SE from 5 rats/group. *t Significantly different from control and DFMO-treated groups, respectively, at P < 0.05.

of putrescine, cadaverine, and spermine, like spermidine, also stimulated significant mucosal repair in rats treated with DMFO. As was the case with spermidine (Fig. 3), spermine totally prevented the effects of DFMO. The diamines, putrescine and cadaverine, were roughly only one-half as effective as the polyamines, spermidine and spermine. Administration of putrescine and cadaverine at 100 mg/kg had no effect on the rate of normal repair 12 h after damage (Fig. 6). When spermidine and spermine were given at a dose of 100 mg/kg, however, repair was accelerated significantly, the recovery rates being 56.6, and 7&O%, respectively, compared with the controls. Effects of spermidine on ODC activity and mucosal DNA, RNA, and protein. Stress significantly increased ODC activity in gastric mucosa (Fig. 7A). The increase

in enzyme activity, which peaked 4 h after the period of stress, was totally inhibited by DFMO. Spermidine at a dose of 100 mg/kg, together with DFMO, had no effect on ODC activity, but it significantly prevented the inhibition of mucosal repair caused by DFMO (Fig. 7B). Statistically significant repair in the DFMO-treated animals began 4 h after administration of spermidine. At 12 h, mucosal repair was considerable and .almost complete by 24 h. As can be seen in Fig. 8, DNA, RNA, and protein content of the gastric mucosa were significantly decreased immediately after the period of stress. The levels of these parameters were gradually restored and reached normal levels 24 h after stress. The recovery of DNA, RNA, and protein content was completely prevented by the administration of DFMO. Twenty-hour hours after stress, in all three cases, the values for the DFMO group were still significantly lower than the corresponding controls. When 100 mg/kg of spermidine were administered together with DFMO, however, DNA, RNA, and protein content recovered at normal rates. After 24 h there were

Downloaded from www.physiology.org/journal/ajpgi at Glasgow Univ Lib (130.209.006.061) on February 14, 2019.

G588

POLYAMINES

AND

MUCOSAL

REPAIR

FIG. 4. Microscopic appearance of gastric mucosa in rats after treatment with Lu-difluoromethylornithine (DFMO, 500 mg/kg) alone (A) and DFMO plus spermidine (100 mg/kg) (B) 24 h after period of stress. Paraffin-embedded with hematoxylin and eosin staining. Original magnification ~180. Bar = 50 pm.

30

I!z. -z

20

v -0 E & 3-0

10

0

Control

100

100

Putrescine DFMO

Cadaverine

100 mglkg

0

Spermine

E s 0 -

I

.-E :: iz a” 100

Dose (mg/kg)

(5OOmg/kg)

FIG. 5. Effects of cu-difluoromethylornithine (DFMO) and putrestine, cadaverine, and spermine given together with DFMO on recovery of stress-induced gastric damage in rats. These agents were given as the same way described in Fig. 4. Animals were killed 24 h after period of stress. Values represent means * SE from 5 rats/group. *t Significantly different from control and DFMO-treated groups, respectively, at P < 0.05.

no significant differences in these parameters between the stressed alone and the stress group treated with DFMO plus spermidine.

FIG. 6. Effects of luminal stress-induced gastric ulcers. after stress and repeated at after stress. Values represent icantly different from control

polyamines on the normal recovery from Polyamines were given orally immediately 6-h intervals. Animals were killed 12 h means + SE from 5 rats/group. * Signifat P < 0.05.

stressed rats treated with DFMO were returned toward control levels after administration of exogenous spermidine. The increases were statistically significant compared to the DFMO group.

Effects of spermidine on endogenous polyamine levels in gastric mucosa. Gastric mucosa polyamine levels were

DISCUSSION

significantly increased following stress (Fig. 9). A single administration of 500 mg/kg of DFMO prevented these increases. The combined administration of 50 or 100 mg/ kg of exogenous spermidine with DFMO had no influence on mucosal putrescine levels. However, the reduced levels of spermidine and spermine in the gastric mucosa of

The present studies confirmed our previous finding (38) that inhibition of ODC activity by repeated administration of DFMO extensively delayed the rate of repair of gastric mucosal stress ulcers. The current studies demonstrated that the repair of these ulcers is accompanied by significant increases in mucosal polyamine

Downloaded from www.physiology.org/journal/ajpgi at Glasgow Univ Lib (130.209.006.061) on February 14, 2019.

POLYAMINES

AND

MUCOSAL

G589

REPAIR

llOOlOOO-

G ,Ii.

a z n

900 800700600-

*

500 -

*

Stress + DFMO

;

*

0f 2200 2000 E

1800 1600

2

2 a z

E,*o-

K

1400

25 -

0ii+ -

15-

2 10 -

35-

5-

30-

O-6

0

4

8

12

24

Hours after Stress FIG. 7. Ornithine decarboxylase (ODC) activity (A) and ulcer indexes (B) of oxyntic gland mucosa from stressed rats and stressed rats treated with DFMO alone and together with spermidine. Values represent means k SE from 5 rats/group. *+ Significantly different from stressed rats and stressed rats treated with DFMO groups, respectively, at P C 0.05.

levels and that these increases in tissue polyamines are totally prevented by inhibiting ODC activity (Fig. 1). The most significant of the new findings reported in this paper, however, is that luminal polyamines can substitute for endogenous mucosal polyamines in rats treated with DFMO to block ODC (Figs. 3-5). Furthermore, luminal polyamines significantly accelerated the normal healing rate of stress ulcers (Fig. 6). Polyamines are polycations found in high concentrations in all eukaryotic cells thus far examined. Physiological concentrations of polyamines stimulate many of the reactions involved in macromolecular synthesis (19). For example, polyamine synthesis usually precedes DNA synthesis, and depletion of polyamines attenuates trophic responses in a number of tissues (23). Treatment with polyamines enhances the translation of a variety of mRNAs (1) and affects the organization of tertiary structures of tRNA (7). Furthermore, these cations influence DNA structure (8) and enhance the reactions involved in transcription and replication of DNA in a variety of cell-free systems (2). Thus polyamines are intimately involved in the regulation of normal cell growth and differentiation. The gastrointestinal mucosa is one of the most rapidly proliferating tissues in the body. Cells lost into the lumen are replaced by new cells. Several experiments suggest the importance of cell renewal in the ulcerogenic process.

252015los 0’41 -6

1 0

I

I

I

4

8

12

/I I/

1 24

Hours after Stress FIG. 8. DNA, RNA, and protein content of oxyntic gland mucosa from rats described in Fig. 7. Values represent means t SE from 5 or 6 rats/group. *+ Significantly different from control and stressed groups, respectively, at P < 0.05.

Vanamee et al. (35) found that growth hormone reduced the incidence of gastric erosion caused by restraint stress in rats. Our laboratory demonstrated that the susceptibility of rats to stress ulcers was directly correlated with the rate of mucosal DNA synthesis (30). In a chronic experimental ulcer model, pentagastrin was shown to accelerate the healing of acetic acid-induced gastric and duodenal ulcers (31). Although the necessity for polyamines during growth of gastrointestinal mucosa is becoming appreciated, no studies have specifically examined the role of endogenous and exogenous polyamines in recovery after gastric mucosal damage. Several, however, have examined whether polyamines are involved in or can influence the ulcerogenie process. Mizui and co-workers (17, 18) found that polyamines prevented mucosal lesions produced by exposure to acidified ethanol for 1 h. They suggested the beneficial effects were due to antiperoxidative properties of the polyamines. We investigated the relationship between ODC activity and gastric lesion formation in rats

Downloaded from www.physiology.org/journal/ajpgi at Glasgow Univ Lib (130.209.006.061) on February 14, 2019.

Gi590

POLYAMINES

AND MUCOSAL

0.6 1

I

25.

l

Control Stress

t

v

I

I

100 mgkg

S5poermidine, 1 DFMO @OOmg/kg)

FIG. 9. Gastric mucosal polyamine levels from controls, stressed, stressed plus a-difluoromethylornithine (DFMO), and stressed treated with DFMO and spermidine groups. DFMO was administered intraperitoneally 10 min before stress. Spermidine was given intragastrically 6 h after DFMO treatment. Animals were killed 1 h after administration of spermidine. Values represent means + SE from 5 rats/group. * P C 0.05 compared with control group; t P < 0.05 compared with stress group; v P < 0.05 compared with rats stressed and treated with DFMO.

given hypertonic NaCl by gastric intubation (33). Enzyme activity increased during NaCl damage, and the increase was directly related to the severity of injury. DFMO totally inhibited the enzyme activity and increased the severity of lesions. Recently, we have used a stress-ulcer model to attempt to decide whether polyamine metabolism influences the damage process itself, the repair process, or both (37, 38). Stress significantly increased ODC activity, and the enzyme activity was inhibited by DFMO. In contrast to the NaCl experiments, the inhibition of ODC activity in stressed rats did not alter the degree of damage but totally prevented healing of damaged mucosa. From these results we concluded that although polyamine synthesis does not play

REPAIR

a role in stress-ulcer formation, the increased ODC is necessary for the normal repair of the mucosa following stress (38). The current and our previous findings demonstrate that the polyamines resulting from increased ODC activity are essential to the normal repair of the mucosa. Careful inspection of the results indicates that newly synthesized polyamines may become part of two separate pools, both of which are required for optimal repair but for different repair processes. Repair of mucosal damage involves at least two processes: replacement of damaged cells by cell division and the migration of undamaged cells into injured areas. This latter process is called early mucosal restitution (26). We are suggesting that newly synthesized intracellular polyamines are necessary for cell division and that another fraction of these substances becomes extracellular and supports early mucosal restitution. As we have already reviewed, there is no doubt the polyamines are required for cell division. The DNA data shown in Fig. 8 indicate that cell renewal is at least partly responsible for the healing we observed after polyamine administration. It is also evident that increases in polyamine levels are necessary for observed increases in DNA content, because DFMO prevented the entire recovery of DNA levels. DNA levels did not increase significantly until 24 h after stress, a period of time more than sufficient for cell division to account for at least a portion of the healing (41). Furthermore, exogenously administered spermidine at a dose of 100 mg/kg totally prevented the inhibition of DNA recovery caused by DFMO. Early mucosal restitution appears to be an initial response to prevent noxious luminal agents from causing deeper mucosal damage and occurs too rapidly to be accounted for by cell division (27). Critchlow et al. (9) demonstrated that Ca2+ was required for complete restitution of the surface epithelium of the frog stomach. These studies were performed in vitro, and restitution occurred by migration of viable gastric pit cells. There is also considerable evidence that endogenous as well as exogenous prostaglandins are involved in mucosal restitution (32, 36). The earliest evidence of repair in the current study was a significant decrease in the ulcer index 4 h after stress in DFMO treated rats given luminal spermidine (Fig. 7B). The ulcer index of this group was -40% lower than that of the group receiving DFMO without spermidine. There was an additional large decrease in the ulcer index between 4 and 12 h. As can be seen in Fig. 6, by 12 h both spermidine and spermine caused significant increases in healing in the absence of DFMO. Based on the 16- to 18-h time interval for an increase in the gastrointestinal mucosal labeling index following aspirin damage (41), the repair occurring before 12 h may largely be due to mucosal restitution, whereas the additional repair after 12 h may be due to cell division. Polyamine levels were significantly elevated immediately after the period of stress and remained so for 12 h (Fig. 1). Our previous study demonstrated that, during stress, ODC activity was significantly increased by 4 h (37). Therefore

Downloaded from www.physiology.org/journal/ajpgi at Glasgow Univ Lib (130.209.006.061) on February 14, 2019.

POLYAMINES

AND

polyamine biosynthesis increases sufficiently early to be involved in mucosal restitution or the early repair process. Because the polyamines contain protonated amine groups, they are excellent substrates for transglutaminase and facilitate protein cross-linking (6). These reactions occur both extracellularly, as in the formation of the cervical plug in rat seminal plasma (40), and intracellularly, as in the case of hepatocyte nuclear protein (12). Physiological concentrations of polyamines can act as promoters as well as inducers of membrane fusion and stabilization (24). Since rapid mucosal repair requires Ca2+ (9) and some transglutaminases respond to Ca2+ (ll), polyamines may function in early mucosal restitution by stabilizing cell membranes and forming a matrix for cell migration into damaged areas. In the current study, spermidine and spermine, which possess three and four positive charges at physiological pH, respectively, stimulated mucosal healing better than putrescine or cadaverine, two groups each. Thus the compounds containing more protonated amines to take part in crosslinking were more efficient in healing. Figure 1 clearly shows that mucosal levels of putrescine and both polyamines increase significantly after damage. Luminal polyamines therefore do not speed normal repair by replacing mucosal stores of polyamines. However, luminal polyamines could speed the process of cell migration by supplying additional polyamines to the serosal surface to participate in cross-linking reactions. Although this is a plausible explanation of our results, it remains to be demonstrated that polyamines are involved in protein cross-linking in the damaged mucosa and that cross-linking is essential for cell migration. In summary, these results indicate that increased levels of polyamines provided by ODC are absolutely required for normal healing of gastric mucosal stress ulcers, at least partly through the process of cell renewal. However, the chemical nature of the polyamines and the time source of the repair process also suggest that polyamines may be necessary for early mucosal restitution that is independent of cell division. That luminal polyamines can effectively substitute for newly synthesized endogenous polyamines supports the hypothesis that they may enhance the migration of cells into the damaged area. The authors thank Dr. Y.-H. Tsai for her assistance with the polyamine measurements and J. Pastore for making the illustrations. This work was supported by the National Institute of Diabetes and Digestive and Kidney Diseases Grant POl-DK-37260. Address for reprint requests: L. R. Johnson, Dept. of Physiology and Biophysics, University of Tennessee Medical School, Nash Research Bldg., 894 Union Ave., Memphis, TN 38163. Received

7 March

1990;

accepted

in final

form

12 June

1990.

REFERENCES 1. ATKINS, J. F., J. B. LEWIS, C. A. ANDERSON, AND R. F. GRESTELAND. Enhanced differential synthesis of proteins in a mammalian cell-free system by the addition of polyamines. J. Biol. Chem. 250: 5688-5695,1975. 2. BACHRACH, U., A. KAYE, AND R. CHAYDEN (Editors). Advances in Polyamine Research. New York: Raven, 1983, vol. 4. 3. BRADFORD, M. A. A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the grincinle of

MUCOSAL

REPAIR

G591

dye binding. Anal. Biochem. 72: 224-254,1976. 4. BURTON, K. A. Study of the conditions and mechanism of the diphenylamine reaction for the calorimetric estimation of deoxyribonucleic acid. Biochem. J. 62: 315-323, 1956. 5. CERIOTTI, G. Determination of nucleic acids in animal tissues. J. Biol. Chem. 214: 59-70,1955. 6. CLARKE, D. D., M. J. MYCEK, A. NEIDLE, AND H. WAELSCH. The incorporation of amines into protein. Arch. Biochem. Biophys. 79: 338-354,1959. 7. COHEN, S. S., S. MORGAN, AND E. STREIBEL. The polyamine content of tRNA of E. coli. Proc. Nutl. Acad. Sci. USA 64: 669676,1969. 8. COHEN, S. S. The function of polyamines. Adv. PoZyumine Res. 1: l-10,1978. 9. CRITCHLOW, J., D. MAGEE, S. ITO, K. TAKEUCHI, AND W. SILEN. Requirements for restitution of the surface epithelium of frog stomach after mucosal injury. Gustroenterology 88: 237-249, 1985. diphenylamine method 10. GILES, K. W., AND A. MYERS. An improved for the estimation of deoxyribonucleic acid. Nature Lond. 206: 93, 1965. 11. GOMIS, R., A. SENER, F. MALAISSE-LAGAE, AND W. J. MALAISSE. Transglutaminase activity in pancreatic islets. Biochim. Biophys. Acta 760: 384-388,1983. 12. HADDOX, M. K., AND D. H. RUSSELL. Differential conjugation of polyamines to calf nuclear and nucleolar proteins. J. Cell. Physiol. 109: 447-452,198l. 13. HARTER, J. L. Critical values for Duncan’s new multiple range test. Biometrics 16: 671-685, 1960. 14. IVY, K. J. Acute hemorrhagic gastrities: modern concepts based on the pathogenesis. Gut 12: 750-762,197l. 15. LIPKIN, M., P. SHERLOCK, AND B. BELL. Cell proliferation kinetics in the gastrointestinal tract of man. II. Cell renewal in the stomach, ileum, colon, and rectum. Gczstroenterology 45: 721-729, 1963. 16. LUK, G. D., AND S. B. BAYLIN. Polyamines and intestinal growthincreased polyamine biosynthesis after jejunectomy. Am. J. Physiol. 245 (Gastrointest. Liver Physiol. 8): G656-G660, 1983. 17. MIZUI, T., AND M. DOTEUCHI. Effect of polyamines on acidified ethanol-induced gastric lesions in rats. Jpn. J. Pharmacol. 33: 939945,1983. 18. MIZUI, T., N. SHIMONO, AND M. DOTEUCHI. A possible mechanism of protection by polyamines against gastric damage induced by acidified ethanol in rats: polyamine protection may depend on its antiperoxidative properties. Jpn. J. Pharmacol. 44: 43-50, 1987. 19. PEGG, A. E. Recent advances in the biochemistry of polyamines in eukaryotes. Biochem. J. 243: 249-262, 1986. 20. PEGG, A. E., AND P. P. MCCANN. Polyamine metabolism and function. Am. J. Physiol. 243 (Cell Physiol. 12): C212-C221, 1982. 21. RICHMAN, R. A., L. E. UNDERWOOD, J. J. W. VAN, AND S. J. BOINA. Synergistic effect of cortisol and growth hormone on hepatic ornithine decarboxylase activity. Proc. Sot. Exp. Biol. Med. 138: 880-884,197l. 22. RUTTEN, M. J., AND S. ITO. Morphology and electrophysiology of guinea pig gastric mucosal repair in vitro. Am. J. Physiol. 244 (Gastrointest. Liver Physiol. 7): G171-G183, 1983. 23. RUSSELL, D. H. Ornithine decarboxylase: a key regulatory enzyme in normal and neoplastic growth. Drug Metab. Rev. 16: l-88,1985. 24. SCHUBERT, F., K. HONG, N. DUZGUNES, AND D. PAPAHADJOPULOS. Polyamines as modulators of membrane fusion: aggregation of liposomes. Biochemistry 22: 6134-6140,1983. 25. SEIDEL, E. R., M. K. HADDOX, AND L. R. JOHNSON. Polyamines in the response to intestinal obstruction. Am. J. Physiol. 246 (Gastrointest. Liver Physiol. 9): G649-G653, 1984. 26. SILEN, W. Gastric mucosal defense and repair. In: Physiology of the Gastrointestinal Tract (2nd ed.), edited by L. R. Johnson. New York: Raven, 1987, p. 1055-1069. 27. SILEN, W., AND S. ITO. Mechanism for rapid-epithelialization of the gastric mucosal surface. Annu. Rev. Physiol. 47: 217-229, 1985. 28. TABOR, C. W., AND H. TABOR. Polyamines. Annu. Rev. Biochem. 53: 749-790, 1984. 29. TAKAGI, K., AND S. OKABE. The effects of drugs on the production and recovery processes of the stress ulcer. Jpn. J. Pharmacol. 18: g-18,1968. TAKEUCHI, K., AND L. R. JOHNSON. Pentagastrin protects against stress ulceration in rats. Gustroenterology 76: 327-334, 1979. TAKEUCHI. K., AND L. R. JOHNSON. Effect of cell nroliferation of

Downloaded from www.physiology.org/journal/ajpgi at Glasgow Univ Lib (130.209.006.061) on February 14, 2019.

G592

32.

33.

34. 35.

POLYAMINES

AND

healing of gastric and duodenal ulcers in rats. Digestion 33: 92100, 1986. TAKEUCHI, K., H. NISHIWAKI, H. OSANO, S. EBARA, AND S. OKABE. Repair of mucosal damage induced by ethanol in the rat stomach: effects of concentration, exposure period and prostaglandins. Digestion 40: l-10, 1988. THIRUMALAI, C. H. R., C. C. TSENG, K. TABATA, L. R. FITZPATRICK, AND L. R. JOHNSON. Relationship between ornithine decarboxylase activity and gastric damage. Am. J. Physiol. 253 (Gastrointest. Liver Physiol. 16): Gl-G6, 1987. TSAI, Y.-H., AND S.-H. LIN. Differential release of polyamines by cultured rat sertoli cells. J. Androl. 6: 348-352, 1985. VANAMEE, P., S. J. WINAWER, P. SHERLOCK, M. SONENBERG, AND M. LIPKIN. Decreased incidence of restraint-stress induced gastric erosions in rats treated with bovine growth hormone. Proc. Sot. BioZ. Med. 135: 259-262, 1970.

MUCOSAL

REPAIR

36. WALLACE, J. L., AND B. J. R. WHITTLE. Acceleration of recovery of epithelial integrity by 16,16-dimethyl prostaglandin EZ. Br. J. Pharmacol. 86: 837-842,1986. 37. WANG, J.-Y., AND L. R. JOHNSON. Induction of gastric and duodenal mucosal ornithine decarboxylase during stress. Am. J. Physiol. 257 (Gastrointest. Liver Physiol. 20): G259-G265, 1989. 38. WANG, J.-Y., AND L. R. JOHNSON. Role of ornithine decarboxylase in repair of gastric mucosal stress ulcers. Am. J. Physiol. 258 (Gastrointest. Liver Physiol. 21): G78-G85, 1990. 39. WANG, J.-Y., AND L. R. JOHNSON. Gastric and duodenal mucosal ornithine decarboxylase and damage after corticosterone. Am. J. Physiol. 258 (Gastrointest. Liver Physiol. 21): G942-G950, 1990. 40. WILLIAMS-ASHMAN, H. G. Transglutaminases and the clotting of mammalian seminal fluids. Mol. CeZZ Biochem. 58: 51-61, 1984. 41. YEOMANS, N. D., D. J. B. ST. JOHN, AND W. G. DEBOER. Regeneration of gastric mucosa after aspirin-induced injury to the rat. Am. J. Dig. Dis. 18: 773-780, 1973.

Downloaded from www.physiology.org/journal/ajpgi at Glasgow Univ Lib (130.209.006.061) on February 14, 2019.