Search PubMed⌕ Search

Biomedical subjects

J F Dixon

Publications and source records attributed to J F Dixon.

At least 37 records · Page 2Linked to original sources
PubMed1989

Kinetic analysis of the formation of inositol 1:2-cyclic phosphate in carbachol-stimulated pancreatic minilobules. Half is formed by direct phosphodiesteratic cleavage of phosphatidylinositol.

The issue as to whether there is direct phosphodiesteratic cleavage of phosphatidylinositol (PI), in addition to that of phosphatidylinositol 4,5-bisphosphate (PIP2), on agonist stimulation of cells has been controversial. In an attempt to resolve this issue, we have studied the kinetics of the formation and breakdown of the cyclic inositol phosphates. This approach is fairly straightforward, since the turnover of the cyclic inositol phosphates is very slow as compared to that of the noncyclic inositol phosphates and proceeds from inositol 1:2-cyclic 4,5-trisphosphate to inositol 1:2-cyclic phosphate (I(c1:2)P) directly by dephosphorylation without any branching pathways, in contrast to the multiple branchpoints of the noncyclic inositol phosphate pathway. Mouse pancreatic minilobules were prelabeled with [3H]inositol for 30 min, followed by washing to remove free inositol. They were then stimulated with carbachol for 30 min. The inositol cyclic polyphosphates reached steady state at 10-15 min, and I(c1:2)P reached steady state at 25 min. We blocked the action of carbachol by addition of an excess of atropine at 30 min, and the rate of disappearance of the three cyclic inositol phosphates was measured. From these data, the contribution of the inositol cyclic polyphosphate pathway to I(c1:2)P was calculated, which was 40-50% of total I(c1:2)P formation. Thus, 40-50% of the I(c1:2)P formed must have been derived from direct phosphodiesteratic cleavage of PI. This approach should prove useful in measuring the relative contributions of PI hydrolysis and PI phosphorylation (phosphatidylinositol 4,5-bisphosphate hydrolysis) in the overall PI cascade.

Animals↗
PubMed1989

Activation of human T lymphocytes by crosslinking of anti-CD3 monoclonal antibodies.

The proliferation of human T lymphocytes induced by anti-CD3 monoclonal antibodies (mAb) is used as a model for antigen-induced activation via the T cell receptor-CD3 complex. Since both systems are accessory cell (AC)-dependent, an understanding of the role of AC in anti-CD3-induced proliferation may provide an understanding of physiological activation via the T cell receptor. Previous work has implicated receptor crosslinking as an important AC function. To determine its necessity in anti-CD3-induced lymphocyte proliferation, we prepared highly purified T lymphocytes and found that these cells did not respond to the anti-CD3 mAb UCHT1, either alone or with interleukin 1 (IL1), interleukin 2 (IL2), or tetradecanoyl phorbol acetate (TPA). However, the response, as measured by appearance of IL2 receptors and proliferation, was restored by crosslinking with immobilized goat anti-mouse antibodies (GAM) and did not require the addition of IL1, IL2, or TPA. Thus, crosslinking of CD3 receptors was a sufficient signal for proliferation of these cells. Cyclosporine A (CsA) inhibited the activation induced by immobilized UCHT1. Since macrophages are the principle targets of CsA-mediated suppression of mitogen-induced proliferation, but macrophages do not participate in the response to immobilized anti-CD3, this may indicate that CsA was inhibiting crosslinking or a signal generated by it.

Adult↗
PubMed1989

Macrophage modulation of retinal pigment epithelial cell migration and proliferation.

Macrophages are fully differentiated cells that do not synthesize an extracellular matrix and do not contract; they do, however, produce substances that modify the behavior and functions of other cells, particularly those involved in the inflammatory and immune responses. Since macrophages are a ubiquitous component of periretinal membranes, we sought to determine whether they modulate proliferation and/or migration of retinal pigment epithelial (RPE) cells, functions that may be essential for the development of proliferative vitreoretinopathy (PVR). Using an in vitro assay, we found that macrophage supernatant contains factors that stimulate proliferation and migration of cultured human RPE cells. Since interleukin-1 (IL-1) is a product of activated macrophages that modulates a number of cellular functions, we also examined its effect on RPE proliferation and migration. We found that IL-1 increased migration but did not affect proliferation, and thus could not duplicate the effect of macrophage supernatant. Injection of activated macrophages into the vitreous of rabbits which had a retinal hole stimulated RPE cell proliferation in the area of the retinal hole, where the RPE cells were exposed. These findings suggest the ability of macrophages to modulate functions of RPE cells that are thought to be critical for the development of PVR. Macrophages may thus be an important part of the vitreous environment that favors the development of PVR.

Animals↗
PubMed1989

Anti-CD4 monoclonal antibodies enhance phorbol 12-myristate, 13-acetate-induced activation of human T cells.

Evidence exists which indicates that the T cell differentiation molecule CD4 may interact with nonpolymorphic determinants of major histocompatibility complex (MHC) class II antigens on accessory cells to stabilize the formation of a ternary complex formed by the T cell receptor (CD3-TcR), antigen, and MHC class II restriction element. However, there is also evidence which suggests alternative or additional functional roles of CD4 in the delivery of signals to T cells independent of MHC class II recognition. In the present study, we examined different anti-CD4 monoclonal antibodies (mAbs) for their ability to influence lymphocyte proliferation induced by phorbol 12-myristate, 13-acetate (PMA). We found that the response of human peripheral blood mononuclear cells to PMA could be enhanced by some anti-CD4 mAbs (OKT4, OKT4A) but not by others (G17-2). This enhancement was due neither to a direct action of the mAbs on the monocytes nor to intercellular crosslinking through an Fc-Fc receptor interaction. We also found that the binding of anti-CD did not influence the down-regulation of CD4 expression induced by PMA, ruling out any correlation between increased stimulation and CD4 modulation. Our results, taken together with those recently published on the ability of a soluble anti-CD4 mAb (B66) to induce lymphocyte activation by itself, provide evidence that CD4 antigen plays a positive functional role in T cell stimulation in addition to stabilizing the antigen-antigen receptor interaction.

Antibodies, Monoclonal↗
PubMed1987

Inositol 1,2-cyclic 4,5-trisphosphate is formed in the rat parotid gland on muscarinic stimulation.

Hawkins et al. [Hawkins, P.T., Berrie, C.P., Morris, A.J., and Downes, C.P. (1987) Biochem J. 243, 211-218] were unable to find any formation of inositol 1,2-cyclic 4,5-trisphosphate on muscarinic stimulation of rat parotid slices, contrary to what has been found in mouse pancreas and in platelets. We have repeated the studies of Hawkins et al. using [3H]inositol-prelabelled rat parotid minilobules and our improved HPLC method for clearly separating the three inositol trisphosphates. Substantial amounts of inositol 1,2-cyclic 4,5-trisphosphate formed on muscarinic stimulation of rat parotid minilobules, amounting to 5% of inositol 1,4,5-trisphosphate at 10 sec and one third of inositol 1,4,5-trisphosphate at 5 min.

Animals↗
PubMed1987

Inositol 1,2-cyclic 4,5-trisphosphate concentration relative to inositol 1,4,5-trisphosphate in pancreatic minilobules on stimulation with carbamylcholine in the absence of lithium. Possible role as a second messenger in long- but not short-term responses.

A method for the extraction of cyclic inositol phosphates in the absence of acid after short incubation times is described. A modified high pressure liquid chromatography method is also described which separates inositol 1,2-cyclic 4,5-trisphosphate (IcP3), inositol 1,4,5-trisphosphate (I(1,4,5)P3), and inositol 1,3,4-trisphosphate (I(1,3,4)P3). Mouse pancreatic minilobules were preincubated with [3H]inositol for 1 h in the absence of lithium, washed, and incubated without and with carbamylcholine without lithium for various times. On adding carbamylcholine, I(1,4,5)P3 peaked at 10 s, followed by a fall to a steady-state level which was two-thirds the peak value. This level was maintained for 20 min. IcP3, on the other hand, rose very slowly; at 10 s, it was only 5% of I(1,4,5)P3. It continued to rise until it equaled the steady-state level of I(1,4,5)P3 at 20 min. I(1,3,4)P3 rose gradually but at a faster rate than IcP3, peaking at 40 s at the same level as that achieved by I(1,4,5)P3 at that time and then falling in parallel with I(1,4,5)P3. Assuming equal potencies of IcP3 and I(1,4,5)P3 in mobilizing intracellular stores of Ca2+ in pancreatic minilobules, as seen in Limulus photoreceptor cells, platelets, and 3T3 cells, IcP3 would appear to play no messenger role at very early times. Thus, I(1,4,5)P3 should be responsible for almost all of Ca2+ release at very early times (10 s), while at later times (20 min) IcP3 and I(1,4,5)P3 should contribute equally to Ca2+ release. The significance of these results is discussed.

Animals↗
PubMed1987

The formation of inositol 1,2-cyclic 4,5-trisphosphate and inositol 1,2-cyclic 4-bisphosphate on stimulation of mouse pancreatic minilobules with carbamylcholine.

When [3H]myoinositol-prelabeled pancreatic minilobules were incubated with carbamylcholine (CCh) for 30 min, followed by ionophoresis on paper of the aqueous extracts, there were distinct peaks of radioactivity immediately preceding inositol 1,4-bisphosphate (IP2) and inositol 1,4,5-trisphosphate (IP3), which, based on earlier studies with inositol 1,2-cyclic phosphate (IcP), are the expected positions for inositol 1,2-cyclic 4-bisphosphate (IcP2) and inositol 1,2-cylic 4,5-trisphosphate (IcP3). These peaks were essentially absent on ionopherograms of extracts from minilobules not incubated with CCh. Similar results were obtained with high performance liquid chromatography (HPLC), except that the putative inositol cyclic phosphate peaks eluted immediately before the non-cyclic inositol polyphosphates, as to be expected. Taking advantage of the unique acid lability of the inositol cyclic phosphates, we demonstrate that the putative inositol cyclic polyphosphate peaks were specifically eliminated by prior hydrolysis of the aqueous extracts, as shown by either ionophoresis or HPLC. After preparative isolation of putative IcP2 and IcP3 by ionophoresis, acid hydrolysis shifted the positions of putative IcP2 and IcP3 peaks to the positions of standard IP2 and IP3, respectively, as shown by either ionophoresis or HPLC. The amounts of IcP, IcP2, and IcP3 formed on CCh stimulation, as measured by ionophoresis, were 0.7, 6.8, and 29.8% of that of, IP, IP2, and IP3, respectively (average of two experiments which agreed within 10%).

Animals↗
PubMed1987

Biochemical aspects of the phosphoinositide signalling system with special reference to the formation of inositol cyclic phosphates and arachidonic acid and metabolites on agonist stimulation.

Several aspects of the phosphoinositide signalling system recently studied in our Laboratory are considered here. 1. The formation of inositol 1:2-cyclic-4,5-trisphosphate (IcP3) and inositol 1:2-cyclic-4-bisphosphate (IcP2) have been shown here to occur in pancreatic minilobules stimulated with carbamylcholine. Identification is based on mobility on ionophoresis on paper and on HPLC, acid lability, and conversion of the inositol cyclic phosphates to their respective non-cyclic inositol phosphates on treatment with acid. The levels of inositol 1:2-cyclic phosphate (IcP), IcP2, and IcP3 were 0.7%, 6.8%, and 29.8% of their respective non-cyclic inositol phosphates. The level of IcP3 is sufficient to evoke release of calcium from the endoplasmic reticulum. 2. In a previous study, we demonstrated that on agonist stimulation of pancreatic minilobules prelabelled with [14C]arachidonate, [14C]stearate, or [3H]glycerol, there was a substantial release of all three of these compounds, amounting to approximately 50% of the total PI loss, which was up to 70% of the total cellular PI (7). It was shown that this loss in PI was due to the sequential actions of phospholipase C and diacylglycerol (DG) lipase. Evidence against the phospholipase A2 pathway was no formation of lysophosphatidylinositol. Further evidence against the phospholipase A2 pathway shown here is the lack of stimulation by agonist of glycerophosphorylinositol formation. We also show here that the stimulation of PI loss in guinea pig brain cortex slices is likely also to be via the sequential actions of phospholipase C and DG-lipase, i.e., there was an increase in the steady-state level of monoacylglycerol and a rise in free arachidonate on stimulation with acetylcholine. The formation of prostaglandin E and prostaglandin F was also increased in brain cortex, corpus striatum, and hippocampus. The effects of acetylcholine were abolished by atropine. 3. Previous studies showed that the DG-lipase inhibitor, RHC 80267, inhibited agonist-stimulated formation of glycerol and fatty acids and raised the steady-state level of DG (7). We have now used RHC 80267 as a tool to elevate the level of DG and to lower the level of arachidonate to see if either of these products might modulate the carbamylcholine-stimulated cGMP levels in pancreatic minilobules. RHC 80267 inhibited formation of cGMP. Addition of arachidonate did not affect this inhibition, nor did addition of free arachidonate to control minilobules have any effect, thus suggesting that liberation of free arachidonate by carbamylcholine was not responsible for the carbamylcholine-induced rise in cGMP.(ABSTRACT TRUNCATED AT 400 WORDS)

Acetylcholine↗
PubMed1987

Highly purified human T-lymphocytes do not respond to mitogens--including calcium ionophore and phorbol ester.

The proliferative, and other, in vitro responses of human lymphocytes are dependent upon interactions with accessory cells. Studies of the nature of these required interactions are dependent upon the ability to isolate functionally pure populations of lymphocytes and accessory cells. We describe here a four-step separation procedure using plastic adherence, passage through columns of nylon wool and Sephadex G-10, and treatment with L-leucine methyl ester which produced T-lymphocyte populations which were functionally free of accessory cells, and accessory cell populations which were highly enriched for macrophages. T-lymphocytes were 91-98% Leu-4-positive, contained fewer than 0.05% Mo1- or Mo2-positive cells and fewer than 0.1% nonspecific esterase-positive cells, showed relatively little change in the T4/T8 ratio, and gave no proliferative response to a panel of mitogens. The accessory cell populations were 72-98% nonspecific esterase-positive, 91-99% OKM-1-positive and only 0-4% Leu-4-positive, and could fully restore the proliferative responses of the mitogen-treated lymphocytes. Using these purified cell populations, we have found that, contrary to a number of published observations, both the calcium ionophore A23187 and the phorbol ester 12-O-tetradecanoyl-phorbol-13-acetate were dependent on the presence of accessory cells to produce a mitogenic response by human lymphocytes.

Adult↗
PubMed1986

Effect of cyclosporine A in erythema nodosum leprosum.

Erythema nodosum leprosum (ENL) is a reactional state of lepromatous leprosy in which the loss of suppressor cell function, decrease in suppressor cell numbers, and increase of interleukin 2 production are observed. We reasoned that cyclosporine A (CsA), by opposing these immune responses, could suppress the ENL reaction and restore patients to the quiescent lepromatous state. We tested this hypothesis in vitro by measuring the effect of CsA on M. leprae-triggered suppressor cells. In 24 of 25 patients with ENL, suppressor cell activity was restored by CsA. The target of CsA appeared to be macrophages. These findings are significant in that they provide the first evidence for the potential efficacy of CsA in the treatment of ENL. Preliminary clinical trials indicate a beneficial therapeutic effect associated with increased T suppressor cells in lesions.

Cyclosporins↗
PubMed1985

The formation of inositol 1,2-cyclic phosphate on agonist stimulation of phosphoinositide breakdown in mouse pancreatic minilobules. Evidence for direct phosphodiesteratic cleavage of phosphatidylinositol.

It is generally thought that formation of inositol 1,2-cyclic phosphate (IcP) on agonist-stimulated "breakdown" of endogenous phosphatidylinositol in intact cells would provide strong evidence for the direct phosphodiesteratic cleavage of phosphatidylinositol. We report here that on ionophoresis of extracts of pancreatic minilobules incubated with the cholecystokinin/pancreozymin congener, caerulein, the usual inositol phosphates, i.e. inositol 1-phosphate (IP), inositol 4,5-bisphosphate (IP2), and inositol 1,4,5-trisphosphate (IP3) were seen. In addition, an [3H]inositol-labeled unknown was present with the correct electrophoretic mobility of IcP. There was only a trace of "IcP" in the unstimulated pancreatic minilobules. Several lines of evidence indicate that the unknown peak was IcP. 1) It ran on ionophoresis with standard [14C]IcP, and the ratio of 3H to 14C for each point on the peak was a constant within experimental error. 2) The putative IcP peak which had been eluted from the electropherogram also coincided with standard [14C]IcP on paper chromatography. 3) On mild acid hydrolysis in the presence of standard 14C-labeled IP, the putative [3H] IcP peak disappeared and appeared in the exact position of the standard [14C]IP peak, as to be predicted of IcP. The formation of IcP on agonist stimulation supports direct phosphodiesteratic cleavage of phosphatidylinositol on stimulation of phosphoinositide breakdown in pancreatic minilobules.

Animals↗
PubMed1985

Evidence for glucose-responsive and -unresponsive pools of phospholipid in pancreatic islets.

The effect of glucose on the metabolism of phospholipids in pancreatic islets was studied with three radioactive phospholipid precursors, [32P]orthophosphate, [3H]myoinositol, and [3H]arachidonic acid, to determine the conditions necessary for studying the breakdown of prelabeled phospholipids. Islets were incubated in the presence of a radioactive precursor for 60 or 90 min and in the presence of either 3.3 or 16.7 mM glucose to prelabel phospholipids. To study the breakdown of prelabeled phospholipid, the unincorporated precursor was removed and the islets were reincubated for 15 or 20 min under conditions that either did or did not stimulate insulin release. Prelabeling in the presence of a noninsulinotropic concentration of glucose (3.3 mM) supported the incorporation of precursors into almost all islet phospholipids studied. Prelabeling in an insulinotropic concentration of glucose (16.7 mM) increased the incorporation of precursors into a number of phospholipids even more; and reincubation in 16.7 mM glucose caused a rapid loss of radioactivity from specific phospholipids (phosphatidylinositol and/or phosphatidylcholine, depending on the precursor). This breakdown was observed only when islets had been prelabeled in 16.7 mM glucose. The amount of radioactivity lost from phospholipid corresponded roughly to the additional amount incorporated during the prelabeling in the high concentration of glucose. Radioactivity in phospholipids in islets prelabeled in 3.3 mM glucose or in nonsecretagogue metabolic fuels, such as malate plus pyruvate, did not decrease when the islets were subsequently exposed to 16.7 mM glucose, nor did it decrease in 3.3 mM glucose when these islets had been prelabeled in 16.7 mM glucose. Glyceraldehyde, an insulin secretagogue, but not galactose or L-glucose which are not insulin secretagogues, stimulated phospholipid breakdown in islets that had been prelabeled in 16.7 mM glucose. Depriving islets of extracellular calcium, a condition that inhibits insulin release, inhibited phospholipid breakdown. The results suggest that pancreatic islets contain a glucose-responsive and a glucose-unresponsive phospholipid pool. The glucose-responsive pool becomes labeled and undergoes rapid turnover only under stimulatory conditions and may play a role in the stimulus-secretion coupling of insulin release.

Animals↗
PubMed1984

Secretogogue-stimulated phosphatidylinositol breakdown in the exocrine pancreas liberates arachidonic acid, stearic acid, and glycerol by sequential actions of phospholipase C and diglyceride lipase.

When mouse pancreatic "minilobules" prelabeled with either [14C]arachidonic acid (AA), [14C]stearic acid (SA), or [3H]glycerol were stimulated with the secretogogue, caerulein, there was a 60-70% loss in radioactivity in phosphatidylinositol (PI) at 30 min. This loss was accompanied by the formation of [14C] phosphatidic acid (PA), [14C]diacylglycerol (DG), [14C] triacylglycerol (TG), and free [14C]AA, [14C]SA, and [3H]glycerol. The loss in radioactive PI was the same as the loss in chemically measured PI-phosphorus. Thirty to fifty per cent of the caerulein-induced loss of prelabeled PI could be accounted for as free [14C]AA, [14C]SA, or [3H]glycerol. Increased incorporation of fatty acid or glycerol residues into DG, PA, and TG accounted for the balance of the loss in PI. The specific DG-lipase inhibitor, RHC 80267, markedly inhibited the caerulein-stimulated release of [14C]AA, [14C]SA, and [3H]glycerol and roughly doubled the caerulein-induced increment in [14C]AA-, [14C]SA-, or [3H]glycerol-labeled DG, showing that the source of the caerulein-induced increment in fatty acids and glycerol was DG. When the PI was prelabeled with either [32P] orthophosphate, [3H]myoinositol, or [3H]glycerol, only 1% or less of the radioactivity in PI was in lysophosphatidylinositol (LPI), and there was no increase in radioactivity in LPI on stimulation with caerulein. These observations, taken together, argue strongly for a phospholipase C-catalyzed breakdown of PI followed by DG-lipase and argue against any significant involvement of phospholipase A2 in PI degradation in mouse pancreas. The formation of substantial amounts of free [14C]AA on stimulation supports the view that, among other things, the phosphoinositide effect in the exocrine pancreas serves to generate arachidonate (and its metabolites). The release of appreciable amounts of free fatty acids and glycerol shows that a significant portion of the DG formed as a result of caerulein-stimulated PI breakdown is not conserved in the phosphoinositide cycle.

Animals↗
PubMed1983

Low-density mononuclear cells. Potent stimulators of the human MLR.

We have applied purification strategies similar to those used to purify murine dendritic cells to human peripheral blood in an attempt to enrich for stimulators of the human mixed lymphocyte reaction (MLR). Equilibrium density centrifugation of peripheral blood mononuclear cells yields a population of low density cells that are potent stimulators of a human MLR. The stimulator cells are Dr+, nonlymphocytic, and weakly adherent. Strongly adherent monocytes, also present in the low-density cell population, do not stimulate a human MLR. This contrasts with other human MLR studies that ascribe stimulatory activity to adherent monocytes, and it indicates functional and morphological heterogeneity among monocytes.

Cell Separation↗
PubMed1982

Ambulatory blood pressure monitoring in diabetic males: a method for detecting blood pressure elevations undisclosed by conventional methods.

Twenty-one diabetic males, ages 20 to 61 years (mean 48.4 +/- 10.5) and 14 healthy males, 22 to 59 years (mean 42.5 +/- 10.4) consented to participate in a study during which 24-h recordings of systolic and diastolic blood pressures, heart rates, and rhythm were obtained. The diabetic subjects were considered normotensive except for two patients, who had been observed to have rare insignificant elevations in blood pressure and were untreated. Diabetic subjects had a higher mean maximal systolic blood pressure (160.7 +/- 49.8 mmHg) than the normal controls (132.4 +/- 12.1 mmHg) (p less than 0.05). They also had a higher frequency (15.1%) of systolic blood pressure readings greater than or equal to 150 mmHg compared to normal men (0.2%) (p less than 0.01). The frequency of diastolic blood pressure readings greater than or equal to 90 mmHg was 22.1% in the former and 9.2% in the latter group (p less than 0.01). Of 21 diabetic males, 14 (66.7%) had systolic pressures greater than or equal to 150 mmHg, diastolic pressures greater than or equal to 100 mmHg, or both. In the normal men, 3 (21.4%) of the 14 had such pressures. Twenty-four hour monitoring of ambulatory blood pressures revealed elevations not detected by routine casual readings in patients with diabetes.

Adult↗
PubMed1982

Prostaglandin E2 derived from phosphatidylinositol breakdown in the exocrine pancreas facilitates secretion by an action on the ducts.

The observation that stimulation of a variety of endocrine and exocrine glands, synaptic tissue and certain other structures with appropriate agonists is associated with a breakdown of phosphatidylinositol (PI) and a synthesis of phosphatidic acid (accompanied by increased turnover) has long been known. Our laboratory recently showed that arachidonate released from PI breakdown forms prostaglandin (PG) E2 and the latter is involved in secretagogue-stimulated secretion from the exocrine pancreas. The data presented here are consistent with the view that there is a "pool" of enzymes in the ducts that is released by PGE2, whether added exogenously or generated by carbamylcholine-stimulated PI breakdown and that this release is probably brought about by relaxation of the ampulla of Vater and possibly effects on smooth muscle of the ducts. This hypothesis is supported by the following observations. Involvement of PGs (inhibition of carbamylcholine-stimulated secretion by indomethacin or phenylbutazone) can be demonstrated only in the intact mouse pancreas and not in sliced pancreas or acinar cells, even though PI breakdown can be demonstrated in the cells. The smooth muscle relaxants, nitroglycerin and sodium nitrite, increase amylase release. Secretin, which stimulates fluid secretion from the ducts (acini in mouse and rat) but not enzyme secretion from the acini, increases secretion to the same extent as PGE2. Secretin is believed to increase enzyme secretion by a "washout" effect in the ducts. PGE2 gave no further increase in enzyme secretion in the presence of secretin. The effects of none of these agents (smooth muscle relaxants, secretin or PGE2) are additive.

Animals↗