The effects of medroxy progesterone acetateon the pro-inflammatory cytokines, TNF-alphaand IL-1beta in the early phase of the spinalcord injury

Berkant Sahin, Baki S. Albayrak, Ozgur Ismailoglu, Askin Gorgulu

Department of Neurosurgery, Suleyman Demirel University, Isparta, Turkey

Objectives: Spinal cord injuries (SCIs) have high morbidity and mortality rates and currently the definitivetreatment of complete SCIs are still not possible. We investigated the effects of the medroxy progesteroneacetate on the pro-inflammatory cytokines, TNF-alpha and IL-1beta in the early phase of the SCI.Methods: Forty-eight Wistar albino male rats were divided equally into four groups each consisting of 12rats. All animals underwent T10–T12 laminectomy. We administered placebo, and 8 mg/kg medroxyprogesterone acetate (MPA) intra-peritoneally into control and progesterone group at 30 minutes after theclip-compression trauma in spinal cord. We performed only T10–T12 laminectomy and clip-compressiontrauma in laminectomy and trauma group, respectively. Half of the rats from each group were killed at1 hour and the other half were killed at 6 hours after the trauma. Spinal cord segments were then removedand stored at 280uC in phosphate buffer. TNF-alpha and IL-1beta levels were determined using ELISA kit.Results: We have found that there was an increase only in the TNF-alpha level at 6 hours after the traumacomparing to control group. MPA appeared to lower the TNF-alpha level significantly in the trauma group.Discussion: This experimentally proven anti-inflammatory effect of MPA via acting upon TNF-alpha mayoffer new therapeutic options in human subjects with SCIs.

Keywords: Medroxy progesterone acetate, TNF-alpha, IL-1beta, Spinal cord injury

IntroductionSpinal cord injuries (SCIs) have high morbidity and

mortality rates and currently the definitive treatment of

complete SCIs are still not possible.1–4 SCIs may result

from traumatic (acute kinetic and static compression,

acceleration–deceleration, distraction, and complete

or partial transition injuries) and non-traumatic

insults including ischemia, tumor compression, and

chemicals.1,5 The extent of the primary injury in spinal

cord trauma is also determined by the mechanism and

the severity of the trauma. Additionally, it was reported

that the regional blood flow and the diameter of the

spinal canal at the involved segment affect the outcome

of the SCIs.6 Furthermore, there are suggested second-

ary mechanisms involved with SCIs including free-

oxygen radicals, calcium theory, opiate receptor theory,

and inflammatory theory.7–10 Based on these theories,

numerous agents, including opiate receptor antagonists,

steroids (methylprednisolone), anti-oxidant agents,

free-radical scavengers, gangliosides, arachidonic acid

modulators, glutamate receptor blockers, calcium

channel antagonists, non-streoidal anti-inflammatory

agents, immunosupressive drugs, and serotonin recep-

tor blockers were tried in the medical treatment of

SCIs.11–15 However, among all these agents, only

methyl prednisolone appeared to have clinical efficacy

in patients with SCI.16–18 Medroxy progesterone acetate

(MPA) is a well-known progesterone agonist, and is a

stimulating factor for cellular differentiation, and there

are experimental trials reporting the neuro-protective

effects of MPA in SCIs in the literature.19–21 However,

currently the extent and the exact mechanism of action

of MPA are not clearly investigated. In our study, we

hypothesized that MPA would exert a neuro-protective

effect by decreasing the post-traumatic inflammatory

reactions by suppressing the pro-inflammatory cyto-

kines, tumor necrosis factor (TNF)-alpha and inter-

leukin (IL)-1beta.

Materials and MethodsThis study was performed at the Animal Laboratory

of the Medical Faculty Hospital of the Suleyman

Demirel University, Isparta, Turkey, after the

Correspondence to: B S Albayrak, Department of Neurosurgery,Suleyman Demirel University, Cunur, Isparta, Turkey. Email: serdarbaki@gmail.com

� W. S. Maney & Son Ltd 2011DOI 10.1179/016164110X12807570510095 Neurological Research 2011 VOL. 33 NO. 1 63

consent of the related ethnic committee. We used 48

Wistar albino rats each weighting between 170 and

361 g. We divided rats equally into 4 main groups as

below:

N Group 1 (n512): laminectomy group;

N Group 2 (n512): laminectomyztrauma group;

N Group 3 (n512): laminectomyztraumazplacebo —control group (0.009% NaCl);

N Group 4 (n512): laminectomyztraumazMPAgroup.

All animals were anesthesized using 60 mg/kg keta-

mine (Ketalar, Parke-Davis, Eczacibasi, Istanbul) and

10 mg/kg xylazine intra-peritoneally. Animals were

then shaved on their dorsal areas and 10% polyvidone

iodine (Batticon, Adeka, Samsun) was applied for

local anti-sepsis. In prone position, throcal (T10–T12)

laminectomy was performed under sterile conditions

and intact dura matters were exposed (Fig. 1).

Aneurysm clips (Sugita no. 07-934-11, closure pres-

sure: 1.37–1.72 N) were then applied horizontally

around the dura matters for 1 minute for the purpose

of creating spinal cord injury (Figs. 2 and 3). The

wound were then closed using 3/0 silk suture

thoroughly. Half of the animal in each group (n56)

were killed at 1 hour and the other half were killed at

6 hours after the trauma and spinal cord segments

were removed carefully and stored at 280uC in

phosphate buffer. Tissue-homogenates were prepared

and TNF-alpha and IL-1beta values were determined.

Removed spinal cord segments in 0.1 M 11% phos-

phate buffer were homogenated in the homogenator

over the ice at 10 000 spins/min for 1 min.

Homogenated samples were then centrifugated at

5000 spins/min for 5 minutes at z4uC. Proteins levels

in supernatants after centrifugation were detected by

using Bio-Rad DC Protein Assay Kit. Rat TNF-alpha

ELISA kit (catalog no. KRC 3011; BioSource,

Wayne, PA, USA) and rat E IL-1beta ELISA kit

(catalog no. KRC 0011; BioSource) were used to

detect TNF-alpha and IL-1beta levels. Measurements

were performed using ELX 808 IU Elisa Plate Reader

(pq/ml) apparatus.

Statistical analysisWe used SPSS 9.0 software. Kruskal–Wallis test was

used to analyse the TNF-alpha, IL-1beta, protein,

TNF-alpha/protein ratio, and IL-1beta/protein ratio

between the groups (P50.05). Groups were com-

pared to each other by applying non-parametric test

(Mann–Whitney U test) since samples’ sizes are less

than 30 and we determined significance level as

P50.01 after Benferroni correction was done.

ResultsWe measured the TNF-alpha, IL-1beta, tissue-

protein levels, TNF-alpha/protein ratio, and IL-

1beta/protein ratios in all groups at 1 and the 6 hours

after the spinal cord trauma (Table 1).

We did not find any significant rise in TNF-alpha/

protein ratio and IL-1beta/protein ratio in trauma

group compared to those in laminectomy group at

1 hour after the trauma, while there was a significant

rise in TNF-alpha/protein ratio in trauma group

compared to that in laminectomy group (P,0.05) at

6 hours after the trauma. This difference was not

significant for IL-1beta/protein ratio.

MPA decreased the IL-1beta/protein levels signifi-

cantly compared to those in trauma group at 1 hour

1 cm

Figure 1 The intact dura was exposed after T10–T12

laminectomy. Figure 2 The view from above after the aneurysm clip was

applied circumferentially around the dura matter.

Table 1 The mean values with standard deviations of TNF-alpha, IL-1beta, tissue-protein levels, TNF-alpha/protein ratio,and IL-1beta/protein ratio in all groups at 1 (a) and 6 (b) hours after the spinal cord trauma

Group (pq/mg) TNF-alpha IL-beta Protein TNF-alpha/protein ratio IL-1beta/protein

Laminectomy (a) 190.1¡38.86 126.6¡43.03 242.5¡53.06 0.80¡0.26 0.50¡0.23Laminectomy (b) 165¡59.68 65.3¡58.67 385¡92.17 0.44¡0.20 0.18¡0.22Trauma (a) 135¡22.17 110.5¡95.51 284¡56.41 0.49¡0.20 0.44¡0.52Trauma (b) 214.2¡42.58 161.9¡36.11 274.5¡36.72 0.76¡0.15 0.55¡0.12Control (a) 184.1¡43.20 123.2¡18.13 231.5¡32.50 0.75¡0.19 0.50¡0.10Control (b) 215.1¡69.72 93.0¡31.36 299.5¡10.23 0.70¡0.22 0.30¡0.10MPA (a) 211.4¡80.58 78.01¡23.92 362¡21.80 0.56¡0.25 0.21¡0.06MPA (b) 150¡30.48 86.2¡40.77 346.5¡45.06 0.47¡0.10 0.25¡0.12Total 176.5¡55.23 92.4¡51.99 296¡68.65 0.63¡0.23 0.37¡0.25

Sahin et al. The effects of MPA on TNF-alfa and IL-1 beta in spinal cord injury

64 Neurological Research 2011 VOL. 33 NO. 1

after the trauma (P50.009). This decrease was not

significant for TNF-alpha/protein ratio.

MPA decreased the TNF-alpha/protein ratio sig-

nificantly compared to that in trauma group at

6 hours after the trauma (P,0.01; Fig. 4).

DiscussionAlthough there are numerous technological assort-

ments in the management of patients with SCIs, the

extent of surgery is still confined with decompression,

stabilization, and rehabilitation. This pessimistic view

has directed the researchers to investigate the

mechanisms of secondary injury of SCIs in animal

models mimicking human subjects. Secondary insult

in SCIs are caused by ischemia, abnormal intracel-

lular ionic shifts including Naz and Ca2z, lipid

peroxidation due to free-oxygen radicals, edema,

leukocyte infiltration, and excito-toxic cell death.8–10

As a result of these events, tissue necrosis with

cavities in the spinal cord develops and neurological

recovery could not be possible due to the failure in

regeneration and glial scar formation.22,23

Inflammation takes place immediately after the SCI

and activated neutrophils, macrophages, and micro-

glias secrete cytotoxic and/or neurotoxic molecules in

the site of injury while scavenging the cellular

debris.24 The source of the secreted pro-inflammatory

cytokines, IL-1beta and TNF-alpha is controversial;

while some authors reported that these cytokines are

the products of neutrophils, and macrophages; there

are studies showing that microglias in the central

nervous system secreted these cytokines.25,26 Yang

et al. have shown that TNF-alpha was found in the

majority of the neurons at 1 hour after severe spinal

trauma; however, the authors could detect very few

TNF-alpha-positive neurons after mild spinal

trauma.27 Similarly, it was also reported that IL-

1beta, IL-6, TNF-alpha levels started to rise at 1 and

3 hours and reached peak at the sixth hour and

started to normalize by the end of the first day

following spinal injury.28 IL-1beta is associated with

neuronal necrosis, apoptosis, leukocyte infiltration,

edema, glial cell activation, and the synthesis of nitric

oxide. Additionally, IL-1beta may cause the stimula-

tion and secretion of IL-6 and basic fibroblast growth

factor from the astrocytes. Additionally, the intracer-

ebral administration of IL-1beta may disrupt the

blood–brain barriers and may results in cerebral

edema and secondary neuronal cell-death. It was also

shown that the rise in the level of IL-1 receptor

antagonist or the blockade of the IL-1-converting

enzyme (caspase) prevented excito-toxic neuronal cell

death.25 Likewise, it was reported that the systemic

injection of IL-1 receptor antagonist has decreased

the neuronal cell death and improved the cognitive

functions in ‘lateral fluid-percussion head trauma

model’.29 TNF-alpha has similar pro-inflammatory

effects to IL-1beta. It induces the release of IL-6 and

growth factor from astrocytes, increases apoptosis,

and activates glial cell proliferation.30 It is also

believed that TNF-alpha increases leukocyte infiltra-

tion by stimulating endothelial adhesion molecules

and surface antigens. On the contrary, it was also

reported that TNF-alpha have neurotrophic effects

by inducing axonal regeneration and increasing the

synthesis of nerve growth factor in astrocytes.31

Currently, the only widely accepted pharmacological

therapy in the treatment of SCI is intravenous (i.v.)

administration of methyl prednisolone (30 mg/kg i.v.

in 15-minute bolus, 5.5 mg/kg/h infusion in 23 hours)

within the first 8 hours of the spinal trauma.16–18

Progesterone receptors are widely distributed in

central nervous system (hypothalamus, preoptic area,

midbrain, cortex, amygdala, hippocampus, caudate,

putamen, and cerebellum). Classically, it is known

that progesterone have interactions with the steroid-

specific cytosolic/nuclear receptors in neuronal cells.32

It was shown that progesterone has neuro-protective

effects when administered at the beginning or 2 hours

after the cerebral ischemia due to middle cerebral

artery occlusion in rats.33 Similarly, Roof et al. have

demonstrated that progesterone have limited the

secondary neuronal loss in the ‘contused animal

model’ and improved the cognitive recovery.34

Figure 3 Hemorrhagic areas and contusions around the

spinal cord after the application of aneurysm clip for

1 minute.

Figure 4 MPA decreased the TNF-alpha/protein ratio sig-

nificantly compared to that in trauma group at 6 hours after

the spinal cord trauma (P,0.01; lam: laminectomy group;

Tra: trauma group; Con: control group; MPA: medrocy

progesterone acetate group).

Sahin et al. The effects of MPA on TNF-alfa and IL-1 beta in spinal cord injury

Neurological Research 2011 VOL. 33 NO. 1 65

Progesterone is synthesized in spinal cord, brain, and

peripheral neurons from pregnenolone or cholesterol

via cytochrome P450 de novo. Pregnenolone is then

converted to progesterone by the enzyme 3-beta-

hydroxysteroid dehydrogenease.35 Macroglial cells,

astrocytes, oligodendrocytes, and Schwann cells have

also the capacity of progesterone synthesis.

Progesterone has a crucial role in the local synthesis

of myelin in neurons.36 It was shown that there was

increase in the muscle strength and neuropathological

changes in the spinal motor neurons in progesterone-

administered animals. Schumacher et al. have first

demonstrated the pro-myelinization effects of pro-

gesterone in mouse-sciatic nerves and sensory neu-

rons and Schwann cell cultures.37 Even though there

are proposed mechanisms of neuro-protective actions

of progesterone, the exact mechanism is not still

known. It was shown that progesterone prevented

glutamate toxicity in cultured spinal cord neurons.

Additionally, recent studies have shown that proges-

terone decreased cerebral edema and increased the

functional recovery. There are also several experi-

mental studies in the literature reporting that

progesterone and allopregnanolone have suppressed

TNF-alpha and IL-1beta. For example, He et al.

have reported that progesterone and allopregnano-

lone have decreased the levels of TNF-alpha and IL-

1beta after traumatic brain injury in rats.38 Similarly,

Pettus et al. have also shown that progesterone has

decreased both mRNA and the protein levels of

TNF-alpha and IL-1beta after traumatic brain injury

in rats.39 However, to the best of our knowledge,

there is no experimental study in the literature

designed to investigate the effects of progesterone

on pro-inflammatory cytokines TNF-alpha and IL-

1beta in spinal cord injury. The clip-compression

model we used in our study was developed by Rivlin

and Tator in 1978.40 In this model, an aneurysm clip

is placed around the spinal cord dura circumferen-

tially after laminectomy is performed. As an advan-

tage, clip closure pressure and compression duration

can be tailored out. Additionally, since compression

is circumferentially applied in this model, it mimics

spinal traumas in human subjects more reliably. In

conclusion, we for the first time were able to clearly

show that MPA has significantly decreased the level

of TNF-alpha, one of pro-inflammatory cytokines

after spinal cord trauma in rats. The verification of

this experimental finding in human subjects by large

prospective clinical trials may offer a new therapeutic

agent, MPA in the treatment of SCIs.

References1 Sanon A, Rengachary SS. The history of spinal biomechanics.

Neurosurgery 1996;39:657–65.2 Naderi S, Zileli M, Ozer AF. The history of spinal surgery. In:

Zileli M, Ozer AF, editors. Spine and spinal cord surgery.Izmir: META Basım; 2002. p. 1–13.

3 Marketos SG, Skiadas PK. Galen: a pioneer of spine research.Spine 1999;24:2358–62.

4 Dohrmann GJ. Experimental spinal cord trauma. A historicalreview. Arch Neurol 1972;7:468–73.

5 Collins WF. A review and update of experiment and clinicalstudies of spinal cord injury. Paraplegia 1983;21:204–19.

6 Osterholm JL. The pathophysiological response to spinal cordinjury. The current status of related research. J Neurosurg1974;40:5–33.

7 Amar AP, Levy ML. Pathogenesis and pharmacologicalstrategies for mitigating secondary damage in acute spinal cordinjury. Neurosurgery 1999;44:1027–39.

8 Demediuk P, Saunders RD, Clendenon NR, Anderson DK,Means ED, Horrocks LA. Changes in lipid metabolism intraumatized spinal cord. Prog Brain Res 1985;63:211–26.

9 Hall ED, Wolf DL. Post-traumatic spinal cord ischemia:relationship to injury severity and physiological parameters.Cent Nerv Syst Trauma 1987;4:15–25.

10 Marx JL. Oxygen free radicals linked to many diseases. Science1987;235:529–31.

11 Kaptanoglu E, Tuncel M, Palaoglu S, Konan A, Demirpence E,Kilinc K. Comparison of the effects of melatonin andmethylprednisolone in experimental spinal cord injury.J Neurosurg 2000;93(Suppl. 1):77–84.

12 Faden AL, Jacops TP. Effect of TRH analogs on neurologicrecovery after experimental spinal trauma. Neurology1985;35:1331–4.

13 Tator CH, Rowed DW, Schwartz ML, Gertzbein SD,Bharatwal N, Barkin M, et al. Management of acute spinalcord injuries. Can J Surg 1984;27:289–93.

14 de la Torre JC. Spinal cord injury. Review of basic and appliedresearch. Spine.1981;6:315–35.

15 Braughler RW, Duncan LA. Interaction of lipid peroxidationand calcium in the pathogenesis of neuronal injury. Cent NervSyst Trauma 1985;2:269–83.

16 Hall ED, McCall JM, Chase RL, Yonkers PA, Braughler JM.A nonglucocorticoid steroid analog of methylprednisoloneduplicates its high-dose pharmacology in models of centralnervous system trauma and neuronal membrane damage.J Pharmacol Exp Ther 1987;242:137–42.

17 Koyanagi I, Tator CH. Effect of a single huge dose ofmethylprednisolone on blood flow, evoked potentials, andhistology after acute spinal cord injury in the rat. Neurol Res1997;19:289–99.

18 Braughler JM, Hall ED. Correlation of methylprednisolonelevels in cat spinal cord with its effects on (Naz z Kz)-ATPase, lipid peroxidation, and alpha motor neuron function.J Neurosurg 1982;56:838–44.

19 Thomas AJ, Nockels RP, Pan HQ, Shaffrey CI, Chopp M.Progesterone is neuroprotective after acute experimental spinalcord trauma in rats. Spine 1999;24:2134–8.

20 Roof RL, Duvdevani R, Heyburn JW, Stein DG. Progesteronerapidly decreases brain edema: treatment delayed up to24 hours is still effective. Exp Neurol 1996;138:246–51.

21 Smith SS, Woodward DJ, Chapin JK. Sex steroids modulatemotor-correlated increases in cerebellar discharge. Brain Res1989;476:307–16.

22 Banik NL, Hogan EL, Hsu CY. Molecular and anatomicalcorrelates of spinal cord injury. Cent Nerv Syst Trauma1985;2:99–107.

23 White BC, Krause GS. Brain injury and repair mechanisms: thepotential for pharmacologic therapy in closed-head trauma.Ann Emerg Med 1993;22:970–9.

24 Katoh S, Ikata T, Tsubo M, Hamada Y, el Masry WS. Possibleimplication of leukocytes in secondary pathological changesafter spinal cord injury. Injury 1997;28:215–7.

25 Holmin S, Mathiesen T. Intracerebral administration ofinterleukin-1beta and induction of inflammation, apoptosis,and vasogenic edema. J Neurosurg 2000;92:108–20.

26 Aloisi F, Care A, Borsellino G, Gallo P, Rosa S, Bassani A,et al. Production of hemolymphopoietic cytokines (IL-6, IL-8,colony-stimulating factors) by normal human astrocytes inresponse to IL-1 beta and tumor necrosis factor-alpha.J Immunol 1992;149:2358–66.

27 Yang L, Blumbergs PC, Jones NR, Manavis J, Sarvestani GT,Ghabriel MN. Early expression and cellular localization ofproinflammatory cytokines interleukin-1beta, interleukin-6,and tumor necrosis factor-alpha in human traumatic spinalcord injury. Spine 2004;29:966–71.

28 Yang L, Jones NR, Blumbergs PC, van Den HC, Moore EJ,Manavis J, et al. Severity-dependent expression of proinflam-matory cytokines in traumatic spinal cord injury in the rat.J Clin Neurosci 2005;12:276–84.

29 Sanderson KL, Raghupathi R, Saatman KE, Martin D, Miller G,McIntosh TK. Interleukin-1 receptor antagonist attenuates

Sahin et al. The effects of MPA on TNF-alfa and IL-1 beta in spinal cord injury

66 Neurological Research 2011 VOL. 33 NO. 1

regional neuronal cell death and cognitive dysfunction afterexperimental brain injury. J Cereb Blood Flow Metab1999;19:1118–25.

30 Klusman I, Schwab ME. Effects of pro-inflammatory cytokinesin experimental spinal cord injury. Brain Res 1997;762:173–84.

31 Ebadi M, Bashir RM, Heidrick ML, Hamada FM, Refaey HE,Hamed A, et al. Neurotrophins and their receptors in nerveinjury and repair. Neurochem Int 1997;30:347–74.

32 Thind KK, Goldsmith PC. Expression of estrogen andprogesterone receptors in glutamate and GABA neurons of thepubertal female monkey hypothalamus. Neuroendocrinology1997;65:314–24.

33 Kumon Y, Kim SC, Tompkins P, Stevens A, Sakaki S, LoftusCM. Neuroprotective effect of postischemic administration ofprogesterone in spontaneously hypertensive rats with focalcerebral ischemia. J Neurosurg 2000;92:848–52.

34 Roof RL, Duvdevani R, Braswell L, Stein DG. Progesteronefacilitates cognitive recovery and reduces secondary neuronalloss caused by cortical contusion injury in male rats. ExpNeurol 1994;129:64–9.

35 Coirini H, Gouezou M, Liere P, Delespierre B, Pianos A,Eychenne B, et al. 3 Beta-hydroxysteroid dehydrogenaseexpression in rat spinal cord. Neuroscience 2002;113:883–91.

36 Schumacher M, Guennoun R, Robert F, Carelli C, Gago N,Ghoumari A, et al. Local synthesis and dual actions ofprogesterone in the nervous system: neuroprotection andmyelination. Growth Horm IGF Res 2004;14:18–33.

37 Schumacher M, Guennoun R, Mercier G, Desarnaud F, Lacor P,Benavides J, et al. Progesterone synthesis and myelinformation in peripheral nerves. Brain Res Brain Res Rev2001;37:343–59.

38 He J, Evans CO, Hoffman SW, Oyesiku NM, Stein DG.Progesterone and allopregnanolone reduce inflammatory cyto-kines after traumatic brain injury. Exp Neurol 2004;189:404–12.

39 Pettus EH, Wright DW, Stein DG, Hoffman SW. Progesteronetreatment inhibits the inflammatory agents that accompanytraumatic brain injury. Brain Res 2005;1049:112–9.

40 Rivlin AS, Tator CH. Regional spinal cord blood flow in ratsafter severe cord trauma. J Neurosurg 1978;49:844–53.

Sahin et al. The effects of MPA on TNF-alfa and IL-1 beta in spinal cord injury

Neurological Research 2011 VOL. 33 NO. 1 67