Robust axonal growth and a blunted macrophage response are associated with impaired functional recovery after spinal cord injury in the MRL/MpJ mouse.

Spinal cord injury (SCI) in mammals leads to a robust inflammatory response followed by the formation of a glial and connective tissue scar that comprises a barrier to axonal regeneration. The inbred MRL/MpJ mouse strain exhibits reduced inflammation after peripheral injury and shows true regeneration without tissue scar formation following an ear punch wound. We hypothesized that following SCI, the unique genetic wound healing traits of this strain would result in reduced glial and connective tissue scar formation, increased axonal growth, and improved functional recovery. Adult MRL/MpJ and C57BL/6J mice were subjected to a mid-thoracic spinal contusion and the distribution of axon profiles and selected cellular and extracellular matrix components was compared at 1, 2, 4 and 6 weeks post-injury. Recovery of hind-limb locomotor function was assessed over the same time period. The MRL/MpJ mice exhibited robust axon growth within the lesion, beginning at 4 weeks post-injury. This growth was accompanied by reduced macrophage staining at 1, 2, 4 and 6 weeks post-injury, decreased chondroitin sulfate proteoglycan staining at 1-2 weeks and increased laminin staining throughout the lesion at 2-6 weeks post-injury. Paradoxically, the extent of locomotor recovery was impaired in the MRL/MpJ mice. Close examination of the chronic lesion site revealed evidence of ongoing degeneration both within and surrounding the lesion site. Thus, the regenerative genetic wound healing traits of the MRL/MpJ mice contribute to the evolution of a lesion environment that supports enhanced axon growth after SCI. However, this response occurs at the expense of meaningful functional recovery.

Proliferation of NG2-positive cells and altered oligodendrocyte numbers in the contused rat spinal cord.

Given the numerous reparative roles glia may play after spinal cord injury (SCI), glial proliferation and cell number were examined in a model of traumatic SCI. Emphasis was placed on analysis of oligodendrocytes and NG2-positive (NG2+) cells, an endogenous cell population that may be involved in oligodendrocyte replacement. Overall, proliferation (assessed by bromodeoxyuridine incorporation) was markedly elevated during the first 2 weeks after injury and declined thereafter; a large portion of these dividing cells likely consisted of microglia-macrophages. Although the total number of NG2+ cells in the epicenter was reduced by half, we noted protracted proliferation in surviving NG2+ cells, with values sevenfold greater than in uninjured controls. Elevated proliferation of NG2+ cells persisted throughout the first 4 weeks after injury. However, the absolute number of NG2+ cells was not increased over controls, suggesting that the daughter cells either did not survive or they differentiated into other cell types. As expected, oligodendrocyte numbers were drastically altered after SCI. By 7 d after injury, the number of oligodendrocytes at the impact site was reduced by 93%. Despite ongoing tissue loss, the number of oligodendrocytes in spared tissue rose threefold at 14 d after injury. Although the function of NG2+ cells within the spinal cord is not completely understood, several studies suggest that they may differentiate into oligodendrocytes. Thus, proliferating NG2+ cells may contribute to the increased oligodendrocyte number observed at 2 weeks after injury. Future studies are required, however, to definitively determine the role NG2+ cells play in oligodendrocyte genesis, remyelination, and other post-injury events.

Gangliosides inhibit platelet-derived growth-factor-stimulated increases in intracellular calcium in Swiss 3T3 cells.

Individual Swiss 3T3 cells stimulated by platelet-derived growth factor delivered by means of a picopump device respond with a brisk, large, and sustained increase in intracellular calcium concentration ([Ca2+]i). Preincubation of cells with either GM1 or GT1b gangliosides inhibited the proportion of responding cells and caused a dose-related diminution in the magnitude of the increase in [Ca2+]i. This effect of ganglioside is probably part of the mechanism through which gangliosides exert their biological effects, including inhibition of platelet-derived growth-factor-induced mitogenesis.

Determination of Fura-2 dissociation constants following adjustment of the apparent Ca-EGTA association constant for temperature and ionic strength.

The accurate calibration of Fura-2 fluorescence in living cells is dependent upon the apparent dissociation constant (Kd) of Fura-2 for Ca2+. If Ca-EGTA calibration buffers are used to construct an in vitro calibration curve, then the calculated value of the apparent Ca-EGTA association constant (K'CaEGTA) will have an important influence on the Kd of Fura-2 and thus the calculated free [Ca2+] in cells. In order to simulate experimental conditions, the individual proton and Ca2+ association constants for EGTA in these experiments were adjusted for both ionic strength and temperature using a semi-empirical form of the Debye-Huckel limiting law and the Van't Hoff isochore, respectively, as described by Harrison and Bers. The modified individual binding constants were then employed in the calculation of K'CaEGTA using the SPECS computer program of Fabatio. At pH = 7.05, ionic strength = 0.15 M, temp = 20 degrees C, K'CaEGTA = 3.232 x 10(6) M-1; at pH = 6.84, temp = 36 degrees C, K'CaEGTA = 1.652 x 10(6) M-1. These values differed substantially from those obtained with unadjusted individual association constants. Calibration buffers of varying [Ca2+] were prepared using the corrected values of K'CaEGTA, and Fura-2 fluorescence ratios were measured during superfusion of these buffers in the experimental chamber at both 20 degrees C and 37 degrees C. The Kd of Fura-2 for Ca2+ was determined to be 236 nM at 20 degrees C and 285 nM at 37 degrees C, utilizing the value of K'CaEGTA adjusted by the method of Harrison and Bers.(ABSTRACT TRUNCATED AT 250 WORDS)

Activation of the phosphatidylinositol pathway in the primate corpus luteum by prostaglandin F2 alpha.

The current study was designed to investigate the ability of prostaglandin F2 alpha (PGF2 alpha) to activate a second messenger system (phosphatidylinositol pathway) in corpora lutea (CL) of rhesus monkeys. Activation of this pathway was assessed by monitoring the hydrolysis of phosphatidylinositol to inositol phosphates. Since inositol triphosphate mobilizes intracellular Ca2+, intracellular free calcium concentrations ([Ca2+]i) were also assessed in individual cells by fura-2 fluorescence photometry. These responses to PGF2 alpha were measured in luteal cells collected from nonpregnant rhesus monkeys. CL were collected during the early (days 4-5 after estimated LH surge; n = 4), mid (days 8-9; n = 4), and late (days 13-14; n = 5) luteal phase and 1 day after in vivo hCG treatment (15 IU/dose, morning and evening), which began during the midluteal phase (n = 5). PGF2 alpha significantly increased the accumulation of inositol phosphates in all groups (P less than 0.05), except the midluteal phase (P = 0.07). The luteal sensitivity to PGF2 alpha, judged by phosphatidylinositol hydrolysis, was low in the early to midluteal phase compared to that in the late luteal phase and after in vivo hCG treatment. PGF2 alpha also caused a rapid, yet transient, increase in [Ca2+]i in a large proportion of primate luteal cells. The proportion of luteal cells that responded to PGF2 alpha with an increase in [Ca2+]i was smaller (P less than 0.05) in CL collected during the early luteal phase than in the other groups. Luteal progesterone production was inhibited by PGF2 alpha in CL collected after in vivo hCG. CL treated in vivo with hCG also displayed in vitro the largest increases in phosphatidylinositol hydrolysis and [Ca2+]i in response to PGF2 alpha. Therefore, this study demonstrates that PGF2 alpha is a potent activator of the phosphatidylinositol pathway in the primate CL. This activation is augmented as the luteal phase progresses and is influenced by in vivo hCG treatment. This study also provides evidence that the inhibitory effects of PGF2 alpha on progesterone production are associated with the activation of the phosphatidylinositol pathway.

Cellular inflammatory response after spinal cord injury in Sprague-Dawley and Lewis rats.

The distribution of microglia, macrophages, T-lymphocytes, and astrocytes was characterized throughout a spinal contusion lesion in Sprague-Dawley and Lewis rats by using immunohistochemistry. The morphology, spatial localization, and activation state of these inflammatory cells were described both qualitatively and quantitatively at 12 hours, 3, 7, 14, and 28 days after injury. By use of OX42 and ED1 antibodies, peak microglial activation was observed within the lesion epicenter of both rat strains between three and seven days post-injury preceding the bulk of monocyte influx and macrophage activation (seven days). Rostral and caudal to the injury site, microglial activation plateaued between two and four weeks post-injury in the dorsal and lateral funiculi as indicated by morphological transformation and the de-novo expression of major histocompatibility class II (MHC II) molecules. Similar to the timing of microglial reactions, T-lymphocytes maximally infiltrated the lesion epicenter between three and seven days post-injury. Reactive astrocytes, while present in the acute lesion, were more prominent at later survival times (7-28 days). These cells were interspersed with activated microglia but appeared to surround and enclose tissue sites occupied by reactive microglia and phagocytic macrophages. Thus, trauma-induced central nervous system (CNS) inflammation, regardless of strain, occurs rapidly at the site of injury and involves the activation of resident and recruited immune cells. In regions rostral or caudal to the epicenter, prolonged activation of inflammatory cells occurs preferentially in white matter and primarily consists of activated microglia and astrocytes. Differences were observed in the magnitude and duration of macrophage activation between Sprague-Dawley (SD) and Lewis (LEW) rats throughout the lesion. Increased expression of complement type 3 receptors (OX42) and macrophage-activation antigens (ED1) persisted for longer times in LEW rats while expression of MHC class II molecules was attenuated in LEW compared to SD rats at all times examined. Variations in the onset and duration of T-lymphocyte infiltration also were observed between strains with twice as many T-cells present in the lesion epicenter of Lewis rats by 3 days post-injury. These strain-specific findings potentially represent differences in corticosteroid regulation of immunity and may help predict a range of functional neurologic consequences affected by neuroimmune interactions.

Quantitative analysis of vascularization and cytochrome oxidase following fetal transplantation in the contused rat spinal cord.

In the normal adult central nervous system, a coupling between energy consumption and vascular density is well established. Likewise, the survival of fetal neural tissue grafts is highly dependent on the establishment of functional vascular integration with the host. However, to what degree graft vascularization and tissue metabolism influence the normal host response to traumatic injury has not been extensively studied. In the present report, embryonic day 14 fetal spinal cord suspension grafts were made into the lesion epicenter of subchronic (10 days) contusion-injured rats. Three months later, intraspinal transplants were analyzed using correlative cytochrome oxidase histochemistry and vascular morphometric analysis. The same approaches were applied to the host spinal cord and injured, non-transplanted animals in order to determine the ability of a graft to alter the level of post-injury vascularization and/or metabolism. In general, graft vascular density was increased over that measured in normal or injured gray matter. Vascular density in gray matter near the host/graft interface was markedly increased when compared to either gray matter of the same spinal level in injured non-grafted animals or normal control spinal gray matter. Vascular changes were not noted in gray matter 3 mm distal to the lesion epicenter (rostral or caudal) in all groups analyzed. Cytochrome oxidase was up-regulated at this time in the graft and gray matter at the host/graft interfaces when compared to either gray matter of the same spinal level in injured, non-grafted animals or that of uninjured controls. These data indicate that an intraspinal transplant placed into the contused adult rat spinal cord reaches a metabolic capacity that is likely to be associated with high levels of oxidative metabolism in the well-vascularized graft neuropil. In addition, transplantation chronically alters vascularization and metabolic patterns of adjacent spinal gray matter following contusion injury.

Assessment of the mechanism by which prolactin stimulates progesterone production by early corpora lutea of pigs.

Previously, we reported that administration of prolactin (PRL) during the early luteal phase in sows increases plasma progesterone concentrations. In the current study, we searched for the mechanisms by which PRL exerts this luteotrophic effect. The objectives of the study were (1) to examine the effect of PRL and/or low-density lipoproteins (LDL) on progesterone production by porcine luteal cells derived from early corpora lutea, and (2) to assess the ability of PRL to activate phosphoinositide-specific phospholipase C (PI-PLC) and protein kinase C (PKC) in these luteal cells. Ovaries with early corpora lutea (day 1-2 of the oestrous cycle) were obtained from the slaughterhouse. Progesterone production by dispersed luteal cells was measured after treatment with PRL, phorbol 12-myristate 13-acetate or inhibitors of PKC in the presence or absence of LDL. LDL increased progesterone concentration in the incubation medium (304.5 vs 178.6 ng/ml in control, P<0.05). PRL augmented LDL-stimulated progesterone secretion by luteal cells (to 416 ng/ml, P<0.05), but PRL alone did not affect progesterone production (209.6 ng/ml, P>0.05). Staurosporine, a PKC inhibitor, inhibited progesterone secretion stimulated by the combined action of LDL and PRL; however, such inhibition was not demonstrated when cells were treated with the PKC inhibitor, H-7. PKC activation was assessed by measuring the specific association of [H]phorbol dibutyrate (H-PDBu) with luteal cells after treatment with PRL or ionomycin (a positive control). PRL and ionomycin increased H-PDBu-specific binding in early luteal cells by 28+/-5.5% (within 5 min) and 70.2+/-19.3% (within 2 min) over control binding respectively (P<0.05). In addition, PRL did not augment the LDL-stimulated progesterone production in PKC-deficient cells. In contrast with PKC, total inositol phosphate accumulation, as well as intracellular free calcium concentrations, were not affected by PRL in the current study. We conclude that PRL, in the presence of LDL, stimulates progesterone production by early corpora lutea in vitro. Moreover, PRL appears to activate PKC, but not PI-PLC, in these cells. Thus intracellular transduction of the PRL signal may involve activation of PKC that is not dependent on PI-PLC.

The effects of gonadotropin on the phosphatidylinositol pathway in the primate corpus luteum.

The current study was designed to investigate the effects of gonadotropin on basal and prostaglandin (PG) F2 alpha-induced activity of the phosphatidylinositol pathway in corpora lutea (CL) of rhesus monkeys. Luteal progesterone production in vitro was significantly stimulated (P < 0.05) by human chorionic gonadotropin (hCG). Neither basal nor PGF2 alpha-induced phosphatidylinositol 4,5-bisphosphate hydrolysis was significantly influenced by hCG in CL of various ages (P > 0.10). Gonadotropin did induce a slight, yet sustained, increase (P < 0.05) in [Ca2+]i in approximately 70% of luteal cells. The maximal increase in [Ca2+]i in response to hCG (approximately 100 nM) was about one-tenth that induced by PGF2 alpha (approximately 1000 nM). hCG treatment did not alter (P > 0.10) the increase in [Ca2+]i induced by PGF2 alpha Treatment-induced changes in [Ca2+]i did not differ between small (17-21 microns) and large (23-28 microns) luteal cells. Therefore, luteolytic agents are more potent activators of the phosphatidylinositol pathway than luteotropins. This is consistent with the hypothesis that the phosphatidylinositol pathway is involved in primate luteal regression. The inability of hCG to acutely alter the responsiveness of this pathway to PGF2 alpha suggests that CG may rescue the CL of early pregnancy via a mechanism other than direct inhibition of the luteolytic actions of PGF2 alpha.

Phosphorus-31 nuclear magnetic resonance spectroscopy of the spinal cord in the pig, rat, and rabbit.

RATIONALE AND OBJECTIVES: To ensure that contamination-free phosphorus-31 nuclear magnetic resonance (31P-NMR) spectra of the spinal cord could be obtained, a porcine model was adopted that provided a large cord sample and a greater area free from adjacent muscle tissue. METHODS: Phosphorus-31 NMR spectra were acquired from the porcine spinal cord under in vivo conditions using a 4.7-T spectrometer. Spectra also were collected from perchloric acid and lipid extracts, and excised freeze trapped samples of the rat, rabbit, and pig spinal cord. RESULTS: The in vivo spectrum showed resonances corresponding to adenosine triphosphate, phosphocreatine, inorganic phosphate, phosphomonoesterase, and phosphodiesterase as confirmed by extracts. In addition, a broad resonance was observed that was assigned to myelin phospholipids. CONCLUSION: Phosphorus-31 NMR spectra of the spinal cord revealed resonances common to brain tissue. Importantly, the existence of a previously undetected resonance, which is likely to correspond to myelin phospholipids, also is reported. This resonance may prove important in future studies monitoring changes in myelin in response to trauma and ischemia.

Phosphorus-31 magnetic resonance spectroscopy studies of pig spinal cord injury. Myelin changes, intracellular pH, and bioenergetics.

RATIONALE AND OBJECTIVES: Phosphorus-31 (31P) nuclear magnetic resonance (NMR) spectroscopy was used to monitor changes in phosphocreatine (PCr), adenosine triphosphate (ATP), inorganic phosphate (Pi), intracellular pH (pHi), and free magnesium in the in vivo pig spinal cord after injury. METHODS: Phosphorus-31 NMR spectra were acquired from healthy (n = 4) and injured pig spinal cords (n = 8) under in vivo conditions using a 4.7-tesla spectrometer. Spinal cords were injured by dropping a 20-g weight from 20 cm onto the surgically exposed cord surface. RESULTS: In vivo spectra of injured cords revealed a reduction in ATP, PCr, pHi, and an increase in Pi. In addition, a broad resonance that is likely to arise from myelin phospholipids was reduced significantly after injury. CONCLUSIONS: Phosphorus-31 NMR can be used to follow in vivo changes in high energy phosphates after injury and may have the potential to follow changes in myelin structure. This technique may prove important in the study of myelin breakdown after secondary, nonreversible spinal cord injury. Changes in high energy phosphates and pHi did not seem to parallel these putative changes in myelin structure.

Experimental models for spinal cord injury research: physical and physiological considerations.

This paper describes historical and current experimental models used to develop our current understanding of the biomechanics and pathophysiology of traumatic spinal cord injury; the advantages and limitations of current experimental models; considerations for selecting an appropriate injury model based on experimental objectives; and key physiological factors in the spinal cord injury response that may interact with the injury response and alter the outcome. All of the above must be considered in the development and selection of an appropriate experimental injury model that meets specific needs. Various experimental models have been developed to study spinal cord injury and the pathophysiological and physical mechanisms responsible for tissue damage and loss of function. Such modeling may involve inherently different biomechanical variables with alternative outcomes and purposes. There is not, therefore, a single "ideal" experimental injury model just as there is no "stereotypical" clinical spinal cord injury. Instead, the goals and objectives of the research dictate specific requirements on the model. In all cases, however, both physical and physiological aspects of the model should be considered, and measured if possible, to ensure interlaboratory comparability and possible clinical relevance. Also, experimental techniques, especially anesthesia, and surgical procedures, should be carefully reviewed for interactions with the injury response or potential therapeutic interventions to ensure validity of interpretation. It is hoped that data correlating physical spinal cord injury parameters with functional outcome will ultimately be combined with data on vertebral injury and spinal failure mechanics to further our understanding of clinical injury. Such approaches should lead to interventions that reduce the incidence and severity of traumatic human spinal cord injury.

Myoelectric evoked potentials versus locomotor recovery in chronic spinal cord injured rats.

The purpose of this study was to determine the utility of descending evoked potentials in evaluating functional recovery in rats after spinal cord contusion injury. Rats received thoracic contusions at T9 using a controlled-displacement impactor. They were evaluated for 5 weeks postinjury using auditory startle responses (ASR) while alert, or by cerebellar motor evoked potentials (CMEP) while anesthetized. ASR and CMEP were recorded electromyographically from forelimb and hindlimb muscles. Open field locomotor performance was also assessed and recovered to almost normal levels by 3 weeks postinjury. Histologic analysis of the injury site indicated that the contusions destroyed approximately 70% of the cross-sectional area of the cord. Although the remaining 30% was sufficient to preserve nearly normal locomotor behavior, ASR and CMEP amplitudes in hindlimb flexors and extensors were reduced by 90% or more after injury and showed virtually no recovery. Significant ASR and CMEP responses were present in the cutaneous trunk muscles of the lower torso after injury. These muscles are innervated via peripheral nerves originating at cord levels above the injury. Multi-wave field potentials normally recorded from the dorsal cord surface in response to cerebellar stimulation were absent in injured rats, suggesting minimal if any activation of segmental neurons via the pathways normally mediating CMEP. The tracts mediating ASR and CMEP thus appear to be highly sensitive to mild spinal cord trauma but are evidently not essential for support or walking.

An electromechanical spinal injury technique with dynamic sensitivity.

Over the past decade, our laboratory has attempted to create a simple, accurate device that could be used to produce reliable and quantifiable spinal cord injuries in the rodent. We report here on our latest of several modifications of a spinal cord impactor that has allowed us to meet these design criteria. The impactor uses the dynamic capacity of an electromagnetic driver (Ling shaker) and a unique pattern generator to briefly compress the dorsal surface of the spinal cord at velocities that may mimic compression injuries seen in the human. Calibrated, independent transducer systems provide open-loop output of the precise movement (displacement) of the impactor probe and the force necessary to achieve a given displacement. Touch sensitivity is accomplished by vibrating the probe slightly as it approaches the dural surface. This also allows a known biomechanical starting point. This combination of improvements in sensitivity and ability to measure all components of the dynamic compression has allowed us to determine detailed biomechanical descriptors of these impact injuries with low coefficients of variation. Furthermore, such descriptors correlate highly with histopathologic and behavioral outcome measures in animal populations with a variety of injury severities.

Differential expression of MHC class II antigen in the contused rat spinal cord.

Following contusion injury to the dorsal surface of thoracic rat spinal cord, major histocompatibility complex (MHC) class II (Ia) antigen expression by microglia was evaluated throughout the developing lesion. Past investigations of various central nervous system (CNS) lesions have examined short-term or acute sequelae of post-traumatic Ia expression. This report demonstrates that in animals allowed to recover for 18 (sub-chronic) and 45 (chronic) days post-injury, MHC class II antigen is expressed differently at rostral and caudal extents of the lesion as compared with the lesion's epicenter. Following contusion injury to the thoracic spinal cord, sub-chronically injured animals demonstrated Ia-positive microglial staining throughout the white matter rostral and caudal to the epicenter of the lesion, whereas Ia-positive microglia and/or perivascular cells are localized within the gray matter adjacent to it. MHC class II immunoreactivity is down-regulated on microglia at chronic survival times but clusters of Ia-positive macrophages are prominent in regions of maximal degeneration at the epicenter of the lesion. Our findings support the theory that two distinct populations of macrophages participate in resolving traumatic injury. One population is the parenchymal CNS microglia and the other is presumably exudate macrophages derived from the blood. Furthermore, the immunocompetence of these cells as measured by MHC expression may be differentially regulated. This hypothesis is based on differences in Ia-positive staining observed between microglia and macrophages over time concomitant with differences in the spatial distribution of these cell types.

Three-dimensional computer-assisted analysis of graded contusion lesions in the spinal cord of the rat.

Histological analysis of spinal cord injury in experimental animals has focused primarily on the microanatomy of damaged tissue. The current study presents an analysis of the three-dimensional structure of lesion sites in the spinal cord of rats contused with an injury device which produces consistent lesions. Three levels of injury were produced by systematically varying the cord displacement and the duration of the displacement during impact. The resulting groups of subjects exhibited mild, moderate, and severe neurological deficits. Comparisons of equivalent mild impacts made at thoracic versus lumbar spinal cord levels were also made. The results indicate that the overall shape of the lesions is generally biconical, with extensions in the base of the dorsal funiculus, irrespective of the degree of damage or the spinal level of the injury. Lower displacement injuries yielded shorter lesions rostrocaudally with less spread into the white matter. Similar impacts in the lumbar versus thoracic spinal cord produced shorter, more truncated lesion sites at lumbar levels with less involvement of the white matter than in the thoracic lesions. Three-dimensional analyses can can provide additional information about the lesion beyond that available from conventional histopathological measures. Such information could be useful in assessing the results of posttraumatic manipulations which are directed at reducing tissue damage or tissue replacement via transplantation.

Neural tissue transplantation and CNS trauma: anatomical and functional repair of the injured spinal cord.

Neural tissue transplantation has become recognized widely as a powerful experimental tool for studying structure-function relationships, development, plasticity, and capacities for regeneration in the adult CNS. In addition, this area of investigation has generated considerable interest in approaches that might be applicable to a variety of catastrophic neurological disorders. In this regard, attention has been given to neural tissue grafting as a potential therapeutic strategy in various forms of neurodegenerative disease. More recently, however, other investigations have begun to focus on the possible application of peripheral and central neural tissue transplants for promoting repair in forms of CNS trauma. This review highlights various neural transplantation approaches that have been explored primarily in the context of injury to the adult CNS, with emphasis on spinal cord injury. An overview is presented of the evolution of this area of research in terms of emerging biological perspectives, technological advances, and experimental modelling. Discussion centers on progress that has been made and a variety of theoretical and practical issues that remain to be resolved.

Operant conditioning of H-reflex increase in spinal cord--injured rats.

Operant conditioning of the spinal stretch reflex or its electrical analog, the H-reflex, is a new model for exploring the mechanisms of long-term supraspinal control over spinal cord function. Primates and rats can gradually increase (HRup conditioning mode) or decrease (HRdown conditioning mode) the H-reflex when reward is based on H-reflex amplitude. An earlier study indicated that HRdown conditioning of the soleus H-reflex in rats is impaired following contusion injury to thoracic spinal cord. The extent of impairment was correlated with the percent of white matter lost at the injury site. The present study investigated the effects of spinal cord injury on HRup conditioning. Soleus H-reflexes were elicited and recorded with chronically implanted electrodes from 14 rats that had been subjected to calibrated contusion injuries to the spinal cord at T8. At the lesion epicenter, 12-39% of the white matter remained. After control-mode data were collected, each rat was exposed to the HRup conditioning mode for 50 days. Final H-reflex amplitudes after HRup conditioning averaged 112% (+/-22% SD) of control. This value was significantly smaller than that for 13 normal rats exposed to HRup conditioning, in which final amplitude averaged 153% (+/-51%) SD of control. As previously reported for HRdown conditioning after spinal cord injury, success was inversely correlated with the severity of the injury as assessed by white matter preservation and by time to return of bladder function. HRup and HRdown conditioning are similarly sensitive to injury. These results further demonstrate that H-reflex conditioning is a sensitive measure of the long-term effects of injury on supraspinal control over spinal cord functions and could prove a valuable measure of therapeutic efficacy.

Operant conditioning of H-reflex in spinal cord-injured rats.

Operant conditioning of the spinal stretch reflex or its electrical analog, the H-reflex, is a new model for exploring the mechanisms of supraspinal control over spinal cord function. Both rats and primates can gradually increase (HRup conditioning mode) or decrease (HRdown conditioning mode) soleus H-reflex magnitude when exposed to an operant conditioning task. This study used H-reflex operant conditioning to assess and modify spinal cord function after injury. Soleus H-reflexes were elicited and recorded with chronically implanted electrodes from rats that had been subjected to calibrated contusion injuries to the spinal cord at T8. From 18 to 140 days after injury, background EMG, M response amplitude, and initial H-reflex amplitude were not significantly different from those of normal rats. HRdown conditioning was successful in some, but not all, spinal cord-injured rats. The H-reflex decrease achieved by conditioning was inversely correlated with the severity of the injury as assessed histologically or by time to return of bladder function. It was not correlated with the length of time between injury and the beginning of conditioning. The results confirm the importance of descending control from supraspinal structures in mediating operantly conditioned change in H-reflex amplitude. In conjunction with recent human studies, they suggest that H-reflex conditioning could provide a sensitive new means for assessing spinal cord function after injury, and might also provide a method for initiating and guiding functional rehabilitation.

Analysis of TGF-beta 1 gene expression in contused rat spinal cord using quantitative RT-PCR.

We have used northern blot analysis and quantitative reverse transcription polymerase chain reaction (RT-PCR) to determine the postinjury expression profile of the transforming growth factor-beta 1 (TGF-beta 1) gene in the contused rat spinal cord. Spectrophotometric estimates of total sample RNA and quantitative analyses of cyclophilin mRNA using RT-PCR served as controls for comparisons between samples. No changes in cyclophilin gene expression were found at any postinjury survival times. The results of the TGF-beta 1 analyses, which were carried out on spinal cord samples taken at postinjury intervals ranging from 6 h to 10 days, show that the amount of TGF-beta 1 mRNA present in spinal cord increases rapidly following injury, reaching maximum levels 7 days postinjury. Unoperated control samples contained approximately 2 x 10(8) molecules of TGF-beta 1 mRNA/0.5 microgram total RNA. By 1 day postinjury, the amount of TGF-beta 1 mRNA in the cord had increased by a factor of 2.5 to 5 x 10(8) molecules/0.5 microgram total RNA. At 7 days postinjury, there were approximately 15 x 10(8) molecules of TGF-beta 1 mRNA/0.5 microgram total RNA. By 10 days postinjury the amount of TGF-beta 1 mRNA present in the spinal cord had declined to 8 x 10(8) molecules of TGF-beta 1 mRNA/0.5 microgram total RNA, a value similar to that observed at 3 days postinjury. The roles that TGF-beta 1 might play in modifying cellular responses in injured spinal cord are discussed.