Short-term prediction of secondary progression in a sliding window: A test of a predicting algorithm in a validation cohort.

INTRODUCTION: The Multiple Sclerosis Prediction Score (MSPS, www.msprediction.com) estimates, for any month during the course of relapsing-remitting multiple sclerosis (MS), the individual risk of transition to secondary progression (SP) during the following year. OBJECTIVE: Internal verification of the MSPS algorithm in a derivation cohort, the Gothenburg Incidence Cohort (GIC, n = 144) and external verification in the Uppsala MS cohort (UMS, n = 145). METHODS: Starting from their second relapse, patients were included and followed for 25 years. A matrix of MSPS values was created. From this matrix, a goodness-of-fit test and suitable diagnostic plots were derived to compare MSPS-calculated and observed outcomes (i.e. transition to SP). RESULTS: The median time to SP was slightly longer in the UMS than in the GIC, 15 vs. 11.5 years (p = 0.19). The MSPS was calibrated with multiplicative factors: 0.599 for the UMS and 0.829 for the GIC; the calibrated MSPS provided a good fit between expected and observed outcomes (chi-square p = 0.61 for the UMS), which indicated the model was not rejected. CONCLUSION: The results suggest that the MSPS has clinically relevant generalizability in new cohorts, provided that the MSPS was calibrated to the actual overall SP incidence in the cohort.

Diffusion tensor imaging in multiple sclerosis at different final outcomes.

OBJECTIVES: Methods to evaluate the relative contributions of demyelination vs axonal degeneration over the long-term course of MS are urgently needed. We used magnetic resonance diffusion tensor imaging (DTI) to estimate degrees of demyelination and axonal degeneration in the corpus callosum (CC) in cases of MS with different final outcomes. MATERIALS AND METHODS: We determined DTI measures mean diffusivity (MD), fractional anisotropy (FA), and axial (AD) and radial (RD) diffusivities in the CC of 31 MS patients, of whom 13 presented a secondary progressive course, 11 a non-progressive course, and seven a monophasic course. The study participants were survivors from an incidence cohort of 254 attack-onset MS patients with 50 years of longitudinal follow-up. As reference, we included five healthy individuals without significant morbidity. RESULTS: In patients with secondary progression, compared to all other groups, the corpus callosum showed increased RD and reduced FA, but no change in AD. None of the parameters exhibited differences among non-progressive and monophasic course groups and controls. CONCLUSION: Increased RD was observed in secondary progressive MS, indicating significant myelin loss. Normal RD values observed in the clinically isolated syndrome and non-progressive groups confirm their benign nature. AD was not a characterizing parameter for long-term outcome. Demyelination revealed by increased RD is a distinguishing trait for secondary progression.

Clinically isolated syndromes with no further disease activity suggestive of multiple sclerosis at the age of population life expectancy.

The proportion of patients with clinically isolated syndrome (CIS) reported to convert to clinically definite multiple sclerosis varied between 30 and 75%. We studied the lifetime probability of remaining in the "CIS only" condition. The study was based on the longitudinally followed Gothenburg 1950-1964 incidence cohort (n = 306). Survival analysis revealed that 17.8% of 236 attack onset patients remained "CIS only". Patients with afferent (optic and sensory) symptoms had a better prognosis with approximately 30% of these patients remaining "CIS only". Patients who had experienced no relapse during the first 25 years remained "CIS only" for the subsequent 25 years of follow-up.

Time to secondary progression in patients with multiple sclerosis who were treated with first generation immunomodulating drugs.

BACKGROUND: It is currently unknown whether early immunomodulatory treatment in relapsing-remitting MS (RRMS) can delay the transition to secondary progression (SP). OBJECTIVE: To compare the time interval from onset to SP in patients with RRMS between a contemporary cohort, treated with first generation disease modifying drugs (DMDs), and a historical control cohort. METHODS: We included a cohort of contemporary RRMS patients treated with DMDs, obtained from the Swedish National MS Registry (disease onset between 1995-2004, n = 730) and a historical population-based incidence cohort (onset 1950-64, n = 186). We retrospectively analyzed the difference in time to SP, termed the "period effect" within a 12-year survival analysis, using Kaplan-Meier and Cox regression analysis. RESULTS: We found that the "period" affected the entire severity spectrum. After adjusting for onset features, which were weaker in the contemporary material, as well as the therapy initiation time, the DMD-treated patients still exhibited a longer time to SP than the controls (hazard ratios: men, 0.32; women, 0.53). CONCLUSION: Our results showed there was a longer time to SP in the contemporary subjects given DMD. Our analyses suggested that this effect was not solely driven by the inclusion of benign cases, and it was at least partly due to the long-term immunomodulating therapy given.

A comparison of the effects of two antispastic drugs, tizanidine and baclofen, on synaptic transmission from muscle spindle afferents to spinal interneurones in cats.

The alpha 2-adrenergic agonist tizanidine was reported to be more efficient than baclofen in reducing muscle tone in some spastic patients. The aim of this study was to investigate if this might be due to more specific depressive actions of tizanidine on transmission from muscle afferents which contribute to muscle tone. This was done by comparing the effects of tizanidine and baclofen on amplitudes of monosynaptic spinal focal field potentials produced by stimulation of muscle nerves in the cat. Such field potentials were recorded in the intermediate zone of the fourth lumbar segment, where they display two distinct components, an early one from group I afferents and a later one from group II afferents. Both reflect EPSPs produced in interneurones in disynaptic pathways to motoneurones. Tizanidine strongly depressed potentials caused by group II afferents, while it had no effect or slightly facilitated potentials produced by group I afferents. In contrast, baclofen had inconsistent effects on the group II potentials; in some cases it caused a depression and in others it caused only an increase in the latency and time to peak, at doses that strongly and consistently depressed the group I potentials. These effects have been found after both local and systemic applications. The antispastic actions of tizanidine may therefore only be related to the depression of transmission from group II muscle afferents, while antispastic actions of baclofen may be secondary to the depression of any sensory fibres. Since tizanidine is as effective in depressing spasticity as baclofen, it is suggested that the enhancement in synaptic transmission from group II muscle afferents may play an important role in the development of exaggerated stretch reflexes in spastic patients.

Dopaminergic control of transmission from group II muscle afferents to spinal neurones in the cat and guinea-pig.

The effects of dopamine and its agonists on transmission from muscle afferents to spinal neurones were investigated in the cat and guinea-pig spinal cord, by measuring the drug effects on the amplitude of monosynaptic field potentials evoked by electrical stimulation of group I and group II muscle afferents. Local iontophoretic application of dopamine, the dopamine D1/D5 agonist SKF-38393 and the D2/D3/D4 agonist quinpirole all depressed the group II field potentials evoked at the base of the dorsal horn. Group II field potentials in the intermediate zone were depressed by dopamine to a similar degree as the dorsal horn field potentials, whereas the dopamine agonists were without effect upon them. The intermediate zone field potentials evoked by group I muscle afferents were not depressed by any of the drugs. The dopamine-evoked depression of the group II-evoked field potentials in the dorsal horn in the guinea-pig spinal cord was reduced by the simultaneous application of haloperidol. The results demonstrate that dopamine receptors mediate the depression of transmission from group II muscle afferents to interneurones in the dorsal horn, but not to neurones in the intermediate zone of the spinal cord.

Depression of transmission from group II muscle afferents by electrical stimulation of the cuneiform nucleus in the cat.

The effects of short trains of electrical stimuli applied within the cuneiform nucleus and the subcuneiform region were examined on transmission from group I and group II muscle afferents to first-order spinal neurons. Variations in the effectiveness of transmission from these afferents were assessed from changes in the sizes of the monosynaptic component of extracellular field potentials evoked following stimulation of muscle nerves. Field potentials evoked from group II muscle afferents in the dorsal horn of the midlumbar and sacral segments and in the intermediate zone of the midlumbar segments were reduced when the test stimuli applied to peripheral nerves were preceded by conditioning stimulation of the cuneiform nucleus or the subcuneiform region. The depression occurred at conditioning-testing intervals of 20-400 ms, being maximal at intervals of 32-72 ms for dorsal horn potentials and 40-100 ms for intermediate zone potentials. At the shortest intervals, both group II and group I field potentials in the intermediate zone were depressed. Conditioning stimulation of the cuneiform nucleus depressed group II field potentials nearly as effectively as conditioning stimulation of the coerulear or raphe nuclei. We propose that the nonselective depression of transmission from group I and II afferents at short intervals is due to the activation of reticulospinal pathways by cells or fibers stimulated within the cuneiform area. We also propose that the selective depression of transmission from group II afferents at long intervals is mediated at least partly by monoaminergic pathways, in view of the similarity of the effects of conditioning stimulation of the cuneiform nucleus and of the brainstem monoaminergic nuclei and by directly applied monoamines (Bras et al. 1990). In addition, it might be caused by primary afferent depolarization mediated by non-monoaminergic fibers (Riddell et al. 1992).

High molecular weight growth hormone (> 160 kD) in human serum characterized with monoclonal antibodies.

Human growth hormone (hGH) was analyzed by six monoclonal antibodies (Mabs) and a polyclonal antiserum (Pas) before and after molecular sieve chromatography of sera from healthy subjects. Their hGH levels were between < 0.2 and 0.4 ng/ml as determined with Pas. The six Mabs reacted with five distinct epitopes and bound to a hGH fragment corresponding to the amino acid sequence 15-125. Two of the Mabs showed reduced binding to 20-kD hGH. The binding of Mabs to dimeric forms of hGH varied. Human GH levels in unfractionated sera as determined with Mabs were < 3.1-390 ng/ml. After molecular sieve chromatography of the sera, one peak of hGH-immunoreactive material of high molecular weight (> 160 kD) and one at the elution volume of monomeric hGH were determined with Pas and Mabs. The major part of the high molecular weight hGH (> 160 kD) seemed to consist of 22-kD hGH molecules, since Pas and all Mabs detected the hGH immunoreactivity (> 160 kD) in a similar manner. This high molecular weight hGH (> 160 kD) was distinguishable from the identified, receptor-like hGH-binding protein in serum.

Gating of transmission to motoneurones by stimuli applied in the locus coeruleus and raphe nuclei of the cat.

1. Neuronal systems activated by stimulation in the region of the locus coeruleus/subcoeruleus (LC/SC) and raphe nuclei have previously been shown to depress transmission from group II muscle afferents in regions of the midlumbar spinal segments in which premotor interneurones are located. The aim of the present investigation was to determine the extent to which such depression is paralleled by depression of the reflex actions of group II afferents on motoneurones. 2. The effects of short trains of conditioning electrical stimuli applied within the LC/SC and raphe nuclei were examined on postsynaptic potentials (PSPs) evoked by group I and group II muscle afferents in hindlimb motoneurones. The effects were examined over a wide range of conditioning-test intervals but particular emphasis was placed on the effects produced at long intervals (> 100 ms) since such effects are more likely to be mediated by the descending noradrenergic and serotonergic neurones of the LC/SC and raphe nuclei which are of slow conduction velocity. In addition, conditioning stimuli alone evoked PSPs in motoneurones (with latencies of 7-15 ms and a duration of 50-80 ms) and effects evoked at short conditioning-test intervals might therefore have been secondary to changes in motoneurone membrane properties. 3. At conditioning-test intervals between 100 and 350 ms synaptic actions of group II origin were strongly and consistently depressed. Both EPSPs and IPSPs were affected, two-thirds of those tested being reduced in amplitude by 50% or more. A similar depression was exerted on PSPs evoked from the quadriceps and deep peroneal nerves mediated predominantly by interneurones located in the midlumbar segments and on PSPs evoked from the hamstring and triceps surae nerves mediated by interneurones located in more caudal segments. It is thus concluded that neuronal systems activated by stimuli applied in the LC/SC and raphe nuclei are capable of gating transmission in all those interneuronal pathways which mediate the reflex actions of group II afferents on motoneurones in anaesthetized animals.

Do noradrenergic descending tract fibres contribute to the depression of transmission from group II muscle afferents following brainstem stimulation in the cat?

Two alpha 2 noradrenaline antagonists, idazoxan and yohimbine, were injected in midlumbar segments of the spinal cord to test whether they counteract depression of field potentials evoked by group II muscle afferents by conditioning stimuli applied in the brainstem. The tested field potentials were those evoked monosynaptically in the intermediate zone of midlumbar segments. Their depression reflected thus the depression of transmission between group II fibres and their first relay neurones. The conditioning stimuli were applied either within the ipsilateral locus coeruleus/subcoeruleus or outside these nuclei (in the raphe magnus, raphe obscurus, or cuneiform nuclei). The brainstem evoked depression of the tested field potentials (n = 12) was reduced following injection of idazoxan or yohimbine to about two thirds of that which was evoked originally but in three cases to about one half. The study leads thus to the conclusion that noradrenergic descending tract neurones contribute to the depression of transmission from group II afferents to spinal interneurones and that such noradrenergic neurones are activated by stimuli applied within as well as outside their nuclei. However, the relative contribution of monoaminergic and non-monoaminergic descending tract neurones to the control of transmission from group II afferents to these neurones remains to be established.

Monoclonal antibodies reveal circulating growth hormone of high molecular weight not detectable by conventional assays.

A radioimmunoassay based on a monoclonal antibody. Mc-ab 1, which was raised against growth hormone but cross-reacted with human placental lactogen yielded higher GH immunoreactivity levels in serum than one based on a polyclonal antiserum. This discrepancy was noted in subjects with normal GH secretion as well as in patients with GH insufficiency. To characterize this GH immunoreactivity detected by Mc-ab 1, affinity purification and molecular sieve chromatography of serum were performed. High molecular weight proteins with GH immunoreactivity were found with both techniques. These proteins were associated with carbohydrates. Affinity cross-linking showed specific binding of radiolabelled GH to high molecular weight proteins in the serum. After fractionation of serum, the GH immunoreactivity became detectable by the polyclonal antiserum assay as well as by an immunoradiometric assay. GH immunoreactive material with an approximate mass of 80 kD was subjected to isoelectric focusing. When GH immunoreactive fractions at pH 5 were re-chromatographed, GH immunoreactivity was recovered in the elution volume corresponding to monomeric GH. Our results show that sera from normal subjects as well as from patients with deficient GH secretion contain notable amounts of high molecular weight GH which is undetectable by antibodies generally used for GH measurements, but which can be revealed after fractionation of serum.

Comparison of Effects of Various Types of NA and 5-HT Agonists on Transmission from Group II Muscle Afferents in the Cat.

A number of noradrenaline and serotonin agonists were tested to investigate which of them replicate the depressive actions of monoamines on transmission from group II muscle afferents in the cat spinal cord. The agonists were applied ionophoretically at the two sites at which maximal monosynaptic focal field potentials are evoked from group II afferents-in the intermediate zone and the dorsal horn of the 4th and 5th lumbar segments. Their effects were estimated from changes in the amplitude of the field potentials. The compounds tested fell into three categories according to the site at which they depressed transmission from group II afferents: one category with highly selective actions in the intermediate zone, a second category with similarly selective actions in the dorsal horn, and a third category with non-selective actions. Drugs in the first category included three noradrenaline agonists (tizanidine, B-HT 933 and clonidine), included in the second were five serotonin agonists (8-OH-DPAT, 5-methoxytryptamine, alpha-methyl serotonin, DOI and 2-methyl-serotonin), and in the third two noradrenaline agonists (phenylephrine and isoproterenol) and two serotonin agonists (RU 24969 and 5-carboxamidotryptamine). Field potentials evoked by group I afferents remained unaffected by all but one compound (8-OH-DPAT). Effects of one noradrenaline agonist and one serotonin agonist (tizanidine and 5-methoxytryptamine) were also tested on responses of single extracellularly recorded neurons. Tizanidine depressed responses induced by stimulation of group II afferents in intermediate zone interneurons, but not in dorsal horn neurons, while 5-methoxytryptamine depressed activation of the latter. Tizanidine had no effect on responses evoked by group I afferents, either in intermediate zone interneurons or in the dorsal spino-cerebellar tract neurons of Clarke's column. It is hypothesized that noradrenaline and serotonin released by descending monoaminergic neurons differ in the potency with which they depress transmission from group II afferents to different functional types of neuron. The results suggest that this depression may involve different membrane receptors at different locations, primarily alpha2 adrenoceptors in the intermediate zone/ventral horn and 5-HT1A serotonin receptors in the dorsal horn.

Biological characterization of charge isomers of human growth hormone.

Since deamidation of the human GH molecule may alter the manner and extent to which the hormone is cleaved by proteases, and since it has been repeatedly suggested that proteolytic processing is required for the expression of certain of the activities of GH, the present study was conducted to determine whether the biological activity profiles of more acidic forms of human GH are altered. Three charge isomers, GH-b, GH-c and GH-d, representing primarily deamidated forms, were isolated from a native human GH preparation (Crescormon) in amounts adequate for characterization of their biological activities. All three were essentially equipotent in a radioimmunoassay for human GH. When assessed for growth-promoting activity in the hypophysectomized rat, the isomers were again equipotent with each other and with the GH preparation from which they were derived. The charge isomers also had significant in vitro insulin-like activity on isolated rat adipose tissue and diabetogenic activity in the ob/ob mouse. Thus, the biological activity profiles of these charge isomers of human GH do not differ greatly from one another.

The acute effects of growth hormone on amino acid transport and protein synthesis are due to its insulin-like action.

GH has acute stimulatory effects on amino acid transport and protein synthesis in a variety of tissues, but it has not been established whether these effects are expressions of the growth-promoting property of GH or of its separate insulin-like action. The 20,000-dalton structural variant of human GH (20K hGH) has been shown to have a high ratio of growth-promoting to insulin-like activity compared to native hGH (22K hGH), suggesting that it could be used as a tool to address the above question. Therefore, experiments were conducted to compare the relative abilities of native 22K hGH and 20K hGH, when added in vitro, to stimulate amino acid transport and protein synthesis in the isolated diaphragm of the female hypophysectomized rat. Paired intact hemidiaphragms were preincubated for 1 h in the absence or presence of various concentrations of 22K or 20K hGH. Then, 3-O-[14C]methylglucose was added to the medium to measure sugar transport as a test of insulin-like activity, and either alpha-[3H]aminoisobutyric acid acid or [3H] phenylalanine was also added to measure amino acid transport or protein synthesis, respectively, during a final hour of incubation. When the responses to the various concentrations of 22K and 20K were compared, 20K hGH was only about 20% as effective as 22K in stimulating 3-O-methylglucose transport, reflecting its markedly attenuated insulin-like activity on the diaphragm. Similarly, 20K hGH was only 20% as effective as 22K hGH in stimulating alpha-aminoisobutyric acid transport and phenylalanine incorporation into protein in the same muscles. Therefore, these findings support the idea that the rapid stimulatory effects of GH on amino acid transport and protein synthesis are expressions of the insulin-like action of GH and are not components of the response of target cells to its growth-promoting action.

Biological characterization of purified native 20-kDa human growth hormone.

Because of the propensity of the 20-kDa variant of human growth hormone (GH) to aggregate with itself and with 22-kDa human GH, it has been difficult to prepare monomeric 20-kDa GH in highly purified form. This has been a major complicating factor in determining whether 20-kDa GH has a biological activity profile distinct from that of 22-kDa GH. In the present study, native 20-kDa GH was isolated from a human GH dimer concentrate and purified by a procedure that included column electrophoresis in agarose suspension as a final separation step. This procedure yielded highly purified monomeric 20-kDa GH, which was contaminated to an extent of less than 1% with 22-kDa GH, and which exhibited only a small degree of dimerization upon storage. The native 20-kDa GH was quite active in stimulating growth in hypophysectomized rats, when growth was assessed by body weight gain, longitudinal bone growth, the stimulation of sulfation of cartilage, and the elevation of serum IGF-1 level. However, in all of these growth assays, the 20-kDa GH was somewhat less active than the native 22-kDa GH to which it was compared; e.g., in the body weight gain and longitudinal bone growth assays, it had an estimated potency of 0.6 relative to the 22-kDa GH. The 20-kDa GH exhibited substantial diabetogenic activity when tested for the ability to raise fasting blood glucose concentration and to impair glucose tolerance in ob/ob mice. Also, the native 20-kDa GH had significant in vitro insulin-like activity, although its potency was approximately 20% that of the native 22-kDa GH to which it was compared. Thus, the biological activity profile of native 20-kDa GH differs from that of 22-kDa GH primarily in that insulin-like activity is markedly attenuated.

Labelling of midlumbar neurones projecting to cat hindlimb motoneurones by transneuronal transport of a horseradish peroxidase conjugate.

When wheat germ agglutinin conjugated with horseradish peroxidase (WGA-HRP) is injected into a hindlimb nerve it is first transported to motoneuronal somata and then, transneuronally, to interneurones in several spinal segments. However, the distribution of the labelled interneurones has been found to be dependent on the experimental conditions. Interneurones of the S1, L7, L6 and L5 segments were labelled in all the preparations used in this study while interneurones of the L4 and L3 segments were labelled only in some of them. The labelling of the L3-L4 interneurones was found when the animals were awake and active (after injection of WGA-HRP under short lasting anaesthesia) or when the contralateral pyramid was stimulated in animals which remained anaesthetized during the whole survival period. Since the transneuronal transport of WGA-HRP is enhanced by synaptic activity, these results are taken to indicate that L3 and L4 interneurones operate as last order neurones of neuronal pathways which subserve centrally initiated movements, in particular those activated by the corticospinal tract fibres.

Purification of pituitary and biosynthetic human growth hormone using monoclonal antibody immunoadsorbent.

This report describes the purification of human growth hormone from crude pituitary extract and lysate of recombinant E. coli by an immunoadsorbent purification with monoclonal antibody coupled to solid phase. By a single-immunoaffinity chromatography step pure hGH can be obtained from either origin as revealed by SDS-PAGE followed by silver staining or immunoblotting. An additional ion-exchange chromatography step results in homogeneous 22 kDa hGH preparations. Furthermore, this method may be used for isolation of a pituitary hGH variant which has higher binding affinity for this monoclonal antibody than the major 22 kDa form. This study clearly illustrates the potential of monoclonal antibody immunoadsorbents for purification of different molecular forms of hGH.

Convergence onto interneurons subserving primary afferent depolarization of group I afferents.

The aim of the study was to investigate whether common or independent neuronal pathways are used to evoke primary afferent depolarization (PAD) from selectively activated group Ia and Ib afferents of different muscles. To this end, the spatial facilitation of effects of various afferents, indicating convergence on the same interneurons, was used as a test. Its occurrence was assessed on dorsal root potentials (DRPs) evoked in unspecified fibers or using intra-axonal recording from identified group Ia muscle spindle afferents or group Ib tendon organ afferents. Spatial facilitation has been found in PAD pathways a) from various Ia-afferents, whether of flexors or extensors; b) from various Ib-afferents, whether of flexors or extensors; and c) from flexor Ib-afferents and flexor or extensor Ia-afferents. In contrast, no indications have been found for common pathways from extensor Ib- and any Ia-afferents under conditions that proved effective in other combinations. Latencies of those components of PAD that appeared as a result of the spatial facilitation ranged from 2 to more than 7 ms, indicating that the convergence occurred in the shortest (trisynaptic) as well as longer pathways. The same patterns of convergence have been found in PAD pathways to extensor and flexor Ia-afferents (in experiments with intraaxonal recording from these afferents). The possibility might thus be considered that some neuronal pathways are used to modulate transmission via Ia-afferents independently of their muscle origin. The same might hold true for extensor and flexor Ib-afferents. Generally, it is concluded that the minimal number of distinct neuronal populations subserving PAD of group I afferents may be two to six. Additionally, actions of cutaneous, joint, and interosseous afferents on DRPs from Ia-afferents were reexamined to further the comparison between neurons mediating PAD and those mediating postsynaptic excitation or inhibition of motoneurons. Only depression of Ia DRPs followed stimulation of these afferents at intensities of 1.5-2.0 times threshold and higher; lower threshold afferents were apparently ineffective. On the basis of lack of convergence of extensor Ib and Ia muscle afferents and of low-threshold cutaneous afferents, interneurons mediating PAD may thus be distinguished from the interneurons subserving Ib and Ia-like-Ib postsynaptic actions in motoneurons. The latter are coexcited by these three groups of afferents.

Immunochemical characterization of charge isomers of bacteria-derived human growth hormone with monoclonal antibodies.

Monoclonal antibodies were used to study the immunochemical nature of charge isomers of bacterially produced methionyl human growth hormone. After isoelectric focusing of the hormone the 12 monoclonal antibodies reacted similarly in immunoblotting experiments and none of them could discriminate between the two isolated charge isomers in ELISA. This indicates that the generation of charge isomers of met-hGH does not result in loss of the determinants recognized by the monoclonal antibodies and that the conformation of the two main charge isomers is identical within these determinants.

Post-synaptic potentials in a population of motoneurones following activity of single interneurones in the cat.

The technique of recording post-synaptic potentials from a population of motoneurones, by recording from ventral roots perfused with isotonic sucrose, has been applied to investigate the action of single last-order interneurones; the target motoneurones were in either caudal L7 or S1 segments. Using spike-triggered averaging, the inhibitory action of 70% of previously identified last-order interneurones (Renshaw cells and lamina VII Ia inhibitory interneurones) has been detected. Previous observations had suggested that interneurones mediating disynaptic non-reciprocal inhibition from group I muscle afferents should be characterized by (i) location in laminae V-VI, (ii) monosynaptic group I input and (iii) ascending collateral axonal projection to upper lumbar segments. 65% of interneurones with these characteristics were found to inhibit motoneurones. In addition, spike-triggered averaging from this group of laminae V-VI interneurones sometimes revealed a depolarizing potential which preceded the inhibitory potential evoked by the interneurone. The depolarizing potential is interpreted as being due to the action of some presynaptic fibres which branch to innervate both the investigated interneurones and motoneurones.