Synthesis and transport of creatine in the CNS: importance for cerebral functions.

J. Neurochem. (2010) 10.1111/j.1471-4159.2010.06935.x Abstract Apart of its well known function of 'energetic buffer' through the creatine/phosphocreatine/creatine kinase system allowing the regeneration of ATP, creatine has been recently suggested as a potential neuromodulator of even true neurotransmitter. Moreover, the recent discovery of primary creatine deficiency syndromes, due to deficiencies in l-arginine : glycine amidinotransferase or guanidinoacetate methyltransferase (the two enzymes allowing creatine synthesis) or in the creatine transporter, has shed new light on creatine synthesis, metabolism and transport, in particular in CNS which appears as the main tissue affected by these creatine deficiencies. Recent data suggest that creatine can cross blood-brain barrier but only with a poor efficiency, and that the brain must ensure parts of its needs in creatine by its own endogenous synthesis. Finally, the recent years have demonstrated the interest to use creatine as a neuroprotective agent in a growing number of neurodegenerative diseases, including Parkinson's and Huntington's diseases. This article aims at reviewing the latest data on creatine metabolism and transport in the brain, in relation to creatine deficiencies and to the potential use of creatine as neuroprotective molecule. Emphasis is also given to the importance of creatine for cerebral function.

Role of the HCF-1 basic region in sustaining cell proliferation.

BACKGROUND: The human herpes simplex virus-associated host cell factor 1 (HCF-1) is a conserved human transcriptional co-regulator that links positive and negative histone modifying activities with sequence-specific DNA-binding transcription factors. It is synthesized as a 2035 amino acid precursor that is cleaved to generate an amino- (HCF-1(N)) terminal subunit, which promotes G1-to-S phase progression, and a carboxy- (HCF-1(C)) terminal subunit, which controls multiple aspects of cell division during M phase. The HCF-1(N) subunit contains a Kelch domain that tethers HCF-1 to sequence-specific DNA-binding transcription factors, and a poorly characterized so called "Basic" region (owing to a high ratio of basic vs. acidic amino acids) that is required for cell proliferation and has been shown to associate with the Sin3 histone deacetylase (HDAC) component. Here we studied the role of the Basic region in cell proliferation and G1-to-S phase transition assays. METHODOLOGY/PRINCIPAL FINDINGS: Surprisingly, much like the transcriptional activation domains of sequence-specific DNA-binding transcription factors, there is no unique sequence within the Basic region required for promoting cell proliferation or G1-to-S phase transition. Indeed, the ability to promote these activities is size dependent such that the shorter the Basic region segment the less activity observed. We find, however, that the Basic region requirements for promoting cell proliferation in a temperature-sensitive tsBN67 cell assay are more stringent than for G1-to-S phase progression in an HCF-1 siRNA-depletion HeLa-cell assay. Thus, either half of the Basic region alone can support G1-to-S phase progression but not cell proliferation effectively in these assays. Nevertheless, the Basic region displays considerable structural plasticity because each half is able to promote cell proliferation when duplicated in tandem. Consistent with a potential role in promoting cell-cycle progression, the Sin3a HDAC component can associate independently with either half of the Basic region fused to the HCF-1 Kelch domain. CONCLUSIONS/SIGNIFICANCE: While conserved, the HCF-1 Basic region displays striking structural flexibility for controlling cell proliferation.

Phenotypic and molecular characterization of Bruck syndrome (Osteogenesis imperfecta with contractures of the large joints) caused by a recessive mutation in PLOD2

Le Syndrome de Bruck (Bruck Syndrome; BS) est une maladie autosomique récessive assemblant la combinaison inhabituelle de fragilité osseuse semblable à celle de l'Ostéogenèse Imparfaite (0I) avec des contractures congénitales tendineuses et cutanées des grandes articulations («ptérygia»). Les cas décrits jusqu'à ce jour mettent en évidence une grande hétérogénéité du tableau clinique, liée en partie au manque d'un diagnostic biochimique ou moléculaire. Nous savons que dans le BS les gènes codant pour le collagène 1 ne sont pas mutés, mais savons néanmoins, grâce à l'étude du collagène extrait de biopsies osseuses, qu'il y a un déficit d'hydroxylation des résidus de lysine dans les télopeptides du collagène 1 qui servent à la formation des liens intermoléculaires (crosslinks) et donc à la stabilisation des fibres de collagène. Un locus génétique du BS à été mappé sur 17q12, mais le gène responsable sur ce locus reste inconnu; plus récemment, deux mutations dans le gène de la lysyl hydroxylase 2 (PLOD2, position chromosomique 3q23-q24) ont été identifiées, démontrant l'hétérogénéité génétique du ES. La proportion de ES liée à 17p22 (BS type 1) et celle liée à une mutation dans PLOD2 (BS type 2) est encore incertaine et nous manquons de données sur la corrélation phenotype-génotype. Nous avons étudié le cas d'un garçon avec des contractures et des ptérygia dès la naissance, combinées à une ostéopénie sévère de type OI menant à des fractures multiples. Ses urines contenaient une quantité élevée d'hydroxyproline, indiquant un remaniement important du tissu osseux, mais peu de produits de dégradation des crosslinks du collagène, indiquant donc une réduction de la proportion de crosslinks dans le collagène in vivo. Nous avons pu démontrer chez lui la présence d'une nouvelle mutation homozygote dans le gène PLOD2 menant à une substitution Arg598His; les deux parents du sujet étaient hétérozygotes pour la mutation et celle-ci était absente dans notre population témoin. La mutation est adjacente aux deux mutations rapportées précédemment (Gly601Val et Thr608Ile), ce qui suggère la présence d'un ''hotspot'' mutationnel mais aussi d'une région de grande importance fonctionnelle sur PLOD2 : cette observation est importante pour la création d'inhibiteurs de PLOD2, recherchés en ce moment pour le traitement de la fibrose. La combinaison de ptérygia et de fragilité osseuse, comme illustrée par notre patient est apparemment contradictoire et donc difficilement explicable mais indique que l'hydroxylation des résidus lysyl des télopeptides est importante non seulement pour la stabilité osseuse mais aussi dans la morphogénèse et la formation des articulations dans la période prénatale. Finalement, la mesure des produits de dégradation du collagène dans l'urine et l'analyse de mutation de PLOD2 permet le diagnostic du syndrome de Bruck et permet de le différencier de l'Osteogénèse Imparfaite. -- Bruck syndrome (BS) is a recessively-inherited phenotypic disorder featuring the unusual combination of skeletal changes resembling osteogenesis imperfecta (0I) with congenital contractures of the large joints. Clinical heterogeneity is apparent in cases reported thus far. While the genes coding for collagen 1 chains are unaffected in BS, there is biochemical evidence for a defect in the hydroxylation of lysine residues in collagen 1 telopeptides. One BS locus has been mapped at 17p12, but more recently, two mutations in the lysyl hydroxylase 2 gene (PLOD2, 3q23-q24) have been identified in BS, showing genetic heterogeneity. The proportion of BS cases linked to 17p22 (BS type 1) or caused by mutations in PLOD2 (BS type 2) is still uncertain, and phenotypic correlations are lacking. We report on a boy who had congenital contractures with pterygia at birth and severe 0I-like osteopenia and multiple frac-tures. His urine contained high amounts of hydroxyproline but low amounts of collagen crosslinks degradation products; and he was shown to be homozygous for a novel mutation leading to an Arg598His substitution in PLOD2. The mutation is adjacent to the two mutations previously reported (Gly601Val and Thr608Ile), suggesting a functionally important hotspot in PLOD2. The combination of pterygia with bone fragility, as illustrated by this case, is difficult to explain; it suggests that telopeptide lysyl hydroxylation must be involved in prenatal joint formation and morphogenesis. Collagen degradation products in urine and mutation analysis ofPLOD2 maybe used to diagnose BS and differentiate it from M.

Changing paradigms--an update on the multidisciplinary management of malignant glioma.

Treatment of malignant glioma requires a multidisciplinary team. Treatment includes surgery, radiotherapy, and chemotherapy. Recently developed agents have demonstrated activity against recurrent malignant glioma and efficacy if given concurrently with radiotherapy in the upfront setting. Oligodendroglioma with 1p/19q deletions has been recognized as a distinct pathologic entity with particular sensitivity to radiotherapy and chemotherapy. Randomized trials have shown that early neoadjuvant or adjuvant administration of procarbazine, lomustine, and vincristine chemotherapy prolongs disease-free survival; however, it has no impact on overall survival. Temozolomide, a novel alkylating agent, has shown modest activity against recurrent glioma. In combination with radiotherapy in newly diagnosed patients with glioblastoma, temozolomide significantly prolongs survival. Molecular studies have demonstrated that the benefit is mainly observed in patients whose tumors have a methylated methylguanine methyltransferase gene promoter and are thus unable to repair some of the chemotherapy-induced DNA damage. For lower-grade glioma, the use of chemotherapy remains limited to recurrent disease, and first-line administration is the subject of ongoing clinical trials. Irinotecan and agents like gefitinib, erlotinib, and imatinib targeting the epidermal growth factor receptor and platelet-derived growth factor receptor have shown some promise in recurrent malignant glioma. This review summarizes recent developments, focusing on the clinical management of patients in daily neuro-oncology practice.

Analysis of a boundary protein delimiting chromatin domains in yeast

SUMMARY The expression state of a eukaryotic gene depends in part on its location in the chromosome. This position effect results from the organization of eukaryotic genomes into discrete functional domains, defined by local differences in chromatin structure. The expression of genes within each domain appears to be defined and maintained by the concerted action of regulatory elements such as promoters, enhancers, silencers and locus control regions. Individual domains may be bordered by boundary elements that separate regions of permissive and silent chromatin. When located next to chromosomal elements such as telomeres, genes can be subjected to epigenetic silencing. In yeast, this is mediated by the propagation of the SIR proteins from telomeres towards more centromeric regions. Particular transcription factors can protect downstream genes from silencing when tethered between the gene and the telomere, and they may thus act as chromatin domain boundaries. Here we have studied one of these transcription factors, CTF-1, that binds directly histone H3. A deletion mutagenesis localized the barrier activity to CTF-1 histone-binding domain. A saturating point mutagenesis of this domain identified several amino-acid substitutions that similarly inhibited the boundary and histone-binding activities. Chromatin immunoprecipitation experiments indicated that the barrier protein efficiently prevents the spreading of SIR proteins, and that it separates domains of hypoacetylated and hyperacetylated histones. Together, these results suggest a mechanism by which proteins such as CTF-1 may interact directly with histone H3 to prevent the propagation of a silent chromatin structure, thereby defining boundaries of permissive and silent chromatin domains. RESUME L'expression des gènes eucaryotes dépend en partie de leur localisation sur les chromosomes. Cet effet de position résulte de l'organisation des génomes eucaryotes en domaines fonctionnels, définis par des changements locaux au niveau de la structure de la chromatine. Dans chacun de ces domaines, l'expression des gènes est définie et maintenue par l'action concertée de différents éléments régulateurs tels que les promoteurs, les amplificateurs, les silenceurs et les locus control régions. Ces domaines peuvent être entourés par des éléments barrière, séparant les régions de chromatine répressive des régions permissive pour l'expression des gènes. Lorsqu'ils se situent à proximité d'éléments chromosomiques comme les telomères, les gènes peuvent être réprimés de manière épigénétique. Chez la levure, cette répression est établie par la propagation des protéines SIR depuis les télomères vers les régions centromériques. Certains facteurs de transcription peuvent empêcher la répression d'un gène, lorsqu'ils sont placés entre ce gène et le télomère. Nous avons étudié un de ces facteurs, CTF-1, qui a la particularité de lier directement l'histone H3. La délétion de certaines parties de CTF-1 a permis de déterminer que la région responsable de l'activité barrière correspond au domaine d'interaction avec H3. Plusieurs mutations points effectuées dans ce domaine inhibent à la fois l'activité barrière et la capacité de lier H3. Des expériences d'immuno-précipitation de la chromatine indiquent que la protéine barrière CTF-1 prévient efficacement la propagation des protéines SIR et sépare des domaines contenant des histones hypo-acétylées de ceux constitués d'histones hyper-acétylées. Ces résultats suggèrent que CTF-1 interagit directement avec l'histone H3 pour empêcher la propagation de la chromatine répressive, délimitant ainsi des domaines de chromatine permissive et des domaines de chromatine silencieuse.

An immuno-electron microscopical analysis of transcribing multinucleosomal templates: what happens to the histones?

Immuno-electron microscopy was used to visualize the structure of reconstituted chromatin after in vitro transcription by purified T7 RNA polymerase. T7 RNA polymerase disrupts the nucleosomal structure in the transcribed region. This disruption is not influenced by the template, linear or supercoiled, and the presence or absence of nucleosomal positioning sequences in the transcribed region. In this study, we used monoclonal autoantibodies reacting with the nucleosome core particles and epitopes within several regions of the four different core histones. Some of the residues recognized by the autoantibodies are accessible on the surface of the nucleosomes and some are more internal and therefore less exposed at the surface. We show that the loss of the nucleosomal configuration during transcription is due to the loss of histone/DNA binding and that at least part of the histones are transferred to the nascent RNA chains. Consequently, after in vitro transcription by T7 RNA polymerase, the nucleosomal template does not conserve its original configuration, and no interaction of antigen/antibodies is observed anymore in the region that has been transcribed. Therefore, we conclude that in our in vitro transcription assay, nucleosomes are detached from the template, and not simply unfolded with histones remaining attached to the DNA.

Critical role of the transcriptional repressor neuron-restrictive silencer factor in the specific control of connexin36 in insulin-producing cell lines.

Connexin36 (Cx36) is specifically expressed in neurons and in pancreatic beta-cells. Cx36 functions as a critical regulator of insulin secretion and content in beta-cells. In order to identify the molecular mechanisms that control the beta-cell expression of Cx36, we initiated the characterization of the human 5' regulatory region of the CX36 gene. A 2043-bp fragment of the human CX36 promoter was identified from a human BAC library and fused to a luciferase reporter gene. This promoter region was sufficient to confer specific expression to the reporter gene in insulin-secreting cell lines. Within this 5' regulatory region, a putative neuron-restrictive silencer element conserved between rodent and human species was recognized and binds the neuron-restrictive silencing factor (NRSF/REST). This factor is not expressed in insulin-secreting cells and neurons; it functions as a potent repressor through the recruitment of histone deacetylase to the promoter of neuronal genes. The NRSF-mediated repression of Cx36 in HeLa cells was abolished by trichostatin A, confirming the functional importance of histone deacetylase activity. Ectopic expression, by viral gene transfer, of NRSF/REST in different insulin-secreting beta-cell lines induced a marked reduction in Cx36 mRNA and protein content. Moreover, mutations in the Cx36 neuron-restrictive silencer element relieved the low transcriptional activity of the human CX36 promoter observed in HeLa cells and in INS-1 cells expressing NRSF/REST. The data showed that cx36 gene expression in insulin-producing beta-cell lines is strictly controlled by the transcriptional repressor NRSF/REST indicating that Cx36 participates to the neuronal phenotype of the pancreatic beta-cells.

Prolonged survival with valproic acid use in the EORTC/NCIC temozolomide trial for glioblastoma.

Objective: This analysis was performed to assess whether antiepileptic drugs (AEDs) modulate the effectiveness of temozolomide radiochemotherapy in patients with newly diagnosed glioblastoma.Methods: The European Organization for Research and Treatment of Cancer (EORTC) 26981-22981/National Cancer Institute of Canada (NCIC) CE.3 clinical trial database of radiotherapy (RT) with or without temozolomide (TMZ) for newly diagnosed glioblastoma was examined to assess the impact of the interaction between AED use and chemoradiotherapy on survival. Data were adjusted for known prognostic factors.Results: When treatment began, 175 patients (30.5%) were AED-free, 277 (48.3%) were taking any enzyme-inducing AED (EIAED) and 135 (23.4%) were taking any non-EIAED. Patients receiving valproic acid (VPA) only had more grade 3/4 thrombopenia and leukopenia than patients without an AED or patients taking an EIAED only. The overall survival (OS) of patients who were receiving an AED at baseline vs not receiving any AED was similar. Patients receiving VPA alone (97 [16.9%]) appeared to derive more survival benefit from TMZ/RT (hazard ratio [HR] 0.39, 95% confidence interval [CI] 0.24-0.63) than patients receiving an EIAED only (252 [44%]) (HR 0.69, 95% CI 0.53-0.90) or patients not receiving any AED (HR 0.67, 95% CI 0.49-0.93). Conclusions: VPA may be preferred over an EIAED in patients with glioblastoma who require an AED during TMZ-based chemoradiotherapy. Future studies are needed to determine whether VPA increases TMZ bioavailability or acts as an inhibitor of histone deacetylases and thereby sensitizes for radiochemotherapy in vivo.

Differential effects of selective HDAC inhibitors on macrophage inflammatory responses to the Toll-like receptor 4 agonist LPS.

Broad-spectrum inhibitors of HDACs are therapeutic in many inflammatory disease models but exacerbated disease in a mouse model of atherosclerosis. HDAC inhibitors have anti- and proinflammatory effects on macrophages in vitro. We report here that several broad-spectrum HDAC inhibitors, including TSA and SAHA, suppressed the LPS-induced mRNA expression of the proinflammatory mediators Edn-1, Ccl-7/MCP-3, and Il-12p40 but amplified the expression of the proatherogenic factors Cox-2 and Pai-1/serpine1 in primary mouse BMM. Similar effects were also apparent in LPS-stimulated TEPM and HMDM. The pro- and anti-inflammatory effects of TSA were separable over a concentration range, implying that individual HDACs have differential effects on macrophage inflammatory responses. The HDAC1-selective inhibitor, MS-275, retained proinflammatory effects (amplification of LPS-induced expression of Cox-2 and Pai-1 in BMM) but suppressed only some inflammatory responses. In contrast, 17a (a reportedly HDAC6-selective inhibitor) retained anti-inflammatory but not proinflammatory properties. Despite this, HDAC6(-/-) macrophages showed normal LPS-induced expression of HDAC-dependent inflammatory genes, arguing that the anti-inflammatory effects of 17a are not a result of inhibition of HDAC6 alone. Thus, 17a provides a tool to identify individual HDACs with proinflammatory properties.

Impact of genomic methylation on radiation sensitivity of colorectal carcinoma.

PURPOSE: To investigate the influence of demethylation with 5-aza-cytidine (AZA) on radiation sensitivity and to define the intrinsic radiation sensitivity of methylation deficient colorectal carcinoma cells. METHODS AND MATERIALS: Radiation sensitizing effects of AZA were investigated in four colorectal carcinoma cell lines (HCT116, SW480, L174 T, Co115), defining influence of AZA on proliferation, clonogenic survival, and cell cycling with or without ionizing radiation. The methylation status for cancer or DNA damage response-related genes silenced by promoter methylation was determined. The effect of deletion of the potential target genes (DNMT1, DNMT3b, and double mutants) on radiation sensitivity was analyzed. RESULTS: AZA showed radiation sensitizing properties at >or=1 micromol/l, a concentration that does not interfere with the cell cycle by itself, in all four tested cell lines with a sensitivity-enhancing ratio (SER) of 1.6 to 2.1 (confidence interval [CI] 0.9-3.3). AZA successfully demethylated promoters of p16 and hMLH1, genes associated with ionizing radiation response. Prolonged exposure to low-dose AZA resulted in sustained radiosensitivity if associated with persistent genomic hypomethylation after recovery from AZA. Compared with maternal HCT116 cells, DNMT3b-defcient deficient cells were more sensitive to radiation with a SER of 2.0 (CI 0.9-2.1; p = 0.03), and DNMT3b/DNMT1-/- double-deficient cells showed a SER of 1.6 (CI 0.5-2.7; p = 0.09). CONCLUSIONS: AZA-induced genomic hypomethylation results in enhanced radiation sensitivity in colorectal carcinoma. The mediators leading to sensitization remain unknown. Defining the specific factors associated with radiation sensitization after genomic demethylation may open the way to better targeting for the purpose of radiation sensitization.

Lithiase cystinique [Cysteine lithiasis].

Cystinuria is a common inherited amino-aciduria resulting in abnormal urinary excretion of cystine and the dibasic aminoacids, lysine, arginine and ornithine. Formation of cystine kidney stones, recurrent infections and subsequent renal failure are the main complications of the disease. Recently, the gene SLC3A1 and SLC7A9, encoding the two subunits rBAT et b0,+AT of the proximal renal transporter complex, have been identified. In this article, we report the medical history of a 30-year-old patient and discuss the recent molecular progress, the clinical evolution, and the medical treatment of the cystinuria.

Epithelioid sarcoma: a clinicopathologic and immunohistochemical analysis of 106 cases from the French sarcoma group.

Epithelioid sarcoma (ES) is rare with a poor prognosis and for which a loss of INI1 expression has been recently reported. We report a study of 106 cases with clinical, histologic, and immunohistochemical data, including INI1 expression, and follow-up data. Of the 106 cases, 70 were the conventional subtype and 36 the large cell subtype. INI1 was negative in 86 cases (81.1%): 57 (81%) of 70 conventional and 29 (81%) of 36 large cell subtypes. Treatment modalities were available for 76 and follow-up for 80 patients. Of the 80 patients, 43 (54%) experienced metastasis and 25 (31%) died of the disease. Univariate analysis indicated that tumor size and mitotic index were significant for metastasis-free survival, whereas proximal location, tumor size, tumor multifocality, and mitotic index were significant for overall survival. Loss of expression of INI1 is frequent in the conventional and large cell subtypes of ES and can be used as a diagnostic marker, but it has no prognostic impact.

The repressor element silencing transcription factor (REST)-mediated transcriptional repression requires the inhibition of Sp1.

The terminal differentiation of neuronal and pancreatic beta-cells requires the specific expression of genes that are targets of an important transcriptional repressor named RE-1 silencing transcription factor (REST). The molecular mechanism by which these REST target genes are expressed only in neuronal and beta-cells and are repressed by REST in other tissues is a central issue in differentiation program of neuronal and beta-cells. Herein, we showed that the transcriptional factor Sp1 was required for expression of most REST target genes both in insulin-secreting cells and neuronal-like cells where REST is absent. Inhibition of REST in a non-beta and a non-neuronal cell model restored the transcriptional activity of Sp1. This activity was also restored by trichostatin A indicating the requirement of histone deacetylases for the REST-mediated silencing of Sp1. Conversely, exogenous introduction of REST blocked Sp1-mediated transcriptional activity. The REST inhibitory effect was mediated through its C-terminal repressor domain, which could interact with Sp1. Taken together, these data show that the inhibition of Sp1 by REST is required for the silencing of its target genes expression in non-neuronal and in non-beta-cells. We conclude that the interplay between REST and Sp1 determines the cell-specific expression of REST target genes.

Vasopressin dilates the rat carotid artery by stimulating V1 receptors.

The acute effects of various vasopressor agents on the diameter of the common carotid artery were studied in halothane-anesthetized normotensive rats. The animals were infused intravenously for 60 min with equipressor doses of angiotensin II (10 ng/min), the alpha1-stimulant methoxamine (5 microg/min), lysine vasopressin (5 mU/min), or vehicle. The arterial diameter was measured by using a high-resolution ultrasonic echo-tracking device. The three vasoconstrictors increased the carotid artery diameter, but this effect was significantly more pronounced with lysine vasopressin. Even a nonpressor dose of lysine vasopressin (1 mU/min) caused a significant increase in the arterial diameter. The lysine vasopressin-induced vasodilatation could be prevented by the administration of d(CH2)5Tyr(Me)AVP (10 microg, i.v.), a selective V1-vasopressinergic receptor antagonist. These data therefore suggest that a short-term increase in blood pressure induces in rats a distention of the carotid artery. The increase in arterial diameter seems to involve an active mechanism with lysine vasopressin caused by the stimulation of V1-vasopressinergic receptors.

Several neuronal and axonal types form long intrinsic connections in the cat primary auditory cortical field (AI).

Intrinsic connections in the cat primary auditory field (AI) as revealed by injections of Phaseolus vulgaris leucoagglutinin (PHA-L) or biocytin, had an anisotropic and patchy distribution. Neurons, labelled retrogradely with PHA-L were concentrated along a dorsoventral stripe through the injection site and rostral to it; the spread of rostrally located neurons was greater after injections into regions of low rather than high characteristic frequencies. The intensity of retrograde labelling varied from weak and granular to very strong and Golgi-like. Out of 313 Golgi like retrogradely labelled neurons 79.6% were pyramidal, 17.2% multipolar, 2.6% bipolar, and 0.6% bitufted; 13.4% were putatively inhibitory, i.e. aspiny or sparsely spiny multipolar, or bitufted. Individual anterogradely labelled intrinsic axons were reconstructed for distances of 2 to 7 mm. Five main types were distinguished on the basis of the branching pattern and the location of synaptic specialisations. Type 1 axons travelled horizontally within layers II to VI and sent collaterals at regular intervals; boutons were only present in the terminal arborizations of these collaterals. Type 2 axons also travelled horizontally within layers II to VI and had rather short and thin collateral branches; boutons or spine-like protrusions occurred in most parts of the axon. Type 3 axons travelled obliquely through the cortex and formed a single terminal arborization, the only site where boutons were found. Type 4 axons travelled for some distance in layer I; they formed a heterogeneous group as to their collaterals and synaptic specializations. Type 5 axons travelled at the interface between layer VI and the white matter; boutons en passant, spine-like protrusions, and thin short branches with boutons en passant were frequent all along their trajectory. Thus, only some axonal types sustain the patchy pattern of intrinsic connectivity, whereas others are involved in a more diffuse connectivity.