Interaction of nerve growth factor with murine alpha-macroglobulin

Platelets Selectively Regulate the Release of BDNF, But Not That of Its Precursor Protein, proBDNF

Background

Brain-derived neurotrophic factor (BDNF) plays a role in synaptic plasticity and neuroprotection. BDNF has well-established pro-survival effects, whereas its precursor protein, proBDNF, induces apoptosis. Thus, it has been suggested that the proBDNF/BDNF ratio could be an indicator of neuronal health. Access to neurons is, understandably, limited. Because of their similarities, platelets have been put forward as a non-invasive biomarker of neuronal health; indeed, they store large quantities of BDNF and can release it into circulation upon activation, similarly to neurons. However, whether platelets also express the precursor proBDNF protein remains unknown. We therefore sought to characterize proBDNF levels in human platelets and plasma.

Methods

The presence of proBDNF was assessed by immunoblotting, cell fractionation, flow cytometry, and confocal microscopy in washed platelets from 10 healthy volunteers. Platelets from 20 independent healthy volunteers were activated with several classical agonists and the release of BDNF and proBDNF into plasma was quantified by ELISA.

Results

Platelets expressed detectable levels of proBDNF (21 ± 13 fmol/250 x 10⁶ platelets). ProBDNF expression was mainly localized in the intracellular compartment. The proBDNF to BDNF molar ratio was ~1:5 in platelets and 10:1 in plasma. In stark contrast to the release of BDNF during platelet activation, intraplatelet and plasma concentrations of proBDNF remained stable following stimulation with classical platelet agonists, consistent with non-granular expression.

Conclusions

Platelets express both the mature and the precursor form of BDNF. Whether the intraplatelet proBDNF to BDNF ratio could be used as a non-invasive biomarker of cognitive health warrants further investigation.

Hepcidin, the hormone of iron metabolism, is bound specifically to alpha-2-macroglobulin in blood

Hepcidin is a major regulator of iron metabolism. Hepcidin-based therapeutics/diagnostics could play roles in hematology in the future, and thus, hepcidin transport is crucial to understand. In this study, we identify alpha2-macroglobulin (alpha2-M) as the specific hepcidin-binding molecule in blood. Interaction of 125I-hepcidin with alpha2-M was identified using fractionation of plasma proteins followed by native gradient polyacrylamide gel electrophoresis and mass spectrometry. Hepcidin binding to nonactivated alpha2-M displays high affinity (Kd 177 +/- 27 nM), whereas hepcidin binding to albumin was nonspecific and displayed nonsaturable kinetics. Surprisingly, the interaction of hepcidin with activated alpha2-M exhibited a classical sigmoidal binding curve demonstrating cooperative binding of 4 high-affinity (Kd 0.3 microM) hepcidin-binding sites. This property probably enables efficient sequestration of hepcidin and its subsequent release or inactivation that may be important for its effector functions. Because alpha2-M rapidly targets ligands to cells via receptor-mediated endocytosis, the binding of hepcidin to alpha2-M may influence its functions. In fact, the alpha2-M-hepcidin complex decreased ferroportin expression in J774 cells more effectively than hepcidin alone. The demonstration that alpha2-M is the hepcidin transporter could lead to better understanding of hepcidin physiology, methods for its sensitive measurement and the development of novel drugs for the treatment of iron-related diseases.

Functions of alpha 2 macroglobulins in pregnancy

The alpha 2 macroglobulins (A2M) are a family of abundant plasma proteins produced predominantly by the mammalian liver. Pregnancy zone proteins (PZP) of humans and rats are A2M family members that bind a wide variety of macromolecules including the important pregnancy-associated molecules such as vascular endothelial growth factor, placenta growth factor and glycodelin (also called PP14). Recently, a mouse gene analogous to PZP (A2M of pregnancy or A2Mp) was cloned. A2Mp has a unique pattern of expression in reproductive and cardiovascular tissues and, unexpectedly, is not expressed by liver. Since changes in heart function and remodeling of renal and uterine vasculature are amongst the earliest maternal responses to pregnancy, the product of the A2Mp gene has been postulated to systemically regulate these changes. A2Ms with and without non-covalently bound ligands also down regulate immune cell activation but promote immune cell migration, additional features associated with gestational success. Here, we review the A2M gene families of mice and humans, the predicted structural relationships between A2M and its pregnancy induced forms and the postulated roles for this gene family in normal pregnancy.

Pregnancy zone protein is a carrier and modulator of placental protein-14 in T-cell growth and cytokine production

A successful pregnancy can only occur when the maternal immune system fails to attack the allogeneic fetus. Two plasma proteins with described immunoregulatory activities, pregnancy zone protein (PZP) and placental protein-14 (PP14; also known as glycodelin-A), increase dramatically during pregnancy, prompting us to examine their potential role in mediating fetal protection. First, we demonstrated that both native PZP and its receptor-recognized monoamine-activated form (MA-PZP) bound non-covalently and specifically to PP14, exhibiting K(d) values greater than 3 microM, as determined by surface plasmon resonance. Our evidence further suggests that PZP is potentially a more effective carrier of PP14 than its relative alpha2-macroglobulin. Second, we found that T-cell activation, as measured by increased proliferation and IL-2 production, was inhibited by either PZP or PP14 in a dose-dependent manner. However, when PZP and PP14 were combined, they acted synergistically to inhibit T cell proliferation and IL-2 production. Interestingly, the combination of PZP and PP14 had little effect on the production of T(H)2 cytokine, IL-4. Based upon these findings, we hypothesize that PZP and PP14 form a stable complex in the plasma of pregnant women and together act synergistically to selectively modulate T-cell activation. Mechanistically, this activity appears to be independent of the PZP receptor (CD91) or PZP's anti-proteinase activity.

Replicate high-density rat genome oligonucleotide microarrays reveal hundreds of regulated genes in the dorsal root ganglion after peripheral nerve injury

BACKGROUND

Rat oligonucleotide microarrays were used to detect changes in gene expression in the dorsal root ganglion (DRG) 3 days following sciatic nerve transection (axotomy). Two comparisons were made using two sets of triplicate microarrays, naïve versus naïve and naïve versus axotomy.

RESULTS

Microarray variability was assessed using the naïve versus naïve comparison. These results support use of a P < 0.05 significance threshold for detecting regulated genes, despite the large number of hypothesis tests required. For the naïve versus axotomy comparison, a 2-fold cut off alone led to an estimated error rate of 16%; combining a >1.5-fold expression change and P < 0.05 significance reduced the estimated error to 5%. The 2-fold cut off identified 178 genes while the combined >1.5-fold and P < 0.05 criteria generated 240 putatively regulated genes, which we have listed. Many of these have not been described as regulated in the DRG by axotomy. Northern blot, quantitative slot blots and in situ hybridization verified the expression of 24 transcripts. These data showed an 83% concordance rate with the arrays; most mismatches represent genes with low expression levels reflecting limits of array sensitivity. A significant correlation was found between actual mRNA differences and relative changes between microarrays (r2 = 0.8567). Temporal patterns of individual genes regulation varied.

CONCLUSIONS

We identify parameters for microarray analysis which reduce error while identifying many putatively regulated genes. Functional classification of these genes suggest reorganization of cell structural components, activation of genes expressed by immune and inflammatory cells and down-regulation of genes involved in neurotransmission.

Low-density lipoprotein receptor-related protein mediates in PC12 cell cultures the inhibition of nerve growth factor-promoted neurite outgrowth by pregnancy zone protein and alpha2-macroglobulin

Human pregnancy zone protein (PZP) is a major pregnancy-associated plasma protein closely related to human alpha(2)-macroglobulin (alpha(2)M). It has been demonstrated that monoamine-activated forms of human and rat alpha(2)M and rat alpha(1)M can bind to TrkA and, respectively, inhibit and stimulate NGF-promoted neurite outgrowth, Trk phosphorylation, and intracellular signal transduction in PC12 cells. However, the effect of PZP on neurons is unknown, and the molecular mechanism of neuroinhibition by monoamine-activated alpha(2)M is still unclear. In this report, we show that methylamine-activated PZP (MA-PZP), like MA-alpha(2)M, inhibits in a dose-dependent way the NGF-promoted neurite extension and TrkA phosphorylation in PC12 cells. On the other hand, normal PZP (N-PZP) had little or no effect. In addition, the inhibitory effect of activated alpha-macroglobulins (alphaMs) was reversible upon its removal from the cell culture. In addition, PZP, as well as alpha(2)M, is neuroinhibitory without being directly cytotoxic. It is known that the activated alphaMs bind to the multiligand receptor termed low-density lipoprotein receptor-related protein (LRP) and that the receptor-associated protein (RAP) specifically blocks uptake of all known LRP ligands. To investigate the potential role of LRP in neuromodulation by activated PZP/alpha(2)M, the effect of RAP on the neuroinhibitory activities of these alphaMs was also studied. Data presented here show that RAP blocked the neurite- and Trk-inhibitory activities of both MA-PZP and MA-alpha(2)M, whereas RAP itself had no neuromodulatory effect. Hence, we conclude that these data suggest that the LRP receptor and its alphaM ligands may play a role in regulating Trk receptors.

Delivery of neurotrophic factors to the central nervous system: pharmacokinetic considerations

Neurotrophic factors are proteins with considerable potential in the treatment of central nervous system (CNS) diseases and traumatic injuries. However, a significant challenge to their clinical use is the difficulty associated with delivering these proteins to the CNS. Neurotrophic factors are hydrophilic, typically basic, monomeric or dimeric proteins, mostly in the size range of 5 to 30 kDa. Neurotrophic factors potently support the development, growth and survival of neurons, eliciting biological effects at concentrations in the nanomolar to femtomolar range. They are not orally bioavailable and the blood-brain and blood-cerebrospinal fluid barriers severely limit their ability to enter into and act on sites in the CNS following parenteral systemic routes of administration. Most neurotrophic factors have short in vivo half-lives and poor pharmacokinetic profiles. Their access to the CNS is restricted by rapid enzymatic inactivation, multiple clearance processes, potential immunogenicity and sequestration by binding proteins and other components of the blood and peripheral tissues. The development of targeted drug delivery strategies for neurotrophic factors will probably determine their clinical effectiveness for CNS conditions. Achieving significant CNS target site concentrations while limiting systemic exposure and distribution to peripheral sites of action will lessen unwanted pleiotropic effects and toxicity. Local introduction of neurotrophic factors into the CNS intraparenchymally by direct injection/infusion or by implantation of delivery vectors such as polymer matrices or genetically modified cells yields the highest degree of targeting, but is limited by diffusion restrictions and invasiveness. Delivery of neurotrophic factors into the cerebrospinal fluid (CSF) following intracerebroventricular or intrathecal administration is less invasive and allows access to a much wider area of the CNS through CSF circulation pathways. However, diffusional and cellular barriers to penetration into surrounding CNS tissue and significant clearance of CSF into the venous and lymphatic circulation are also limiting. Unconventional delivery strategies such as intranasal administration may offer some degree of CNS targeting with minimal invasiveness. This review presents a summary of the neurotrophic factors and their indications for CNS disorders, their physicochemical characteristics and the different approaches that have been attempted or suggested for their delivery to the CNS. Future directions for further research such as the potential for CNS disease treatment utilising combinations of neurotrophic factors, displacement strategies, small molecule mimetics, chimaeric molecules and gene therapy are also discussed.

Comparative binding of biotinylated neurotrophins to alpha(2)-macroglobulin family of proteins: relationship between cytokine-binding and neuro-modulatory activities of the macroglobulins

Human alpha(2)-macroglobulin (alpha(2)M), pregnancy zone protein (PZP), rat alpha(1)M and acute-phase rat alpha(2)M belong to the alpha(2)M gene family of proteins, which can react covalently with nucleophilic monoamines to yield monoamine-activated (MA) macroglobulins. The MA forms of human alpha(2)M, PZP and rat alpha(2)M have been demonstrated previously to inhibit various neurotrophin-promoted neuronal activities, whereas MA-alpha(1)M is neurostimulatory and all native macroglobulins are generally inactive. The mechanism of neuromodulation is unknown, but it has been postulated that MA macroglobulins might inhibit neurons via their binding and sequestration of neurotrophins. This study employed a novel biotinylation-Western blot technique to compare the neurotrophin-binding properties of the four macroglobulins, and to correlate their binding activities with their known neuro-modulatory activities. In comparison with their respective native counterparts, human and rat MA-alpha(2)M bound slightly more NGF, but significantly less BDNF or NT-3. Native human alpha(2)M and PZP in general have no neuro-modulatory activity, but native PZP bound significantly more NGF, BDNF or NT-3 than either native alpha(2)M or MA-alpha(2)M, which is neuro-inhibitory. It is known that MA-PZP is neuro-inhibitory, but it fails to bind more NGF, BDNF, or NT-3 than native PZP. MA-alpha(1)M is the only macroglobulin known to stimulate NGF-promoted neurite outgrowth, but it bound NGF with similar affinities as native alpha(1)M and rat alpha(2)M; in addition, it bound significantly less BDNF or NT-3 than native alpha(1)M. All the bindings were non-covalent and appeared specific. In conclusion, PZP and rat macroglobulins are versatile carriers of neurotrophins with diverse binding capacities, and the neurotrophin-binding property does not appear to mediate the neuro-modulatory activity of these human and rat macroglobulins.

Identification of the high affinity binding site in transforming growth factor-beta involved in complex formation with alpha 2-macroglobulin. Implications regarding the molecular mechanisms of complex formation between alpha 2-macroglobulin and growth factors, cytokines, and hormones

The biological activities of transforming growth factor-beta isoforms (TGF-beta(1,2)) are known to be modulated by alpha(2)-macroglobulin (alpha(2)M). alpha(2)M forms complexes with numerous growth factors, cytokines, and hormones, including TGF-beta. Identification of the binding sites in TGF-beta isoforms responsible for high affinity interaction with alpha(2)M many unravel the molecular basis of the complex formation. Here we demonstrate that among nine synthetic pentacosapeptides with overlapping amino acid sequences spanning the entire TGF-beta(1) molecule, the peptide (residues 41-65) containing Trp-52 exhibited the most potent activity in inhibiting the formation of complexes between (125)I-TGF-beta(1) and activated alpha(2)M (alpha(2)M*) as determined by nondenaturing polyacrylamide gel electrophoresis and by plasma clearance in mice. TGF-beta(2) peptide containing the homologous sequence and Trp-52 was as active as the TGF-beta(1) peptide, whereas the corresponding TGF-beta(3) peptide lacking Trp-52, was inactive. The replacement of the Trp-52 with alanine abolished the inhibitory activities of these peptides. (125)I-TGF-beta(3), which lacks Trp-52, bound to alpha(2)M* with an affinity lower than that of (125)I-TGF-beta(1). Furthermore, unlabeled TGF-beta(3) and the mutant TGF-beta(1)W52A, in which Trp-52 was replaced with alanine, were less potent than unlabeled TGF-beta(1) in blocking I(125)-TGF-beta(1) binding to alpha(2)M*. TGF-beta(1) and TGF-beta(2) peptides containing Trp-52 were also effective in inhibiting I(125)-nerve growth factor binding to alpha(2)M*. Tauhese results suggest that Trp-52 is involved in high affinity binding of TGF-beta to alpha(2)M*. They also imply that TGF-beta and other growth factors/cytokines/hormones may form complexes with alpha(2)M* via a common mechanism involving the interactions between topologically exposed Trp and/or other hydrophobic residues and a hydrophobic region in alpha(2)M*.

Rat alpha1-macroglobulin enhances nerve growth factor-promoted neurite outgrowth, TrkA phosphorylation, and gene expression of pheochromocytoma PC12 cells

Monoamine-activated human alpha2-macroglobulin (alpha2M) has been previously demonstrated to inhibit TrkA-, TrkB-, and TrkC-mediated signal transduction. Rat alpha1-macroglobulin (alpha1M) and alpha2M are structural homologues of human alpha2M, but rat alpha1M is distinctly different from rat alpha2M in many ways and its role in the mammalian nervous system is unknown. In this report, monoamine-activated rat alpha1M was demonstrated to enhance in a dose-dependent manner nerve growth factor (NGF)-promoted neurite outgrowth in pheochromocytoma PC12 cells. Monoamine-activated alpha1M by itself, however, was neither neurotrophic nor mitogenic to PC12 cells. To investigate further its possible mode of action, the ability of monoamine-activated alpha1M and normal alpha1M to bind and to activate the NGF receptor (TrkA) was investigated. Monoamine-activated alpha1M formed a more stable complex with TrkA than normal alpha1 M, but the binding of monoamine-activated alpha1M to TrkA was adversely affected by prior stimulation of TrkA with NGF. In addition, monoamine-activated alpha1M enhanced the NGF-promoted TrkA phosphorylation and up-regulated the expression of NGF-inducible immediate-early genes (c-jun and NGFI-A) and delayed-response genes (SCG10 and transin) in PC12 cells; normal alpha1M, in contrast, produced little or no effect. This study demonstrates that alpha1M, the constitutive form of alpha-macroglobulin in the rat, possesses the ability to promote NGF-mediated differentiation in PC12 cells, possibly via its direct action on TrkA receptors and TrkA-mediated signal transduction and gene expression.

Oxidized α2-Macroglobulin (α2M) Differentially Regulates Receptor Binding by Cytokines/Growth Factors: Implications for Tissue Injury and Repair Mechanisms in Inflammation

α2M binds specifically to TNF-α, IL-1β, IL-2, IL-6, IL-8, basic fibroblast growth factor (bFGF), β-nerve growth factor (β-NGF), platelet-derived growth factor (PDGF), and TGF-β. Since many of these cytokines are released along with neutrophil-derived oxidants during acute inflammation, we hypothesize that oxidation alters the ability of α2M to bind to these cytokines, resulting in differentially regulated cytokine functions. Using hypochlorite, a neutrophil-derived oxidant, we show that oxidized α2M exhibits increased binding to TNF-α, IL-2, and IL-6 and decreased binding to β-NGF, PDGF-BB, TGF-β1, and TGF-β2. Hypochlorite oxidation of methylamine-treated α2M (α2M*), an analogue of the proteinase/α2M complex, also results in decreased binding to bFGF, β-NGF, PDGF-BB, TGF-β1, and TGF-β2. Concomitantly, we observed decreased ability to inhibit TGF-β binding and regulation of cells by oxidized α2M and α2M*. We then isolated α2M from human rheumatoid arthritis synovial fluid and showed that the protein is extensively oxidized and has significantly decreased ability to bind to TGF-β compared with α2M derived from plasma and osteoarthritis synovial fluid. We, therefore, propose that oxidation serves as a switch mechanism that down-regulates the progression of acute inflammation by sequestering TNF-α, IL-2, and IL-6, while up-regulating the development of tissue repair processes by releasing bFGF, β-NGF, PDGF, and TGF-β from binding to α2M.

Reversible sedimentation and masking of nerve growth factor (NGF) antigen by high molecular weight fractions from rat brain

Nerve growth factor (NGF) was recently found to be largely associated with sedimentable fractions of adult rat brain and treatments of the fractions by alkaline pH increased the measurable amount of their NGF antigen as well as its solubilization [M.C. Hoener, E. Hewitt, J.M. Conner, J.W. Costello and S. Varon, Nerve growth factor (NGF) content in adult rat brain tissues is several-fold higher than generally reported and is largely associated with sedimentable fractions, Brain Res., 728 (1996) 47-56; M.C. Hoener and S. Varon, Effects of sodium chloride, Triton X-100, and alkaline pH on the measurable contents and sedimentability of the nerve growth factor (NGF) antigen in adult rat hippocampal tissue extracts, J. Neurosci. Res., in press (1997); C. Zettler, D.C.McL. Bridges, X.-F. Zhou and R.A. Rush, Detection of increased tissue concentrations of nerve growth factor with improved extraction procedure, J. Neurosci. Res., 46 (1996) 581-594]. We have further investigated the reversibility of these pH effects. Reversal of the pH of an adult rat hippocampal tissue extract from 10.5 to 7.4 led to an almost complete transfer of NGF back from nonsedimentable to sedimentable fractions and to a remasking of the previously unmasked portion of NGF antigen. Thus, molecules causing masking and sedimentation of NGF at pH 7.4 were likely to be present in the alkaline extract. A gel filtration column in PBS, pH 10.5 was used to separate such putative binding molecules from the NGF. All of the NGF antigen from rat hippocampal alkaline extract was found to elute with 19 kDa fractions. The same apparent molecular weight was found for mouse submaxillary beta-NGF and recombinant human beta-NGF. Masking and sedimentation no longer occurred when newly generated 19 kDa rat brain NGF was returned to pH 7.4. When high molecular weight fractions derived from the same gel filtration (in PBS, pH 10.5) were added back to the 19 kDa NGF pool at pH 7.4 and the mixture incubated and centrifuged, the measurability of 19 kDa rat brain NGF antigen was markedly reduced and half of the antigen was recovered in sedimentable fractions. Similar but less dramatic results were obtained when mixing the same high molecular weight fractions with 19 kDa mouse or human beta-NGF. These findings provide new opportunities to identify molecules to which NGF may be bound within intact brain tissues.

Acute phase proteins and transformed cells

Acute phase proteins (APP) are plasma proteins whose concentration and glycosylation alters in response to tissue injury, inflammation, or tumor growth. Significant interspecies and sex differences in APP response exist. APP are produced mainly by hepatocytes, and their synthesis and glycosylation are controlled by a network consisting of cytokines, their soluble receptors, and glucocorticoids. The major cytokines involved in these processes belong to a group of interleukin-6-type cytokines that act through the hematopoietin receptor complex on hepatocytes and JAK-STAT signal transduction pathway. Transformed cells (hepatoma) display significant differences in synthesis of APP, cytokine responsiveness, expression of cytokine-receptor subunits and signal-transduction machinery. The most striking variability relates to the glycosylation alterations induced by cytokines. However, transformed cells (hepatoma) form a basic model for studying and understanding mechanisms controlling the synthesis and glycosylation of APP. Furthermore, APP may be secreted by transformed (tumor) cells of various origins and may display a growth factor-like function in certain cancer types.

Detection of increased tissue concentrations of nerve growth factor with an improved extraction procedure

Nerve growth factor (NGF) is a protein essential for the survival and normal function of sympathetic neurons. Two-site immunoassays have been developed over the past decade in several laboratories and used to estimate its endogenous concentrations in a variety of effector tissues. However, levels appear restricted to a narrow range, display only a poor correlation with innervation density, and show obvious inter- and intralaboratory variations, the origins of which are unclear. This led us to examine alternative extraction procedures for NGF before quantification. In particular, we have found treatment of tissue extracts with high and low pH in the presence of detergent results in the detection of higher NGF concentrations in immunoassays using either polyclonal or commercially available monoclonal antibodies. These increases were tissue-specific (sciatic nerve, mesenteric arteries, and thoracic aorta > heart and brain > sympathetic ganglia > abdominal aorta) and as much as 10 times greater than the amounts detected by traditional procedures. The method should also prove useful for the assay of other members of the neurotrophin family when appropriate antibodies become available.

Alteration of dopamine release by rat caudate putamen tissues superfused with alpha 2-macroglobulin

Monoamine-activated alpha-2-macroglobulin (alpha 2M) has been shown to decrease the dopamine concentrations in rat caudate putamen (CP) in vivo as well as inhibit choline acetyltransferase activities in the culture of basal forebrain neurons. In this study, we further investigated the effects of methylamine-activated alpha 2M (MA-alpha 2M) upon striatal dopaminergic function by determining whether a direct infusion of this glycoprotein will alter dopamine (DA) release in vitro from superfused CP tissue fragments. In experiment 1, an infusion of 2.8 microM MA-alpha 2M produced a statistically significant increase in DA release compared with control superfusions. In experiment 2, varying doses (0, 0.7, 1.4, 2.8, 4.1 microM) of MA-alpha 2M were tested for their capacity to alter DA release. Only the 2.8 microM dose of MA-alpha 2M was effective in producing a significant increase of DA release. In experiment 3, the normal form of alpha 2M (N-alpha 2M) at 2.8 microM was compared with the control superfusions. The infusion of N-alpha 2M produced an increase in DA release which was substantially lower than the DA increase induced by MA-alpha 2M, and not significantly different from that of the control superfusion. These results show that MA-alpha 2M, like some other neurotoxins, can markedly alter CP dopaminergic function as indicated by the acute increase in DA release following infusion of this glycoprotein, and these effects are exerted at a relatively narrow range of doses. Taken together, these data suggest that this glycoprotein, if allowed to accumulate in the central nervous system (CNS), may promote some neurodegenerative changes that can occur in disorders like Parkinson's disease.

Update of the NGF saga

Nerve growth factor (NGF), initially characterized for its survival and differentiating actions on embryonic sensory and sympathetic neurons, is now known to display a greatly extended spectrum of biological functions. NGF exerts a profound modulatory role on sensory nociceptive nerve physiology during adulthood which appears to correlate with hyperalgesic phenomena occurring in tissue inflammation. Other newly detected NGF-responsive cells belong to the hematopoietic-immune and neuroendocrine systems. In particular, mast cells and NGF both appear to be involved in neuroimmune interactions and tissue inflammation, with NGF acting as a general "alert" molecule capable of recruiting and priming both local tissue and systemic defense processes following stressful events. NGF can thus be viewed as a multifactorial mediator modulating neuroimmune-endocrine functions of vital importance to the regulation of homeostatic interactions, with potential involvement in pathological processes deriving from dysregulation of either local or systemic homeostatic balances.

Evidence for the binding of a biologically active interleukin-2 to human alpha 2-macroglobulin

Human alpha 2-macroglobulin (alpha 2M), which irreversibly entraps proteinases through a drastic conformational change, has also been reported to bind various cytokines. The meaning of cytokine binding to native and/or transformed alpha 2M molecules is, however, not understood. In an attempt to elucidate this question, we have studied the interaction of radioiodinated recombinant human interleukin-2 (125I-rhIL-2) with native and chymotrypsin (alpha 2M-C)- or methylamine-transformed (alpha 2M-MA) alpha 2M. Our results show that native and alpha 2M-MA are able to bind 125I-rhIL-2, with binding occurring only with the latter in a covalent manner, whereas the labeled cytokine is proteolyzed when incubated with alpha 2M-entrapped chymotrypsin. The degradation of uncomplexed 125I-rhIL-2 has also been observed in the presence of trypsin, whereas 125I-rhIL-2 bound to alpha 2M-MA is protected. Moreover, the proliferative activity of this cytokine on responsive cells is still maintained either with native alpha 2M- or alpha 2M-MA-complexed rhIL-2 in comparison with that observed with the cytokine alone. Our results, which lead us to consider alpha 2M molecules as IL-2-binding proteins, emphasize the possible role of these molecules as immune response regulators.

Inhibition of platelet-derived growth factor-BB-induced fibroblast proliferation by plasmin-activated alpha 2-macroglobulin is mediated via an alpha 2-macroglobulin receptor/low density lipoprotein receptor-related protein-dependent mechanism

alpha 2-Macroglobulin (alpha 2M) is a potentially important regulator of platelet-derived growth factor-BB (PDGF-BB)-stimulated cell growth due to our previous observation that PDGF-BB binds to alpha 2M noncovalently (Bonner, J. C., Goodell, A. L., Lasky, J. A., and Hoffman, M. R. (1992) J. Biol. Chem. 267, 12837-12844). We examined the in vitro effect of native and plasmin-activated (receptor-recognized) alpha 2M on the PDGF-BB-induced proliferation of mouse Swiss 3T3 and rat lung fibroblasts. Nondenaturing polyacrylamide gel electrophoresis showed that plasmin converted alpha 2M to its electrophoretically "fast" form at a 2:1 molar ratio and that 125I-PDGF-BB bound both alpha 2M and alpha 2M-plasmin. PDGF-BB-induced growth was not affected by native alpha 2M (0.3 microM) or plasmin (0.6 microM). The combination of plasmin and alpha 2M (2:1 molar ratio) inhibited PDGF-BB-induced cell proliferation 80-90%. Complexes of PDGF-BB.alpha 2M purified by gel filtration chromatography retained growth promoting activity, but the PDGF-BB.alpha 2M-plasmin complex did not. Preincubation of fibroblasts (37 degrees C for 24 h) with alpha 2M-plasmin did not change 125I-PDGF-BB binding or affect gene expression of the 6.5-kilobase PDGF-alpha receptor or 5.2-kilobase PDGF-beta receptor mRNA. However, preincubation with alpha 2M-plasmin (0-4 degrees C for 4 h) increased 125I-PDGF-BB binding 2-fold, and this increase was blocked by a receptor-associated protein antagonist of the alpha 2M-receptor/low density lipoprotein receptor-related protein. The receptor-associated protein antagonist blocked 125I-alpha 2M-methylamine binding, inhibited PDGF-BB-alpha 2M-plasmin uptake from fibroblast-cultured supernatants, and abolished the inhibitory effect of alpha 2M-plasmin on PDGF-stimulated growth. These data suggest that inhibition of PDGF-stimulated proliferation by alpha 2M-plasmin is mediated in part by clearance of PDGF-BB-alpha 2M-plasmin through the lipoprotein receptor-related protein.

Multiple neurotrophic factors including NGF-like activity in nerve regeneration chamber fluids

Silicone nerve regeneration chambers were implanted between the cut ends of the sciatic nerve of adult rats. Neurotrophic activities in cell-free fluids collected from the chambers were determined using bioassays for survival of embryonic chick ciliary and sympathetic neurons in culture. Separation by molecular exclusion HPLC of the components of fluids collected 1, 2 or 3 days after implantation revealed the presence of a multitude of neurotrophic factors differing in their molecular weights, specificity towards the two types of neurons, and time course. Antiserum to nerve growth factor partially blocked sympathetic activity of fluids collected at 1 day. Affinity purified antibody was also effective and completely eliminated bioactivity of HPLC fractions corresponding to the molecular weight of nerve growth factor. The presence in the fluids of 13-18 and 20-32 kD components active towards ciliary neurons is consistent with the release of fibroblast growth factor and ciliary neurotrophic factor respectively. The stimulation of sympathetic neurons by the 13-18 kD material, and also by 4-6 and 7-11 kD components cannot be entirely accounted for by known factors. This study demonstrates that a number of neurotrophic factors, which differ in their specificity towards sympathetic and parasympathetic neurons, are made available to the region of axonal regrowth over the first few days of regeneration. Contrary to earlier reports, nerve growth factor-like activity was shown to be present in nerve regeneration chambers.

Nerve growth factor released by transgenic astrocytes enhances the function of adrenal chromaffin cell grafts in a rat model of Parkinson's disease

Previous studies have demonstrated that astrocytes genetically modified to express recombinant nerve growth factor (NGF) support the survival and neuronal transdifferentiation of intrastriatal adrenal chromaffin cell grafts at 2 weeks post-transplantation [15]. The present study was performed to determine whether these effects would be maintained at longer times post-transplantation and, if so, whether the co-grafts would reduce rotational behavior in the unilateral 6-hydroxydopamine-lesioned rat. In the present study, we have demonstrated that primary type I rat astrocytes infected with a replication-defective retrovirus conferring expression of a mouse beta-NGF cDNA sequence secrete NGF at a rate that is approximately 40-fold higher than that of controls (i.e., 8.0 vs. 0.2 pg NGF/h/10(5) cells, respectively). The genetically modified astrocytes were also found to express recombinant NGF following intrastriatal transplantation, as indicated by a 23% increase in striatal NGF content compared with controls, measured at 4 weeks post-transplantation. When NGF-producing astrocytes and adrenal chromaffin cells were co-grafted into the dopamine-denervated striatum of the unilateral 6-hydroxydopamine-lesioned rat, the chromaffin cells displayed extensive neurite outgrowth and a 5-12-fold increase in survival compared to controls at 10 weeks post-grafting. These effects were paralleled by a 60% reduction of apomorphine-induced rotational behavior, suggesting a partial normalization of striatal function. These results suggest that genetically modified astrocytes promote the prolonged survival and function of adrenal chromaffin cell grafts in a rat model of Parkinson's disease.

Intracranial infusion of monoamine-activated alpha 2-macroglobulin decreases dopamine concentrations within the rat caudate putamen

Monoamine-activated alpha 2-macroglobulin (alpha 2M) has been shown to inhibit choline acetyltransferase in basal forebrain neurons as well as neurotrophin-dependent neuronal functions. The objective of this study was to determine whether monoamine-activated alpha 2M can affect the caudate putamen (CP) dopaminergic system in vivo. Male rats received intracranial infusions of methylamine-activated alpha 2M (0.6 nmole) and contralateral infusions of its vehicle, phosphate-buffered saline (PBS). Five days following infusion, the animals were killed, the CP dissected into three rostral-caudal segments, and assayed for dopamine (DA) using a high-performance liquid chromatography system. Within the two rostral CP segments (the approximate site of cannula placement), statistically significant (26%) reductions of DA concentrations were obtained on the alpha 2M-infused side of the CP with 90-100% of the animals showing decreases. At a more distal (caudal) site of the CP, DA concentrations showed only an insignificant (12%) reduction. No differences in DA concentrations between sides infused with bovine serum albumin versus PBS or from olfactory tubercle samples were obtained in these animals. These results demonstrate that monoamine-activated alpha 2M is capable of producing significant degeneration of the nigrostriatal dopaminergic system in vivo and suggest that this factor may play a role in age-related neurodegenerative disorders such as Parkinson's disease.

Monoamine-activated alpha 2-macroglobulin inhibits choline acetyltransferase of embryonic basal forebrain neurons and reversal of the inhibition by NGF and BDNF but not NT-3

Monoamine-activated alpha 2-macroglobulin (alpha 2M) has recently been shown to inhibit the growth and survival of cholinergic neurons of the basal forebrain (Liebl and Koo: J Neurosci Res 35:170-182, 1993). The mechanism of this inhibitory effect is believed to involve the regulation of growth factor activities by alpha 2M. The objectives of this study are to determine whether monoamine-activated alpha 2M can inhibit choline acetyltransferase (ChAT) activity of cholinergic basal forebrain neurons, and whether some common neurotrophins in the CNS can reverse the inhibition. This study demonstrates that both methylamine-activated alpha 2M (MA-alpha 2M) and serotonin-activated alpha 2M (5HT-alpha 2M) can dose-dependently suppress the expression of normal basal levels of ChAT activity in embryonic rat basal forebrain cells in vitro, while normal alpha 2M has little or no effect. As little as 0.35 microM monoamine-activated alpha 2M can suppress the ChAT activity, whereas either nerve growth factor (NGF) or brain-derived neurotrophic factor (BDNF), but not neurotrophin-3 (NT-3), stimulates ChAT expression of these cells. The addition of either NGF or BDNF to the alpha 2M-suppressed cells can increase ChAT activity back to its normal levels, while NT-3 can not. These results demonstrate that (1) monoamine-activated alpha 2M is a potent non-cytotoxic inhibitor of the ChAT activity in cholinergic basal forebrain neurons, and (2) NGF and BDNF are capable of not only stimulating the ChAT activity but can also specifically reverse the alpha 2M inhibition.(ABSTRACT TRUNCATED AT 250 WORDS)

Binding of lipoprotein lipase to alpha 2-macroglobulin

The interaction between bovine lipoprotein lipase (bLPL) and human alpha 2-macroglobulin (alpha 2M) was studied by use of non-denaturing PAGE and gel-permeation, Zn(2+)-Sepharose and heparin-Sepharose chromatography. It was demonstrated that bLPL in vitro binds non-covalently to native alpha 2M, but not to the receptor-recognized form produced by treatment of alpha 2M with chymotrypsin or methylamine. A small amount of bLPL was bound covalently to alpha 2M by disulphide interchange, when incubated together with chymotrypsin or methylamine. Whereas alpha 2M in complex with bLPL still bound to Zn(2+)-Sepharose, bLPL lost the ability to bind to heparin-Sepharose. Preincubation of bLPL with heparin prevented complex-formation with alpha 2M, suggesting that alpha 2M interacts with the heparin-binding domain of bLPL. Experiments in which 125I-bLPL was incubated with human plasma at 20 degrees C demonstrated an 11-17% binding of the labelled lipase to alpha 2M, indicating that this interaction may be of physiological significance.

Expression of the alpha 2-macroglobulin-encoding gene in rat brain and cultured astrocytes

The alpha 2-macroglobulin (alpha 2M), a protease inhibitor, is a major acute-phase protein in rats, and is produced in the liver during acute inflammation. Recently, it has been demonstrated that alpha 2M is also produced by cultured astrocytes from newborn rat brain and has neurite-promoting activity. Here, we found that the expression of the alpha 2M gene was significantly enhanced in the brain following intraperitoneal injection of the neurotoxicant, kainic acid (KA), suggesting that alpha 2M acts as an acute-phase protein in the brain, as in the case of the liver, and may be involved in neural repair processes. Expression of alpha 2M in cultured astrocytes was shown to be stimulated by interleukin-6 (IL-6) and/or leukemia inhibitory factor (LIF) in the presence of glucocorticoid. The amount of mRNAs for IL-6 and LIF increased in the brain of KA-injected rats prior to alpha 2M induction. These results strongly suggested that IL-6 and LIF are involved in alpha 2M induction in the brain, as in the case of the liver. Analysis of the cis-acting element(s) and the trans-acting factor(s) suggested that the regulatory mechanism for alpha 2M expression in astrocytes was similar to that in inflamed liver.

Distribution and metabolism patterns of plasma 7S- and beta-NGF in the adult male rat

There is a great deal of controversy on the existence of NGF in body fluids and tissues. To date it remains unknown whether this peptide accumulates preferentially at significant levels in different organs. Thus we undertook the evaluation of kinetic parameters of the disappearance of blood of 125I-7S-NGF and 125I-beta-NGF after intravenous injection in male adult rats. Our results indicate that the plasma half-life of 125I-7S-NGF is approximately twice as long as for 125I-beta-NGF (respectively 61.7 +/- 11.7 min and 36.3 +/- 2.2 min) while the distribution volume is not significantly different between both peptides. Furthermore, the uptake of radioactive NGF by different tissues seems very low as shown by 125I-7S-NGF and 125I-beta-NGF content of the sampled organs compared to the plasma concentration at the same time. These results indicate that the tissue uptake of circulating 7S and beta-NGF is very low in the adult rat. Thus in these animals NGF did not cross the blood-brain barrier and did not accumulate in peripheral organs which are known to contain subsequent amounts of this peptide. This lack of deposition might be due to a binding with plasma proteins (probably alpha 2-macroglobulin).

Natural autoantibodies against the nerve growth factor in autoimmune diseases

High titers of natural autoantibodies against the nerve growth factor (NGF) were detected in the sera of patients with systemic lupus erythematosus, autoimmune thyroiditis and rheumatoid arthritis. Autoantibodies to NGF from these pathological cases displayed higher avidity for NGF and a higher polyreactivity with certain cytoskeletal proteins and with DNA as compared to those from control human subjects. The biological activity, immunoglobulin composition and physiological relevance of these autoantibodies are discussed.

Production of conformation-specific monoclonal antibodies against alpha 2 macroglobulin and their use for quantitation of total and transformed alpha 2 macroglobulin in human blood

Monoclonal antibodies against the human proteinase inhibitor, alpha 2 macroglobulin, have been produced by immunizing BALB/c mice with alpha 2 macroglobulin reacted with methylamine. Two antibodies have been characterized in detail with respect to their binding to native alpha 2 macroglobulin and to different derivatives of the inhibitor. The antibody alpha-1 was found to recognize only those forms of the inhibitor which were transformed by reaction with different proteinases or with methylamine. Binding of alpha-1 was mapped to a specific epitope localized within a distance of 138 amino acid residues from the C terminal end of alpha 2 macroglobulin. The C terminal end is assumed to be exposed during the transformation of the inhibitor and harbours the receptor recognition site. The monoclonal antibody alpha-11 was found to bind to all forms of the inhibitor indicating that its epitope is located in a region not involved in major conformational changes of the inhibitor. On the basis of the different reactivity patterns of alpha-1 and alpha-11 two enzyme-linked immunosorption assays were established for quantitation of total and transformed alpha 2 macroglobulin in human blood. The concentration of the two forms have been determined in a population of 114 healthy individuals giving values of 254 +/- 6.6 mg/dl (mean +/- SEM) of total alpha 2 macroglobulin and 1.07 +/- 0.05 mg/dl (mean +/- SEM) of the transformed inhibitor.

Serotonin-activated alpha 2-macroglobulin inhibits neurite outgrowth and survival of embryonic sensory and cerebral cortical neurons

Methylamine-modified alpha-2-macroglobulin (MA-alpha 2M) has been recently shown to inhibit the biological activity of beta-nerve growth factor (NGF) in promoting neurite outgrowth by embryonic dorsal root ganglia in culture (Koo PH, Liebl DJ, J Neurosci Res 31:678-692, 1992). The objectives of this study are to determine whether alpha 2M can also be modified by larger aromatic biogenic amines such as 5-hydroxytryptamine (5HT; serotonin), the nature of interaction between NGF and 5HT-modified alpha-2-M (5HT-alpha 2M), and the effect of 5HT-alpha 2M on the neurite extension and the growth of embryonic sensory and cholinergic neurons in 2 disparate animal species (chicken and rats). This study demonstrates that each mole of alpha 2M can combine with 15.2 +/- 1.8 moles of 5HT, in which up to 4.5 +/- 0.4 moles may be covalently bonded. As determined by gel filtration and polyacrylamide gel electrophoresis studies, both 5HT-alpha 2M and normal alpha 2M combine noncovalently with NGF, but 5HT-alpha 2M by comparison can combine with NGF somewhat more effectively. In contrast to normal alpha 2M, 5HT-alpha 2M at concentrations greater than about 0.17 microM exerts a dose-dependent inhibition on the NGF-stimulated neurite outgrowth by embryonic dorsal root ganglia and dissociated cells in culture, and the inhibitory effect can be overcome by higher NGF concentrations. Both 5HT-alpha 2M and MA-alpha 2M at 1.0 microM inhibit neurite extension by embryonic rat cerebral cortical cells and seriously damage these cells in culture. Such neurite-inhibitory activity, however, can only be partially blocked by extraneously added NGF alone. Normal alpha 2M (at 1.0 microM) and 5HT (at 188 microM), on the other hand, under the identical conditions produce very little or no effect on the normal cellular and axonal growth of these cells. We conclude that alpha 2M can potentially interact with nucleophilic monoamines, including neurotransmitters, to form inhibitory complexes which may inhibit/regulate NGF-promoted neurite outgrowth and neuronal survival. In addition, higher concentrations of such complexes can seriously damage certain CNS neurons which do not depend solely on NGF for survival.

Differential isoform profiles of alpha 2-macroglobulin from plasma of patients with chronic-progressive or relapsing-remitting multiple sclerosis

Multiple sclerosis (MS) is a human neurological disease for which no clinically useful marker has been identified in blood. This study examined alpha 2-macroglobulin (alpha 2M) from the plasma of six patients with chronic-progressive MS and six with relapsing-remitting disease. The alpha 2M trypsin-binding activity in the plasma from both groups of patients did not differ from normal controls. However, after column isoelectric focusing, consistently less alpha 2M activity was recovered from the MS samples: those from the chronic-progressive and relapsing-remitting disease groups were an average of 43% and 68%, respectively, of controls. The number and isoelectric point (pI) values of the isoforms of the alpha 2M from patients with chronic-progressive disease were similar to controls. The average pI of the major form for both groups was 6.6. By contrast, the average pI of the major form from the patients with relapsing-remitting MS was significantly elevated to 7.1, and this group displayed a significantly higher percentage of total recovered activity above pH 7.0. In eleven of the twelve cases examined, the pI of the major form of alpha 2M correctly correlated with the clinical status of the patient. The original clinical diagnosis of the patients was reassessed by a 9-year retrospective interview which verified that 9 of the 10 patients in the follow-up group retained their original clinical diagnosis. These studies demonstrate differential isoform profiles of native alpha 2M from MS patients with progressive versus remitting disease which may be useful in subclassifying MS patients.

Developmental changes in nerve growth factor level in rat serum

In serum, nerve growth factor (NGF) forms a complex with alpha 2-macroglobulin (alpha 2M), which formation inhibits the immunoreactivity between NGF and its antibodies. For measuring the serum level of NGF, it is thus necessary to liberate NGF from the NGF-alpha 2M complex and prevent reformation of such complex. The pretreatment of rat serum with 1 M guanidine hydrochloride for a few hours and operation of the enzyme immunoassay (EIA) in the presence of guanidine hydrochloride provided a reliable means for determination of the NGF level in serum. By this procedure we followed the serum NGF level in rats developmentally. It increased from prenatal day 2 to postnatal day 5 and decreased slightly at postnatal week 3, thereafter remaining constant throughout adulthood. In pregnant rats, the NGF level in serum increased threefold to fivefold before birth and then decreased rapidly. These data suggest that serum NGF level may reflect the demand for this molecule during establishment of the peripheral nervous system.

Inhibition of nerve growth factor-stimulated neurite outgrowth by methylamine-modified alpha 2-macroglobulin

alpha 2-Macroglobulin (alpha 2M) is a rather ubiquitous protein in extracellular spaces of mammals. It is an inhibitor of endopeptidases, can be modified by aliphatic amines, and combines with a number of hormones/cytokines such as beta-nerve growth factor (NGF) [Koo PH, Stach RW (1989): J Neurosci Res 22:247]. The objective of this study is to compare the NGF-binding properties of methylamine-modified human alpha 2M (MA-alpha 2M) versus normal alpha 2M and their effects on the biological activity of NGF and neurite extension by embryonic chicken dorsal root ganglia. As determined by gel filtration, polyacrylamide gel electrophoresis, and equilibrium binding studies, these two forms of alpha 2M are similar in their binding affinities, with MA-alpha 2M binding about twice as much NGF as normal alpha 2M. Both normal alpha 2M and MA-alpha 2M combine noncovalently with NGF, and prior modification of alpha 2M is unnecessary for the binding to occur. In contrast to normal alpha 2M, MA-alpha 2M potently inhibits the biological activity of NGF and exerts a dose-dependent inhibition on the NGF-stimulated neurite outgrowth by embryonic chicken dorsal root ganglia in culture. The inhibitory effect of MA-alpha 2M can be overcome by higher NGF concentrations, but is irreversible at lower NGF concentrations. Trypsin-modified alpha 2M combines covalently and noncovalently with more NGF than normal alpha 2M but has very little neurite inhibitory activity. The mechanism of inhibition by MA-alpha 2M is discussed.

Binding of transforming growth factor-beta 1 to methylamine-modified alpha 2-macroglobulin and to binary and ternary alpha 2-macroglobulin-proteinase complexes

The binding of 125I-labelled transforming growth factor-beta 1 (TGF-beta 1) to human alpha 2-macroglobulin (alpha 2M) was studied by native PAGE and autoradiography. TGF-beta 1 bound preferentially to alpha 2M-methylamine and minimally, if at all, to native alpha 2M. Preparations of alpha 2M-proteinase complex were generated by incubating a standard concentration of alpha 2M (0.4 microM) with different concentrations of trypsin, chymotrypsin or neutrophil elastase (0.04-2.0 microM). The 125I-TGF-beta 1-binding activity depended on the initial ratio of active proteinase to alpha 2M, or r value, used to form the alpha 2M-proteinase complex. With all three proteinases, r values of 2 or greater yielded preparations with unchanged or decreased TGF-beta 1-binding activity relative to native alpha 2M. By contrast, r values near 1 yielded preparations with significantly increased TGF-beta 1-binding activity. The results of [3H]thymidine-incorporation studies performed in mouse keratinocytes were consistent with the 125I-TGF-beta-binding experiments. alpha 2M-trypsin and alpha 2M-chymotrypsin prepared at an r value of 1.0 counteracted the activity of TGF-beta 1, whereas the equivalent complexes prepared at an r value of 3.0 had no effect. As determined by SDS/PAGE, 125I-TGF-beta 1 binding to alpha 2M-methylamine was at least 80% non-covalent. Reaction of alpha 2M-methylamine with iodoacetamide or 5,5'-dithiobis-(2-nitrobenzoic acid) decreased the percentage of covalent binding but had no effect on total binding. Neuraminidase treatment had no effect on the binding of 125I-TGF-beta 1 to alpha 2M-methylamine. Cleavage of the 'bait regions' in alpha 2M-methylamine by prolonged treatment with trypsin also had no effect. These studies suggest that TGF-beta 1 binding to alpha 2M is enhanced by conformational change in the proteinase inhibitor resulting from reaction with proteinase or amine. If both proteinase-binding sites in a single alpha 2M molecule are occupied, TGF-beta 1-binding activity is decreased or perhaps eliminated.

Binding of transforming growth factor-β1 to immobilized human α2-macroglobulin

Native alpha 2-macroglobulin (alpha 2M) and alpha 2M-methylamine were immobilized in 96-well microtiter plates. 125I-labeled transforming growth factor-beta 1 (TGF-beta 1) bound to both alpha 2M variants; however, greater binding was observed with alpha 2M-methylamine. Binding of 125I-TGF-beta 1 (0.2 nM) to immobilized alpha 2M-methylamine was inhibited by nonradiolabeled TGF-beta 1 (up to 74% with 0.4 microM TGF-beta 1). Approximately 10% of the TGF-beta 1-alpha 2M-methylamine complex was covalent. Treatment of alpha 2M-methylamine with iodoacetamide prior to immobilization completely eliminated covalent TGF-beta 1 binding; the total amount of 125I-TGF-beta 1-alpha 2M-methylamine complex detected was unchanged. The binding of 125I-TGF-beta 1 to immobilized alpha 2M-methylamine was not significantly inhibited by increasing the ionic strength to 1.0 M. Binding and complex dissociation were also unaffected by changes in pH within the range 6.9-8.9. Acidic pH dramatically decreased binding and promoted complex dissociation; no binding of 125I-TGF-beta 1 to immobilized alpha 2M-methylamine was detected at pH 3.5. The interaction of TGF-beta 1 with immobilized alpha 2M-methylamine was not significantly changed by 1.0 mM EDTA or 1.0 mM CaCl2. ZnCl2 (1.0 mM) completely eliminated binding. This result was not due to TGF-beta 1 precipitation or aggregation. Inhibition of 125I-TGF-beta 1 binding to alpha 2M-methylamine was 50% complete (IC50) with 30 microM ZnCl2. Native alpha 2M, thrombospondin, and alpha 2M-methylamine (in solution) decreased binding of 125I-TGF-beta 1 to immobilized alpha 2M-methylamine. The IC50 values for these three proteins were 520, 160, and 79 nM, respectively. The TGF-beta 1-binding activity of native alpha 2M may have reflected, at least in part, trace-contamination with alpha 2M-proteinase complex.

Ultrastructure of alpha 2-macroglobulins

New results concerning the ultrastructure of human alpha 2-macroglobulin (alpha 2M) molecules are presented in connection and comparison with the historical, the current and our own most recent, even unpublished results on the structure and function of alpha 2M and related proteins. The electron microscopic approach uses classical negative staining, combined with the new imaging mode "Electron Energy Loss Spectroscopy", which provides unusual contrast, resolution and readability of the electron micrographs. Immuno- and cryoelectron microscopy, as well as image processing has provided new data necessary to the building of tentative 3D models of the molecule. A model for the native tetrameric alpha 2M is described for the first time, and tries to explain and gather the various observations, sometimes contradictory, taken from different laboratories. A revised version for a model of the methylamine- and proteinase-transformed forms of alpha 2M is also shown. The probable positions of the bait regions and the thiol esters are given on both models. We confirm that alpha 2M is a twin trap capable of inactivating one or two proteinases by partial immobilization. Preliminary results on the production of crystals of alpha 2M-chymotrypsin complexes are also presented. A critical analysis of our models is presented in comparison with others. The technical limitations reached with some techniques and some possible extensions of future research in the field are also presented.

Dissociation of neurite-promoting activity and protease-inhibiting function of alpha 2-macroglobulin in culture

The relationship between the neurite-promoting effect and the protease-inhibiting function of alpha 2-macroglobulin (alpha 2M) was examined in a culture of dissociated neurons from embryonic rat neocortical tissue. The neurite-promoting effect of various protease inhibitors (soybean trypsin inhibitor, alpha 1-antitrypsin, aprotinin, phenylmethanesulfonyl fluoride, leupeptin, antipain, human alpha 2M and purified rat alpha 2M) was investigated. At lower concentrations, only alpha 2M significantly promoted neurite outgrowth. Furthermore, alpha 2M-trypsin complexes, which had lost their protease-inhibiting function and exposed their receptor-recognition site, promoted neurite outgrowth more efficiently than native alpha 2M. These results indicate the possibility that the neurite-promoting effect of alpha 2M is mediated by its receptor-binding activity but is not directly related to its protease-inhibiting function.

Reaction of alpha 2-macroglobulin with plasmin increases binding of transforming growth factors-beta 1 and beta 2

The binding of 125I-transforming growth factors-beta 1 and beta 2 (TGF-beta 1 and TGF-beta 2) to alpha 2-macroglobulin (alpha 2M) was studied before and after reaction with plasmin, thrombin, trypsin, or methylamine. Complex formation between TGF-beta and native or reacted forms of alpha 2M was demonstrated by non-denaturing polyacrylamide gel electrophoresis and autoradiography. Reaction of native alpha 2M with plasmin or methylamine markedly increased the binding of 125I-TGF-beta 1 and 125I-TGF-beta 2 to alpha 2M. The alpha 2M-plasmin/TGF-beta complexes were minimally dissociated by heparin. Reaction of alpha 2M with thrombin or trypsin reduced the binding of 125I-TGF-beta 1 and 125I-TGF-beta 2; the resulting complexes were readily dissociated by heparin. Complexes between TGF-beta 2 and native or reacted forms of alpha 2M were less dissociable by heparin than the equivalent complexes with TGF-beta 1. These studies demonstrate that the TGF-beta-binding activity of alpha 2M is significantly affected by plasmin, thrombin, trypsin and methylamine. Observations that alpha 2M-plasmin preferentially binds TGFs-beta suggest a mechanism by which alpha 2M may regulate availability of TGFs-beta to target cells in vivo.

Structure and biosynthesis of nerve growth factor

Most of our knowledge about NGF comes from extensive study of the mouse submaxillary gland protein. NGF from this source is isolated as a high molecular weight complex consisting of beta-NGF and two subunits, alpha and gamma, belonging to the kallikrein family of serine proteases. There are few other tissues where NGF is found in sufficient quantities for protein purification and study, although new molecular biological techniques have accelerated the study of NGFs from a variety of species and tissues. Mouse submaxillary gland NGF is synthesized as a large precursor that is cleaved at both N- and C-terminals to produce mature NGF. This biologically active molecule can be further cleaved by submaxillary gland proteases. The roles of the alpha and gamma subunits in the processing of the beta-NGF precursor, the modulation of the biological activity of beta-NGF, and the protection of mature beta-NGF from degradation have been well studied in the mouse. However, the apparent lack of alpha and gamma subunits in most other tissues and species and the existence of a large family of murine kallikreins, many of which are expressed in the submaxillary gland, challenge the relevance of murine high molecular weight NGF as a proper model for NGF biosynthesis and regulation. It is important therefore to identify and characterize other NGF complexes and to study their subunit interactions, biosynthesis, processing, and regulation. This review points out a number of other species and tissues in which the study of NGF has just begun. At this time, there exist many more questions than answers regarding the presence and the functions of NGF processing and regulatory proteins. By studying NGF in other species and tissues and comparing the processing and regulation of NGF from several sources, we will discover the unifying concepts governing the expression of NGF biological activity.

Alpha 2-macroglobulin is an astroglia-derived neurite-promoting factor for cultured neurons from rat central nervous system

The neurite promoting factors in the astroglial conditioned medium (As-CM) were characterized by using primary cultures of embryonic rat neocortical neurons. The factors in the As-CM bind to lectins such as wheat germ agglutinin (WGA), suggesting that they contain sugar moieties. When the WGA-bound fractions were applied on a Superose 6 column, the activity was recovered mainly in two fractions, peak I and peak II. The peak II fraction was further purified by Mono Q anion exchange chromatography. A single protein band of 180 kDa was detected in the final Mono Q fraction by sodium dodecylsulfate polyacrylamide gel electrophoresis. The molecular weight coincided with that of alpha 2-macroglobulin (alpha 2M). Western blotting showed that the single protein band was reacted with anti-alpha 2M antibody but not with anti-fibronectin and anti-laminin antisera. The neurite-promoting activity of the Mono Q fraction was inhibited by anti-alpha 2M antibody. Furthermore, commercially available alpha 2M also promotes neurite outgrowth in our assay system. These results strongly suggested that alpha 2M is one of the neurite-promoting factors in the As-CM.