No enhancement of fusion probability by the neutron halo of 6He.

Quantum tunnelling through a potential barrier (such as occurs in nuclear fusion) is very sensitive to the detailed structure of the system and its intrinsic degrees of freedom. A strong increase of the fusion probability has been observed for heavy deformed nuclei. In light exotic nuclei such as 6He, 11Li and 11Be (termed 'halo' nuclei), the neutron matter extends much further than the usual nuclear interaction scale. However, understanding the effect of the neutron halo on fusion has been controversial--it could induce a large enhancement of fusion, but alternatively the weak binding energy of the nuclei could inhibit the process. Other reaction channels known as direct processes (usually negligible for ordinary nuclei) are also important: for example, a fragment of the halo nucleus could transfer to the target nucleus through a diminished potential barrier. Here we study the reactions of the halo nucleus 6He with a 238U target, at energies near the fusion barrier. Most of these reactions lead to fission of the system, which we use as an experimental signature to identify the contribution of the fusion and transfer channels to the total cross-section. At energies below the fusion barrier, we find no evidence for a substantial enhancement of fusion. Rather, the (large) fission yield is due to a two-neutron transfer reaction, with other direct processes possibly also involved.

CD4 T cell recovery is slower in patients experiencing viral load rebounds during HAART.

To determine whether viral load rebounds during HAART impact on CD4+ T cell recovery and immune reconstitution, we studied a prospective cohort of 355 antiretroviral naive patients enrolled to be randomized in a trial of three strategies of induction/maintenance HAART. The extent of immune reconstitution in blood through 72 weeks of antiretroviral treatment was evaluated. Lymphocyte subset markers (CD4, CD8, CD45RA, CD62L, CD16, CD19), activation markers (HLA-DR, CD38, CD25) were performed by cytometry analysis. Our results showed that plasma HIV-1 RNA was suppressed to below 500 copies per ml through week 72 in 240 patients (group 1) while the remaining 115 patients experienced at least one viral rebound (group 2). At baseline, CD4 cell count was higher and HIV-1 RNA was lower in group 1 than in group 2. Over 72 weeks, mean increase in CD4+ T cell count was 0.32 cell/mm3/day in group 1 and only 0.14 cell/mm3/day in group 2 (P < 0.0001). However, the patterns of changes in CD4+ and CD8+ T cell subsets during therapy were very similar across the two groups with only subtle and very limited differences. We conclude that permanent control of HIV replication could be necessary for faster immune reconstitution.

Frequency and phenotyping of human immunodeficiency virus (HIV)-specific CD8+ T cells in HIV-infected children, using major histocompatibility complex class I peptide tetramers.

HLA-A*02 tetramers complexed to human immunodeficiency virus (HIV) Gag SLYNTVATL and HIV Pol ILKEPVHGV peptides were used to characterize HLA class I-restricted CD8(+) T cells in 41 HIV-infected children. The frequencies and the phenotype of specific circulating CD8(+) T cells were determined in whole-blood samples by means of cytometric analysis. Background staining of 13 HLA-A*02-negative patients showed that the frequency of CD8(+) T cells was <0.01%. Of the 28 HLA-A*02-positive patients, blood samples from 26 stained positive at least once the Gag tetramer (mean CD8(+) T cells, 0.87%; range, 0.1%-3.9%), and blood samples from 21 stained positive for the Pol tetramer (mean CD8(+) T cells, 0.59%; range, 0.1%-5.5%). The tetramer-binding cells were CD28(-), CD45RA(-), CD45RO(+), HLA-DR(+), and CD69(-) T lymphocytes. HIV-specific CD8(+) T cells can be detected easily in peripheral blood of HIV-infected children, using HLA tetramers combined with HIV peptides. These cells are memory activated CD28(-)CD8(+) T lymphocytes.

Functional monocyte-derived dendritic cells can be generated in chronic lymphocytic leukaemia.

Chronic lymphocytic leukaemia (CLL) remains an incurable disease. Although modern available treatments are able to induce disease regression, relapse almost inexorably occurs. Therefore, novel therapeutic strategies aimed at reducing the disease relapse rate are very much needed. Among these, the induction of tumour-associated antigen-specific cytotoxic T lymphocytes (CTL), through either DNA vaccines or injection of idiotype pulsed dendritic cells (DCs), has been actively investigated with encouraging preliminary results in B-cell malignancies. As the CLL B lymphocyte characteristically expresses low amounts of surface immunoglobulin (Ig) and T cells from these patients have been reported to display impaired functional activity, there are concerns related to the possibility of generating specific cytotoxic antitumoral T cells in this disease. In addition, no information is presently available regarding the functional ability of CLL-derived DCs. In the present work, freshly purified monocytes from CLL patients and normal donors were induced to differentiate in granulocyte-macrophage colony-stimulating factor (GM-CSF) and interleukin (IL)-4 serum-free medium and compared for their morphological, phenotypic and functional characteristics. Our results demonstrate that: (1) functional DCs can be generated from CLL patients with similar phenotype and function to those observed from normal donors; (2) in contrast to normal control subjects, monocyte-derived DCs from CLL patients spontaneously secrete endogenous IL-10; and (3) interferon (IFN)-gamma in combination with CD40L plays a major role in priming DCs from CLL patients for IL-12 and IL-15 production. Overall, these results indicate that it is possible to derive functionally competent DCs from circulating monocytes in CLL patients.

Large enhancement of the sub-barrier fusion probability for a halo nucleus

The fusion-fission cross sections of the 4He+238U and 6He+238U systems have been measured, at Louvain-la-Neuve, for energies around and below the Coulomb barrier, using an array of Si detectors surrounding a UF4 target. The data taken with 4He are in good agreement with previous data and with the coupled channel fusion calculation performed with ECIS. The 6He data show a regular trend with a large enhancement below the barrier which is attributed to the halo structure of the 6He nucleus.

A peptide derived from a polyreactive monoclonal anti-DNA natural antibody can modulate lupus development in (NZBxNZW)F1 mice.

In lupus-prone (NZBxNZW)F1 (B/W) mice, elevated levels of polyreactive autoantibodies bearing the D23 idiotype (Id), characteristic of natural antibodies, were detected before and after the appearance of pathological anti-DNA antibodies. While these D23 Id+ antibodies were able to regulate anti-DNA antibodies in the early stage of the disease, we found that during disease evolution they had lost their normal ability to regulate anti-DNA antibodies and furthermore could participate in the lupus-like syndrome. To explore further the role of the D23 Id+ antibodies, we injected young B/W mice with a peptide corresponding to the VH CDR3 region of the D23 monoclonal natural antibody (mNAb). High levels of monospecific antipeptide, as well as polyreactive antibodies, were induced. Among them, the most markedly enhanced antibody population was DNA-reactive immunoglobulin G1 (IgG1). Compared with controls, these immunized mice had a delayed 50% survival rate and proteinuria developed later. Furthermore, IgG1 able to react with IgG2a anti-DNA monoclonal antibodies derived from B/W mice were also produced after peptide immunization. Thus, a peptide corresponding to the CDR3 of the D23 mNAb antibody might play a role in the regulation of murine lupus.

Natural autoantibodies are involved in the haemolytic anaemia of NZB mice.

NZB is a mouse strain that spontaneously develops autoimmune haemolytic anaemia at 10-12 months of age. We analysed the autoantibodies present throughout their life and compared them to natural autoantibodies found in the normal mouse. Sera and Coombs' antibodies eluted from red blood cells (RBC) were tested for their activities against RBC and a panel of antigens: actin, myoglobin, myosin, tubulin, spectrin, DNA and trinitrophenyl bovine serum albumin (TNP-BSA), F(ab')2 and Fc fragments of IgG by using enzyme immunoassays (EIA) and Western blotting analysis of RBC membrane extracts. In NZB mouse sera, activities of IgM and IgG against the whole panel, compared to those of sera from age-matched BALB/c mice, increased progressively throughout life with oscillating values in parallel with the anti-RBC activity. Two periods of autoantibody production seem to exist: the first is characterized by a fluctuating high level of IgM and stable level of IgG natural autoantibodies, and the second by a rise of IgG natural autoantibodies in parallel with IgG anti-RBC antibodies. The presence of idiotype D23 (IdD23), which is characteristic of natural polyspecific autoantibodies, was high on serum IgM and low on IgG autoantibodies throughout life. To further analyse autoantibody level oscillations, we tested IgM and IgG fractions after their separation from whole serum and observed highly enhanced autoantibody activities of both IgM and IgG. These autoreactivities markedly diminished when the separated IgM and IgG fractions were recombined, suggesting humoral control of the autoreactivity as we had already noted for IgG in normal animals. During the first period of autoantibody production, IgM and IgG antibodies eluted from RBC (Combs' antibodies) and those eluted from serum using an RBC-immunoadsorbent (circulating antibodies) reacted with all RBC membrane components, with all antigens of the panel and with F(ab')2 and Fc. Some of these reactivities were comparable to those exhibited by a monoclonal antibody recognizing bromelain-treated RBC. In the second period, both IgM and IgG Coombs' antibodies reacted more strongly with spectrin, and exhibited new specificities, for example against the band 3 polypeptide. IdD23 was abundant on Combs' IgG antibodies in the second period. Taken together, these data suggest that IgM and IgG natural autoantibodies, able to recognize not only RBC antigens but also other antigens, particularly F(ab')2 and Fc fragment of IgG, predominate in Coomb's antibody population.(ABSTRACT TRUNCATED AT 400 WORDS)

Antigen presentation capacity of murine macrophages infected with Leishmania amazonensis amastigotes.

Leishmania-infected M phi are potential candidates for the presentation of parasite Ag to Leishmania-specific CD4+ T lymphocytes. To assess whether infected cells could function as APC, we examined the ability of bone marrow-derived M phi infected with Leishmania amazonensis amastigotes to stimulate various CD4+, l-Ad- or l-Ed-restricted T-cell hybridomas specific for the bacteriophage lambda repressor cl protein, the human chorionic gonadotropin or OVA. A reduced capacity of infected M phi to present native Ag to most T-cell hybridomas tested was noted that was probably a result of a lower expression on their plasma membrane of stimulatory [la-peptide] complexes. Neither a reduced Ag uptake nor an altered Ag processing appeared to be at the origin of the partial inability of infected M phi to present Ag. As regards the level of plasma membrane la expression, no quantitative difference could be detected between uninfected and infected M phi. Moreover, after fixation with paraformaldehyde, the ability of plasma membrane la molecules to bind immunogenic peptides was apparently not reduced in infected M phi. So, these cells most likely expressed functional la molecules on their cell surface. Interestingly, infected M phi and M phi infected then cured by a treatment with a leishmanicidal compound were similarly impaired in their capacity to present native Ag or peptides to the hybridomas, and no recovery was noted even 24 h after the leishmanicidal treatment. Furthermore, infected M phi and M phi incubated with heat-killed amastigotes or with an amastigote homogenate exhibited similar inhibitions of Ag presentation. Taken together, these results suggest that the functional failure of infected M phi to present exogenous Ag could be because either of interferences with the events leading to the meeting of la molecules with peptides derived from these exogenous Ag or to a competition for binding to la molecules between these peptides and parasite molecules.

Localization of major histocompatibility complex class II molecules in phagolysosomes of murine macrophages infected with Leishmania amazonensis.

Leishmania-infected macrophages are potential antigen-presenting cells for CD4+ T lymphocytes, which recognize parasite antigens bound to major histocompatibility complex class II molecules (Ia). However, the intracellular sites where Ia and antigens may interact are far from clear, since parasites grow within the modified lysosomal compartment of the host cell, whereas Ia molecules seem to be targeted to endosomes. To address this question, the expression and fate of Ia molecules were studied by immunocytochemistry in Leishmania amazonensis-infected murine macrophages stimulated with gamma interferon. In uninfected macrophages, Ia molecules were localized on the plasma membrane and in perinuclear vesicles, but they underwent a dramatic redistribution after infection, since most of the intracellular staining was then associated with the periphery of the parasitophorous vacuoles (p.v.) and quite often polarized towards amastigote-binding sites. The Ii invariant chain, which is transiently associated with Ia during their intracellular transport, although well expressed in infected macrophages, apparently did not reach the p.v. Similar findings were observed with macrophages from mice either resistant or highly susceptible to Leishmania infection. In order to determine the origin of p.v.-associated Ia, the fate of plasma membrane, endosomal, and lysosomal markers, detected with specific antibodies, was determined after infection. At 48 h after infection, p.v. was found to exhibit a membrane composition typical of mature lysosomes. Overall, these data suggest that (i) Ia located in p.v. originate from secondary lysosomes involved in the biogenesis of this compartment or circulate in several endocytic organelles, including lysosomes and (ii) p.v. could play a role in antigen processing and presentation. Alternatively, the presence of high amounts of Ia in p.v. could be due to a Leishmania-induced mechanism by means of which this organism may evade the immune response.

Parasitophorous vacuoles of Leishmania amazonensis-infected macrophages maintain an acidic pH.

Leishmania amastigotes are intracellular protozoan parasites of mononuclear phagocytes which reside within parasitophorous vacuoles of phagolysosomal origin. The pH of these compartments was studied with the aim of elucidating strategies used by these microorganisms to evade the microbicidal mechanisms of their host cells. For this purpose, rat bone marrow-derived macrophages were infected with L. amazonensis amastigotes. Intracellular acidic compartments were localized by using the weak base 3-(2,4-dinitroanilino)-3'-amino-N-methyldipropylamine as a probe. This indicator, which can be detected by light microscopy by using immunocytochemical methods, mainly accumulated in perinuclear lysosomes of uninfected cells, whereas in infected cells, it was essentially localized in parasitophorous vacuoles, which thus appeared acidified. Phagolysosomal pH was estimated quantitatively in living cells loaded with the pH-sensitive endocytic tracer fluoresceinated dextran. After a 15- to 20-h exposure, the tracer was mainly detected in perinuclear lysosomes and parasitophorous vacuoles of uninfected and infected macrophages, respectively. Fluorescence intensities were determined from digitized video images of single cells after processing and automatic subtraction of background. We found statistically different mean pH values of 5.17 to 5.48 for lysosomes and 4.74 to 5.26 for parasitophorous vacuoles. As for lysosomes of monensin-treated cells, the pH gradient of parasitophorous vacuoles collapsed after monensin was added. This very likely indicates that these vacuoles maintain an acidic internal pH by an active process. These results show that L. amazonensis amastigotes are acidophilic and opportunistic organisms and suggest that these intracellular parasites have evolved means for survival under these harsh conditions and have acquired plasma membrane components compatible with the environment.

Megasomes as the targets of leucine methyl ester in Leishmania amazonensis amastigotes.

Certain L-amino acid esters, such as L-leucine methyl ester (Leu-OMe), can kill intracellular and isolated Leishmania amazonensis amastigotes. Killing appears to involve ester trapping and hydrolysis within an acidified parasite compartment (M. Rabinovitch and S. C. Alfieri, 1987, Brazilian Journal of Medical and Biological Research 20, 665-74). We show here by acid phosphatase light microscopic cytochemistry and by ultrastructural morphometry that megasomes, lysosome-like amastigote organelles, are the putative parasite targets of Leu-OMe. This conclusion is supported by the following observations. (a) Control amastigotes displayed a string of electron-dense, acid phosphatase-positive megasomes mostly located in the cellular poles opposite the flagellar pockets. Incubation of the amastigotes with Leu-OMe resulted in concentration-dependent swelling and fusion of the organelles as well as decreased electron density of the internal contents. These changes, which preceded parasite disruption, were followed by the progressive loss of parasite viability and the release of acid phosphatase activity into the medium. (b) Incubation of the amastigotes with L-isoleucine methyl ester, a non-leishmanicidal compound, induced only moderate fusion of the megasomes. (c) Pre-incubation of the parasites with the proteinase inhibitors antipain and chymostatin, previously shown to confer protection from Leu-OMe toxicity, nearly completely prevented the morphological changes of megasomes. (d) Exposure of amastigotes to tryptophanamide (Trp-NH2), the leishmanicidal activity of which is not reduced by antipain and chymostatin, did not result in swelling and fusion of the megasomes. This last finding suggests that different mechanisms underlie the destruction of amastigotes by Trp-NH2 and Leu-OMe.(ABSTRACT TRUNCATED AT 250 WORDS)

Leishmania amazonensis: acidic organelles in amastigotes.

Leishmania amastigotes are intracellular protozoan parasites which exclusively invade cells of the macrophage series and multiply within phagolysosomes. Recent studies showed that intracellular and isolated amastigotes of L. amazonesis are killed by amino acid esters which appear to be trapped within as yet unidentified, possibly acidified, "lysosome-like" parasite compartments and cleaved by hydrolytic enzyme(s) (M. Rabinovitch, V. Zilberfarb, and C. Ramazeilles, 1986, Journal of Experimental Medicine 163, 520-535). In the present study, we have localized acidic compartments of Leishmania amastigotes using as a probe the weak base 3-(2,4 dinitroanilino)-3'-amino-N-methyldipropylamine (DAMP). This indicator, which can be detected within cells by light and electron microscopy using immunocytochemical immunocytochemical methods, mainly accumulates within megasomes and in dense inclusion vacuoles. With the help of quantitative assays to titrate cell-associated DAMP, it was found that (a) its uptake is temperature dependent and thus probably requires an energy supply, (b) the proton ionophore monensin partially inhibits the trapping of DAMP, and (c) monensin greatly increases its efflux from cells. These results, as well as those obtained by quantitative ultrastructural immunocytochemistry of cells incubated with DAMP in the absence or presence of monensin, show that megasomes and inclusion vacuoles have a low pH probably maintained by an active process. Furthermore, confirming the report of H. F. Hassan and G. H. Coombs (1987, Molecular and Biochemical Parasitology 23, 285-296) megasomes were found to display acid phosphatase activity at both light and electron microscope levels. This, together with the demonstration that megasomes are acidified, suggests that these organelles may be targets for amino acid derivatives.

Leishmania mexicana: a cytochemical and quantitative study of lysosomal enzymes in infected rat bone marrow-derived macrophages.

The cellular localization and activity of the lysosomal enzymes acid phosphatase, trimetaphosphatase, and arylsulfatase were studied in rat bone marrow-derived macrophages infected with Leishmania mexicana amazonensis amastigotes. The specific activity of acid phosphatase normalized for protein content was similar in normal macrophages and in isolated amastigotes, whereas the latter were markedly deficient in trimetaphosphatase and arylsulfatase activities. It is thus likely that trimetaphosphatase and arylsulfatase activities detected in infected macrophages were of host cell origin. The activities of the three enzymes, assayed biochemically, varied independently in the infected macrophages. While arylsulfatase activity was unchanged after infection, the activity of acid phosphatase increased by 19, 40, and 94% at 6, 24, and 48 hr, respectively. Trimetaphosphatase activity rose only slightly during the first 24 hr after infection but increased by 74% at 48 hr. The rise in acid phosphatase activity could be accounted for only partially by multiplication of the amastigotes. Thus, as for trimetaphosphatase, these results suggest enhanced macrophage synthesis of acid phosphatase and/or reduced enzyme degradation by the infected macrophages. The reduction in host cell lysosomes previously described (Ryter et al. 1983; Barbieri et al. 1985) was confirmed but appearance of lysosomal enzyme activity in the parasitophorous vacuole is documented in the present report. Thus, Leishmania do not seem to reduce the amount and the activity of host lysosomal enzymes.

Multiple effects of the phenylhydrazone derivative FCCP on the secretory pathway in rat plasma cells.

We studied the sensitivity of the last steps of the secretory process of antibody-producing cells to carbonylcyanide-p-trifluoromethoxyphenylhydrazone (FCCP) and sodium azide (NaN3), agents which lower the cellular ATP content by inhibiting oxidative phosphorylation and mitochondrial electron transport, respectively. Popliteal lymph node cells or purified plasma cells from rats immunized against horseradish peroxidase were incubated with the drugs. The rate of secretion of anti-HRP antibodies was measured by an enzyme-linked immunoadsorbent assay or after biosynthetic labeling with L-[3H]fucose. FCCP as well as NaN3 were shown to rapidly inhibit (in less than 5 min) the secretion of immunoglobulins (Ig) and to partially block the release of fucosylated Ig. This indicates that the drugs inhibit the transport of Ig from the Golgi apparatus (GA) (where fucose is added to Ig) to the plasma membrane. However, the degree of inhibition reached 40 to 50% with NaN3 and 70 to 80% with FCCP, whereas both drugs similarly depleted ATP stores by 45 to 55%. These results are consistent with multiple effects of FCCP on the secretion pathway of Ig. As a tentative explanation, we suggest that FCCP, because of its protonophore properties, not only reduces cellular ATP levels but may also neutralize the Golgi or post-Golgi acidic compartments recently shown to be involved in the transport of plasma membrane and secretory proteins.

Ammonium chloride, methylamine and chloroquine reversibly inhibit antibody secretion by plasma cells.

The effects of the lysosomotropic weak bases, NH4Cl, methylamine and chloroquine, on the secretory process of antibody-synthesizing cells were studied. Popliteal lymph node cells taken from rats immunized against horseradish peroxidase (HRP) were incubated with the lysosomotropic agents. The rate of secretion of anti-HRP antibodies was measured using an indirect enzyme-linked immunosorbent assay. These agents induced an inhibition of antibody release within 5 min, and for all four concentrations tested, maximal inhibition was reached after 15 min. A 50% inhibition was obtained with 20 mM NH4Cl, 21.7 mM methylamine and 8.8 X 10(-4) M chloroquine. This effect was rapidly and entirely reversible, regardless of the weak base used, and it increased as the pH of the extracellular media was raised. Under these conditions, intracellular ATP contents remained normal, and protein synthesis did not undergo marked changes except with chloroquine. Inhibition of secretion was accompanied by an intracellular accumulation of antibodies which was equal to the degree of inhibition of antibody release. Immunocytochemical studies of the weak base-treated cells performed by light and electron microscopy showed that this accumulation probably occurred within certain dilated Golgi saccules. In addition, reduced incorporation of fucose into immunoglobulins as well as partial inhibition of the secretion of fucosylated immunoglobulins were observed in the presence of weak bases. These results are consistent with the hypothesis that weak bases inhibit antibody secretion by acting within saccules of the Golgi apparatus. These saccules could maintain an acidic pH important for the migration and/or sorting of immunoglobulins.

Reversible pinocytosis of horseradish peroxidase in lymphoid cells.

A detailed study of fluid phase endocytosis of horseradish peroxidase (HRP) in rat lymph node cells (LNC) is presented in this paper. Preliminary experiments have shown that HRP was internalized by non-receptor-mediated endocytosis and interacted minimally or not at all with plasma membrane of LNC, and can then be considered as a true fluid phase marker for these cells. Kinetics of uptake of HRP was found not to be linear with incubation time at 37 degrees C and deviation from linearity can be attributed to constant exocytosis of HRP. The kinetics of exocytosis cannot be described by a single exponential process. Rather, a minimum of two exponentials is required to account for exocytosis. This suggests that at least two intracellular compartments are involved in this process. The first turns over very rapidly with a t 1/2 release of about 3 min and is saturated after 10 min of exposure with HRP. The second, which turns over very slowly, is characterized by a t 1/2 release of about 500 min and accounts for the intracellular accumulation of HRP. Similar biphasic kinetics of exocytosis were observed with unfractionated LNC, with T lymphocyte-enriched LNC and with lymphocytes purified according to their density. This suggests that most, if not all, LNC are able to release HRP and that each cell type is endowed with the two intracellular compartments. Kinetics of uptake of HRP in these two compartments indicated that they are probably filled by two endocytic pathways, at least partially independent. Taken together, these results seem to indicate that a rapid membrane recycling occurs in lymphocytes. Furthermore, the weak base ammonium chloride and the carboxylic ionophore monensin were shown in our study to inhibit fluid phase endocytosis of HRP. The inhibition was time-dependent and required a preincubation of the cells with the drugs to be observed. Our results suggest that a perturbation of the vesicular traffic or a sequestration of membranes involved in HRP uptake is induced by these drugs. Under these conditions the release of cell-associated HRP was also reduced and to the same extent as the inhibition of uptake. Distribution of HRP between the two compartments and the t 1/2 release of HRP from either compartment were not perturbed. Taken together these results seem to indicate that exocytosis is not specifically affected by these drugs. Inhibition of uptake in drug-treated cells could result from a general decrease of membrane recycling or to the formation of smaller pinocytic vesicles with a different surface to volume ratio.

An amplification system using BSA-antibody conjugate for sensitive enzyme immunoassay.

A procedure is described for sensitive titration of antibodies, macromolecular antigens and haptens by enzyme immunoassay. It involves using first antigen or antibody labelled with bovine serum albumin (BSA) and then an anti-BSA antibody conjugated with an enzyme. The performance characteristics of this assay are indicated and compared with those for conventional enzyme immunoassay. The present procedure allowed fast sensitive titration of human IgE, rabbit type III anti-streptococcal antibody and cAMP.