New function of type I IFN: induction of autophagy

Autophagy is a highly conserved cellular process responsible for recycling of intracellular material. It is induced by different stress signals, including starvation, cytokines, and pathogens. Type I interferons (IFN) are proteins with pleiotropic functions, such as antiviral, antiproliferative, and immunomodulatory activities. Several recent studies showed type I IFN-induced autophagy in multiple cancer cell lines as evidenced by autophagic markers, for example, the conversion of microtubule-associated protein 1 light chain 3 beta (MAP1LC3B, also known as LC3-I) to LC3-II and the formation of autophagosomes by electron microscopy. In addition, studies suggest the involvement of Janus kinase (JAK)/signal transducer and activator of transcription (STAT) and phosphatidylinositol-4,5-bisphosphate 3-kinase (PI3K)/v-akt murine thymoma viral oncogene homolog (AKT) and mechanistic target of rapamycin, serine/threonine kinase (mTOR) pathways in the induction of autophagy. This review highlights a new function of type I IFN as an inducer of autophagy. This new function of type I IFN may play an important role in viral clearance, antigen presentation, inhibition of proliferation, as well as a positive feedback loop for the production of type I IFN.

Immunomodulatory effects of interferons in malignancies

Investigation of the antitumor and immunomodulatory activities of interferon (IFN) began shortly after the cytokine was discovered in 1957. Early work showed a direct correlation between administration of IFN and inhibition of symptoms associated with virally induced leukemia in mice as well as an increase in their survival time. Subsequent studies with purified IFNs confirmed the direct and indirect stimulation of immune cells, resulting in antitumor activities of IFN. Clinically, IFN-alphas (αs) have been shown to have activity against a variety of tumors. Initially, the U.S. Food and Drug Administration licensed 2 recombinant IFN-αs for the treatment of hairy-cell leukemia and then later for several other cancers. The success rate seen with IFNs and certain tumors has been varied. Unfortunately, some neoplasms show no response to IFN. Monocytes/macrophages play an important role in cancer progression. Monocytes in combination with IFN may be an important therapy for several cancers. This article focuses on the role of IFN and monocytes alone or in combination in affecting malignancies.

Induction of autophagy by Type I interferon in human cancer cells (P6217)

Type I IFNs (interferons) are cytokines that induce an antiviral state in cells and have antiproliferative, pro-apoptotic, and immunomodulatory activities. Autophagy is a basic cellular mechanism that involves degradation of cellular components via lysosomes. While induction of autophagy is well documented for Type II IFN, induction of autophagy by Type I IFN has not been previously reported. We determined that Type I IFN can induce autophagy in multiple human cancer cell lines. IFN-α2c and IFN-β induces autophagy in Daudi B cells, as indicated by an increase in autophagy markers LC3-II, Atg5-Atg12 complexes, and a decrease in p62. An increase in LC3-II was also detected post IFN-α2c treatment in HeLa S3, MDA-MB-231, T98G and A549 cell lines. The presence of autophagosomes in selected cell lines was confirmed by electron microscopy analysis. An increase in autophagy markers correlated with inhibition of mTORC1 activity and cell proliferation as well as changes in the cell cycle progression in Daudi cells. Treatment of Daudi and T98G cells with IFN-α2c and rapamycin (mTOR inhibitor), or LY294002 (PI3K inhibitor) increased the level of LC3-II, indicating that the PI3K/Akt/mTORC1 signaling pathway may affect IFN-induced autophagy in Daudi and T98G cells. The role of mTOR in Type I IFN-induced autophagy was confirmed by siRNA silencing experiments. Taken together, our findings demonstrate a novel function for Type I IFN as inducer of autophagy in multiple cancer cell lines.

Type I interferons induce autophagy in certain human cancer cell lines

Autophagy is an evolutionarily conserved cellular recycling mechanism that occurs at a basal level in all cells. It can be further induced by various stimuli including starvation, hypoxia, and treatment with cytokines such as IFNG/IFNγ and TGFB/TGFβ. Type I IFNs are proteins that induce an antiviral state in cells. They also have antiproliferative, proapoptotic and immunomodulatory activities. We investigated whether type I IFN can also induce autophagy in multiple human cell lines. We found that treatment with IFNA2c/IFNα2c and IFNB/IFNβ induces autophagy by 24 h in Daudi B cells, as indicated by an increase of autophagy markers MAP1LC3-II, ATG12–ATG5 complexes, and a decrease of SQSTM1 expression. An increase of MAP1LC3-II was also detected 48 h post-IFNA2c treatment in HeLa S3, MDA-MB-231, T98G and A549 cell lines. The presence of autophagosomes in selected cell lines exposed to type I IFN was confirmed by electron microscopy analysis. Increased expression of autophagy markers correlated with inhibition of MTORC1 in Daudi cells, as well as inhibition of cancer cell proliferation and changes in cell cycle progression. Concomitant blockade of either MTOR or PI3K-AKT signaling in Daudi and T98G cells treated with IFNA2c increased the level of MAP1LC3-II, indicating that the PI3K-AKT-MTORC1 signaling pathway may modulate IFN-induced autophagy in these cells. Taken together, our findings demonstrated a novel function of type I IFN as an inducer of autophagy in multiple cell lines.

Potent antitumor effects of combination therapy with IFNs and monocytes in mouse models of established human ovarian and melanoma tumors

Interferon-activated monocytes are known to exert cytocidal activity against tumor cells in vitro. Here, we have examined whether a combination of IFN-α2a and IFN-γ and human monocytes mediate significant antitumor effects against human ovarian and melanoma tumor xenografts in mouse models. OVCAR-3 tumors were treated i.t. with monocytes alone, IFN-α2a and IFN-γ alone or combination of all three on day 0, 15 or 30 post-tumor implantation. Mice receiving combination therapy beginning day 15 showed significantly reduced tumor growth and prolonged survival including complete regression in 40% mice. Tumor volumes measured on day 80 in mice receiving combination therapy (206 mm(3)) were significantly smaller than those of mice receiving the IFNs alone (1,041 mm(3)), monocytes alone (1,111 mm(3)) or untreated controls (1,728 mm(3)). Similarly, combination therapy with monocytes and IFNs of much larger tumor also inhibited OVCAR-3 tumor growth. Immunohistochemistry studies showed a large number of activated macrophages (CD31(+)/CD68(+)) infiltrating into OVCAR-3 tumors and higher densities of IL-12, IP10 and NOS2, markers of M1 (classical) macrophages in tumors treated with combination therapy compared to the controls. Interestingly, IFNs-activated macrophages induced apoptosis of OVCAR-3 tumor cells as monocytes alone or IFNs alone did not mediate significant apoptosis. Similar antitumor activity was observed in the LOX melanoma mouse model, but not as profound as seen with the OVCAR-3 tumors. Administration of either mixture of monocytes and IFN-α2a or monocytes and IFN-γ did not inhibit Lox melanoma growth; however, a significant inhibition was observed when tumors were treated with a mixture of monocytes, IFN-α2a and IFN-γ. These results indicate that monocytes and both IFN-α2a and IFN-γ may be required to mediate profound antitumor effect against human ovarian and melanoma tumors in mouse models.

Structural variants of IFNα preferentially promote antiviral functions

IFNα, a cytokine with multiple functions in innate and adaptive immunity and a potent inhibitor of HIV, exerts antiviral activity, in part, by enhancing apolipoprotein B mRNA-editing enzyme-catalytic polypeptide-like 3 (APOBEC3) family members. Although IFNα therapy is associated with reduced viral burden, this cytokine also mediates immune dysfunction and toxicities. Through detailed mapping of IFNα receptor binding sites, we generated IFNα hybrids and mutants and determined that structural changes in the C-helix alter the ability of IFN to limit retroviral activity. Selective IFNα constructs differentially block HIV replication and their directional magnitude of inhibition correlates with APOBEC3 levels. Importantly, certain mutants exhibited reduced toxicity as reflected by induced indoleamine 2,3-dioxygenase (IDO), suggesting discreet and shared intracellular signaling pathways. Defining IFN structure and function relative to APOBEC and other antiviral genes may enable design of novel IFN-related molecules preserving beneficial antiviral roles while minimizing negative effects.

Near eradication of clinically relevant concentrations of human tumor cells by interferon-activated monocytes in vitro

We have previously reported that low concentrations of interferon (IFN)-activated monocytes exert near-eradicative cytocidal activity against low concentrations of several human tumor cells in vitro. In the present study, we examined 7 human tumor cell lines and 3 diploid lines in the presence or absence of 10 ng/mL IFNα2a and monocytes. The results confirmed strong cytocidal activity against 4 of 7 tumor lines but none against 3 diploid lines. To model larger in vivo tumors, we increased the target cell concentration and determined the concentration of IFNα2a and monocytes, required for cell death. We found that increasing the tumor cell concentration from 10- to 100-fold (10(5) cells/well) required an increase in the concentration of IFNs by over 100-fold and monocytes by 10-fold. High concentrations of monocytes could sometimes kill tumor or diploid cells in the absence of IFN. We may conclude that killing of high concentrations of tumor or diploid cells required high concentrations of monocytes that could sometimes kill in the absence of IFN. Thus, high concentrations of tumor cells required high concentrations of IFN and monocytes to cause near eradication of tumor cells. These findings may have clinical implications.

Structural Determinants of IFNα Distinguish Antiviral and Potentially Toxic Functions. (39.7)

Interferon alpha (IFNα), a cytokine with multiple functions in innate and adaptive immunity and a potent inhibitor of human immunodeficiency virus (HIV), exerts antiviral activity, in part, by enhancing apolipoprotein B mRNA-editing enzyme-catalytic polypeptide-like 3(APOBEC3) family members. Although IFNα therapy is associated with reduced viral burden, this cytokine also mediates immune dysfunction and toxicities. Through detailed mapping of IFNα receptor binding sites, we generated IFNα hybrids and mutants and determined that structural changes in the C-loop influencing receptor interactions alter IFN’s ability to limit retroviral replication. Selective IFNα constructs differentially block HIV replication and their directional magnitude of inhibition correlates with elevated APOBEC3 levels with less significant changes in transcription of other antiviral molecules such as TRIM22 and PKR. Besides participation of the JAK-STAT pathway, we uncover previously unappreciated roles for NFκB, CAMK and PI3K signaling pathways in induction of APOBEC3 by the IFN constructs. Importantly, certain mutants exhibited reduced toxicity as reflected by induced indoleamine 2,3-dioxygenase and TNFα compared with intact IFN, suggesting discreet and shared intracellular signaling pathways. Defining IFN structure and function relative to APOBEC and other antiviral genes may enable the design of novel IFN-related molecules preserving beneficial antiviral roles while minimizing negative effects.

Antiproliferative Properties of Type I and Type II Interferon

The clinical possibilities of interferon (IFN) became apparent with early studies demonstrating that it was capable of inhibiting tumor cells in culture and in vivo using animal models. IFN gained the distinction of being the first recombinant cytokine to be licensed in the USA for the treatment of a malignancy in 1986, with the approval of IFN-α2a (Hoffman-La Roche) and IFN-α2b (Schering-Plough) for the treatment of Hairy Cell Leukemia. In addition to this application, other approved antitumor applications for IFN-α2a are AIDS-related Kaposi's Sarcoma and Chronic Myelogenous Leukemia (CML) and other approved antitumor applications for IFN-α2b are Malignant Melanoma, Follicular Lymphoma, and AIDS-related Kapoisi's Sarcoma. In the ensuing years, a considerable number of studies have been conducted to establish the mechanisms of the induction and action of IFN's anti-tumor activity. These include identifying the role of Interferon Regulatory Factor 9 (IRF9) as a key factor in eliciting the antiproliferative effects of IFN-α as well as identifying genes induced by IFN that are involved in recognition of tumor cells. Recent studies also show that IFN-activated human monocytes can be used to achieve >95% eradication of select tumor cells. The signaling pathways by which IFN induces apoptosis can vary. IFN treatment induces the tumor suppressor gene p53, which plays a role in apoptosis for some tumors, but it is not essential for the apoptotic response. IFN-α also activates phosphatidylinositol 3-kinase (PI3K), which is associated with cell survival. Downstream of PI3K is the mammalian target of rapamycin (mTOR) which, in conjunction with PI3K, may act in signaling induced by growth factors after IFN treatment. This paper will explore the mechanisms by which IFN acts to elicit its antiproliferative effects and more closely examine the clinical applications for the anti-tumor potential of IFN.

Clinical model: interferons activate human monocytes to an eradicative tumor cell level in vitro

Eradicative levels of antitumor activity by cytokines and leukocytes have not yet been reached experimentally and are needed clinically. Only a limited number of human cancers respond to therapy with interferon (IFN), other cytokines, or mononuclear leukocytes despite significant antitumor activity in vitro. We studied the IFN and monocytic cell conditions that would lead to an eradicative effect using human cells in vitro. Targets of the IFN-activated monocytic cells were either four human tumor cell lines (human osteosarcoma [HOS], LOX melanoma, A549 lung tumor, and SNB-19 glioblastoma) or two diploid cell lines (WI38 and MRC5). An average of 30-90 colony-forming tumor target cells were cultured overnight in 96-well tissue culture plates prior to treatment with serially diluted IFN with or without activated elutriation-purified monocytes or lymphocytes. The target cell colonies were treated for 3 days. The colonies were then stained with crystal violet to determine the levels of antitumor activity. IFN-activated human monocytes reached an eradicative level (95%-100%) against three of four tumor cell lines. The eradicative level (1) was induced best in human monocytes activated by combined type I and II IFNs, (2) was effective against tumor cells that were growing for 24 h, (3) was specific for human tumors, as diploid human cells were not inhibited, and (4) required contact between the macrophage and the tumor cells. Also, for the first time, the minimal effective concentration (MEC) of IFNs to activate monocytes can approach those needed for antiviral activity. To our knowledge, this is the first report of near total eradication of many tumor cells, but not diploid cells, by IFN-activated monocytes. Because of its potency and specificity, the IFN-activated monocyte arm of the innate immune system may be a candidate for therapy of established tumors.

Attenuating mutations in the P/C gene of human parainfluenza virus type 1 (HPIV1) vaccine candidates abrogate the inhibition of both induction and signaling of type I interferon (IFN) by wild-type HPIV1

Recombinant human parainfluenza virus type 1 (HPIV1) and mutants containing point and deletion (Delta) mutations in the P/C gene (r-CDelta10-15HNT553A, r-CR84G, r-CF170S and r-CDelta170), which have previously been evaluated as HPIV1 vaccine candidates, were evaluated for their effect on the type I interferon (IFN) response in vitro. HPIV1 wt infection inhibited the IFN response by inhibiting IFN regulatory factor-3 (IRF-3) activation and IFN production in A549 cells and IFN signaling in Vero cells. In contrast, r-CR84G, r-CF170S and r-CDelta170 were defective for inhibition of IRF-3 activation and IFN production and r-CF170S and r-CDelta170 did not inhibit IFN signaling. Thus, HPIV1 antagonizes the IFN response at both the level of induction and signaling, and antagonism at both levels was disrupted by mutations in the P/C gene. Because CF170S affects C and not P, the anti-IFN function can be attributed to the C proteins. These data, in the context of previous in vivo studies, suggest that the loss of antagonism of the IFN response at both the level of induction and signaling, observed with the P/C mutants, r-CF170S and r-CDelta170, was necessary for significant attenuation in African green monkeys (AGMs).

Protein engineering of interferon alphas

Interferon (IFN)-alphas constitute a family of proteins exhibiting high degree of homology in primary, secondary, and tertiary structure and display a high level of species specificity in their biological properties. However, small structural differences in these proteins may be responsible for a significant variety of biological actions. Understanding the structure and function of human IFN-alpha is very important. Recombinant techniques are important tools for the production and modification of IFN proteins. The first IFN hybrid, IFN-alpha1/alpha2 was constructed using recombinant technology in 1981. Subsequently, a number of IFN hybrids and mutants have been constructed, expressed and characterized. These hybrids and mutants have resulted in novel IFNs that either combine different biological properties from the parental proteins or have significantly different biological activity. Therefore, IFN hybrids and mutants have provided a powerful tool for studying the structure and function of these molecules. Also, these engineered IFNs may have important new therapeutic applications and may provide greater sights into understanding of the clinical activities of these molecules.

Gateway cloning is compatible with protein secretion from Pichia pastoris

Secretion of a recombinant protein from the yeast Pichia pastoris requires the presence of a signal peptide at the amino terminus. Maintaining the full amino acid sequence of the signal peptide is thought to be important for proper signal processing and protein secretion. We show that at least for one protein, a synthetic human interferon, the presence of a Gateway recombination site within the signal peptide is fully compatible with high levels of protein secretion. The amino termini of the secreted interferon proteins cloned with Gateway and cloned with restriction enzymes and ligase are identical, and the proteins were highly active in biological assays. Compatibility with Gateway cloning simplifies construction of plasmids directing secretion of recombinant proteins from P. pastoris.

Human interferons alpha, beta and omega

Type I interferons (IFNs), IFN-alpha, IFN-beta, IFN-omega, IFN-delta and IFN-tau are a family of structurally related, species-specific proteins found only in vertebrates. They exhibit a variety of biological functions, including antiviral, antiproliferative, immunomodulatory and developmental activities. Human Type I IFNs interact with the human IFN alpha receptor (IFNAR), which is composed of two identified subunits (IFNAR-1 and IFNAR-2). The interaction of IFN-alpha/beta with its receptor components results in the activation of a number of signaling pathways. The regulation of specific genes and proteins contributes to the numerous biological functions of Type I IFNs.

Amino acid substitutions in loop BC and helix C affect antigenic properties of helix D in hybrid IFN-alpha21a/alpha2c molecules

We compared the antigenic properties of human interferon-alpha2c (IFN-alpha2c), IFN-alpha21a, hybrids IFN-alpha21a/alpha2c, and their mutants, using a panel of 27 anti-IFN-alpha1, anti-IFN-alpha2, and anti-IFN-alpha8/1/8 monoclonal antibodies (mAb). After immunoanalysis by ELISA, we found parental IFN-alpha2c and IFN-alpha21a to be antigenically distinct. Lack of reactivity of anti-IFN-alpha1 mAb with IFN-alpha21a indicated an antigenic distinction between subtypes alpha1 and alpha21a. The antigenic properties of hybrid IFNs consisting of the N-terminal portion (1-75) of IFN-alpha21a and the C-terminal portion (76-166) of IFN-alpha2c were analyzed with mAb recognizing defined regions of IFN-alpha2c, IFN-alpha1, and IFN-alpha8/1/8. We found that extending the sequence of IFN-alpha21a up to position 95 in hybrid molecule decreased the immunoreactivity of mAb specific for the antigenic structure formed by residues --112-132-- (helix D) of IFN-alpha2c. Inserting the sequence 76-81 (loop BC) of IFN-alpha2c into the sequence of 1-95 of IFN-alpha21a restored the reactivity of anti-IFN-alpha2c mAb. Some amino acid substitutions at positions 86 and 90 (helix C) of hybrid IFN-alpha21a/alpha2c also affected the immunoreactivity of C-terminal-specific mAb, which recognize helix D, but did not influence the structure of C-terminus of IFN (aa 151-165). Changes in the structure of constructs affected not only their antiproliferative activity but also their antiviral activity on human cells.

Human IFN-alpha protein engineering: the amino acid residues at positions 86 and 90 are important for antiproliferative activity

Human IFN-alpha is a family of structurally related proteins that exhibit a wide range of antiproliferative activities. To understand the structural basis for these different antiproliferative activities, eight recombinant human IFN-alpha hybrids (HY) of alpha21a/alpha2c (HY-4, HY-5) and mutants (site-directed mutagenesis (SDM)-1, 2 and cassette mutagenesis (CM)-1, 2, 3, and 4) have been expressed, purified, and characterized. The data showed that the amino acid region 81-95 is important for antiproliferative activity. Site-directed mutagenesis and cassette mutagenesis studies showed that if serine (S) 86 and asparagine (N) 90 were replaced by tyrosine (Y), the antiproliferative activity was increased. We have also observed that if Y86 was replaced by isoleucine (I), the antiproliferative activity was comparable. However, if Y86 was replaced by aspartic acid (D), lysine (K), or alanine (A), the antiproliferative activity was substantially decreased. Our results indicate that Y and/or I at position 86 and Y at position 90 are very important in antiproliferative activity of human IFN-alpha. Circular dichroism spectra showed that the amino acid replacements at position 86 did not change the secondary structure. Thus the biological activity changes among those mutants do not appear to be due to conformational changes. The results also suggest that hydrophobic residue(s) at position 86 may be important for the interaction of the molecule with its receptor. The competitive binding data correlated with the antiproliferative activity. The N-terminal region of the molecule and the hydrophobic residues (including Y and I) on the C-helix region at positions 86 and/or 90 are important for binding and antiproliferative activities of human IFN-alphas.

Divergence of binding, signaling, and biological responses to recombinant human hybrid IFN

Three human IFN-alpha hybrids, HY-1 [IFN-alpha21a(1-75)/alpha2c(76-165)], HY-2 [IFN-alpha21a(1-95)/alpha2c(96-165)], and HY-3 [IFN-alpha2c(1-95)/alpha21a(96-166)], were constructed, cloned, and expressed. The hybrids had comparable specific antiviral activities on Madin-Darby bovine kidney (MDBK) cells but exhibited very different antiproliferative and binding properties on human Daudi and WISH cells and primary human lymphocytes. Our data suggest that a portion of the N-terminal region of the molecule is important for interaction with components involved in binding of IFN-alpha2b while the C-terminal portion of IFN is critical for antiproliferative activity. A domain affecting the antiproliferative activity was found within the C-terminal region from amino acid residues 75-166. The signal transduction properties of HY-2 and HY-3 were evaluated by EMSA and RNase protection assays. Both HY-2 and HY-3 induced activation of STAT1 and 2. However, HY-2 exhibited essentially no antiproliferative effects at concentrations that activated STAT1 and 2. Additionally, at concentrations where no antiproliferative activity was seen, HY-2 induced a variety of IFN-responsive genes to the same degree as HY-3. RNase protection assays also indicate that, at concentrations where no antiproliferative activity was seen for HY-2, this construct retained the ability to induce a variety of IFN-inducible genes. These data suggest that the antiproliferative response may not be solely directed by the activation of the STAT1 and STAT2 pathway in the cells tested.

Divergence of Binding, Signaling, and Biological Responses to Recombinant Human Hybrid IFN

Three human IFN-α hybrids, HY-1 [IFN-α21a(1-75)/α2c(76-165)], HY-2 [IFN-α21a(1-95)/α2c(96-165)], and HY-3 [IFN-α2c(1-95)/α21a(96-166)], were constructed, cloned, and expressed. The hybrids had comparable specific antiviral activities on Madin-Darby bovine kidney (MDBK)3 cells but exhibited very different antiproliferative and binding properties on human Daudi and WISH cells and primary human lymphocytes. Our data suggest that a portion of the N-terminal region of the molecule is important for interaction with components involved in binding of IFN-α2b while the C-terminal portion of IFN is critical for antiproliferative activity. A domain affecting the antiproliferative activity was found within the C-terminal region from amino acid residues 75–166. The signal transduction properties of HY-2 and HY-3 were evaluated by EMSA and RNase protection assays. Both HY-2 and HY-3 induced activation of STAT1 and 2. However, HY-2 exhibited essentially no antiproliferative effects at concentrations that activated STAT1 and 2. Additionally, at concentrations where no antiproliferative activity was seen, HY-2 induced a variety of IFN-responsive genes to the same degree as HY-3. RNase protection assays also indicate that, at concentrations where no antiproliferative activity was seen for HY-2, this construct retained the ability to induce a variety of IFN-inducible genes. These data suggest that the antiproliferative response may not be solely directed by the activation of the STAT1 and STAT2 pathway in the cells tested.

Identification of cysteine 354 of beta-tubulin as part of the binding site for the A ring of colchicine

The colchicine analog 3-chloroacetyl-3-demthylthio-colchicine (3CTC) is a competitive inhibitor of colchicine binding to tubulin, binds to tubulin at 37 degrees C, but not at 0 degree C, and covalently reacts with beta-tubulin at 37 degree C, but not at 0 degree C, in a reaction inhibited by colchicine site drugs. The approximate intramolecular distance between the oxygen at position C-3 in 3CTC and the chlorine atom of the 3-chloroacetyl group is 3 A. using decylagarose chromatography, we purified beta-tubulin that had reacted with 3-(chloromethyl-[14C] Carbonyl)-3- demethylthiocolchicine ([14C]3CTC). This beta-tubulin that had reacted with 3-(chloromethyl-[14C]carbonyl)- 3-demethythiocolchicine ([14C]3CTC). This beta-tubulin was digested with formic acid, cyanogen bromide, endoproteinase Glu-C, or endoproteinase Lys-C, and the radio-labeled peptide(s) were isolated. The sequences of these peptides indicated that as much as 90% of the covalent reaction between the [14C]3CTC and beta-tubulin occurred at cysteine 354. This finding indicates that the C-3 oxygen atom of colchicinoids is within 3 A of the sulfur atom of the Cys-354 residue, suggests that the colchicine A ring lies between Cys-354 and Cys-239, based on the known 9 A distance between these residues, and may indicate that the tropolone C ring lies between the peptide region containing Cys-239 and the amino-terminal beta-tubulin sequence, based on the labeling pattern observed following direct photoactivation of tubulin-bound colchicine.

Purification and characterization of multiple components of human lymphoblastoid interferon-alpha

Twenty-two components of human interferon-alpha (IFN-alpha) derived from Sendai virus-induced Namalwa cells were purified by sequential immunoadsorbent affinity chromatography using four monoclonal antibody affinity columns followed by ultrafiltration and reversed-phase high-performance liquid chromatography. The specific activity ranged from 0.2 to 2.6 x 10(8) IU/mg protein on Madin-Darby bovine kidney cells, 0.3 to 4.6 x 10(8) IU/mg protein on human WISH cells, and 10(4) to 7 x 10(5) units/mg protein on mouse L929 cells. The apparent molecular weights of the components ranged from 17,500 to 23,300 using nonreducing sodium dodecyl polyacrylamide-gel electrophoresis and 17,500 to 27,600 using reducing sodium dodecyl polyacrylamide-gel electrophoresis. The amino-terminal amino acid sequences were similar among the components as well as to those reported for the cloned human IFN-alpha genes (Pestka, S. (1986) Methods Enzymol. 119, 3-14). However, four components, f, i, l, and m, have amino-terminal amino acid sequences which appear to be unique when compared to those predicted from the cDNA clones. One component, pre-a, has a potential N-linked glycosylation site on the Asn of residues 2 through 4, Asn-Leu-Ser.

Purification and characterization of a human Mx protein

Human interferon-beta (IFN-beta) induces in human embryonic foreskin fibroblasts a cytoplasmic protein with antigenic similarities to mouse Mx protein, a nuclear protein implicated in inhibition of influenza virus replication. The human protein was purified to virtual homogeneity by immunoaffinity chromatography using a monoclonal antibody to mouse Mx protein. The purified protein has an apparent Mr of 78,000 and displays a strong tendency to self-aggregate. It can be resolved on two-dimensional gels into four spots with pIs between 6.0 and 6.2, each of which reacts with antibodies to mouse Mx protein. The partial amino-terminal sequence was determined for the affinity-purified protein. Cytoplasmic microinjection of the affinity-purified protein does not lead to efficient protection against infection with influenza virus. Cytoplasmic microinjection of the monoclonal Mx antibody, which increases suceptibility of IFN-treated mouse Mx cells to influenza virus, does not alter the viral susceptibility of IFN-treated human cells. These results suggest that, unlike the mouse Mx protein, the human Mx protein studied in this communication may not be sufficient to confer to cells a high degree of protection against influenza virus.

A mouse monoclonal antibody that reacts specifically with immune-complexed human IgG

After human IgG binds to antigen, it attains biological functions that are not properties of monomeric, uncomplexed IgG, including the ability to activate complement and to bind to cellular receptors. Associated with antigen binding, we have recently demonstrated that IgG itself has neoantigenic epitopes. Antibodies to these neoantigens on immune-complexed IgG may represent a significant proportion of circulating anti-human IgG in rabbits immunized with immune complexes. In contrast, mice immunized in an identical fashion have very little circulating anti-neoantigen antibody. This is true whether the mice are genetically easy to tolerize to monomeric human IgG (DBA/2 and C57Bl/6) or difficult to tolerize (BALB/c). Fusions were made between the NS-1 myeloma cell line and spleen cells from mice of each strain, which had been made tolerant to monomeric human IgG and then immunized with immune complexes containing IgG. Like the serum antibody, antibodies made by these fusions showed little specificity for immune complexes since 99% of the hybridoma antibodies that recognized IgG in immune complexes also bound to uncomplexed IgG. Only 1 hybridoma produced antibody that preferentially recognized human IgG in immune complexes. This antibody, called CE9, is an IgM that binds to IgG in plate-bound immune complexes with 100-1000-fold greater avidity than it does to plate-bound uncomplexed IgG. Because CE9 will not bind to immune complexes made with F(ab')2 antibody, the epitope it recognizes requires the Fc fragment of IgG. The minimal binding of CE9 to uncomplexed IgG is easily inhibited by soluble aggregates of IgG, but binding to immune complexes is not inhibited by aggregated IgG. CE9 does recognize fluid-phase immune complexes as well as solid-phase immune complexes. We conclude that, while mice produce much less anti-immune complex antibody than rabbits, anti-neoantigen is still a component of their response to immunization with immune complexes. Using hybridoma techniques to amplify these anti-neoantigen antibodies, we have shown that they resemble rheumatoid factors in their isotype and binding properties.

Neoantigens appear in human IgG upon antigen binding: detection by antibodies that react specifically with antigen-bound IgG

We raised rabbit antibodies that specifically recognize antigen-bound but not monomeric human IgG. These rabbit IgG antibodies (RAb) were induced in rabbits that were made tolerant to monomeric human IgG. They bound to immune complexes (IC) made with human IgG and various antigens including tetanus toxoid, sheep erythrocytes (E), rabbit E, or human Rh(D) + E, and were very poorly inhibitable with monomeric IgG compared to conventional rabbit anti-human IgG. RAb did not recognize complement components bound to the IgG containing IC. Cleavage of the Fc domain from human IgG markedly decreased binding of RAb to IC. Surprisingly, RAb did not bind to heat-aggregated human IgG (agg-IgG) better than to monomeric IgG. We conclude that human IgG expresses an Fc neoantigen when it binds its own antigen, and that this determinant is not expressed by agg-IgG. The implications of these findings for the biologic functions of antigen-complexed IgG are discussed.

Neoantigens appear in human IgG upon antigen binding: detection by antibodies that react specifically with antigen-bound IgG

We raised rabbit antibodies that specifically recognize antigen-bound but not monomeric human IgG. These rabbit IgG antibodies (RAb) were induced in rabbits that were made tolerant to monomeric human IgG. They bound to immune complexes (IC) made with human IgG and various antigens including tetanus toxoid, sheep erythrocytes (E), rabbit E, or human Rh(D) + E, and were very poorly inhibitable with monomeric IgG compared to conventional rabbit anti-human IgG. RAb did not recognize complement components bound to the IgG containing IC. Cleavage of the Fc domain from human IgG markedly decreased binding of RAb to IC. Surprisingly, RAb did not bind to heat-aggregated human IgG (agg-IgG) better than to monomeric IgG. We conclude that human IgG expresses an Fc neoantigen when it binds its own antigen, and that this determinant is not expressed by agg-IgG. The implications of these findings for the biologic functions of antigen-complexed IgG are discussed.