Structural studies on equine chorionic gonadotropin

The amino acid sequence of the subunit of equine chorionic gonadotropin (eCG, also pregnant mare serum gonadotropin, PMSG) has been determined. Overlapping peptides from tryptic and chymotrypic digests were isolated by a two-dimensional peptide mapping technique and sequenced by the Edman procedure. The proposed amino acid sequence of eCG is: (**Denotes carbohydrate attachment points.) This sequence differs significantly from that proposed by Rathnamet al. (1978) for equine follitropin subunit; in particular, their sequence lacked the first fourteen residues.For the subunit we have placed in sequence 104 amino acid residues by direct sequence determination and peptide overlap procedures; in addition, 37 residues have been placed provisionally by homology with the human chorionic gonadotropin (hCG) sequence and composition and/or sequence data for the peptides isolated in the present studies. Difficulties in the procurement of the hormone have stalled completion of the -subunit amino acid sequence determination. The data now available indicate that eCG -subunit is highly homologous to hCG subunit and the subunits of luteinizing hormone from the pituitary gland of the several species so far described. The proposed partial sequence of eCG is:     

Beta-subunits of equine chorionic gonadotropin and lutenizing hormone with an identical amino acid sequence have different asparagine-linked oligosaccharide chains.

The glycoprotein hormones, equine chorionic gonadotropin (eCG) and lutenizing hormone (eLH), possess a beta-subunit with an identical amino acid sequence. The Asn-linked oligosaccharide chains of eCG beta and eLH beta were quantitatively liberated as tritium-labeled oligosaccharides by hydrazinolysis followed by N-acetylation and NaB3H4-reduction. Paper electrophoresis in combination with sialidase digestion and solvolytic desulfation indicated that eCG beta contained neutral and sialylated oligosaccharides, while eLH beta contained neutral, sialylated, sulfated, and both sialylated and sulfated oligosaccharides. In addition, elution profiles on a Bio-Gel P-4 column of the neutralized oligosaccharide mixtures of eCG beta and eLH beta were different, indicating that the molecular masses of oligosaccharides of the two glycoproteins are different. Therefore, this suggests that the structures of the Asn-linked oligosaccharide chains of eCG beta and eLH beta are different although they have an identical amino acid sequence.     

A nicked beta-subunit of human chorionic gonadotropin purified from pregnancy urine.

Human chorionic gonadotropin (hCG) is a glycoprotein hormone composed of two dissimilar subunits (alpha and beta) and normally excreted in urine of pregnant women. An uncommon beta-subunit of hCG was purified from fresh early normal pregnancy urine by Sepralyte C8, resin adsorption. Sephadex G-100 column chromatography, and reverse-phase HPLC. SDS-PAGE under non-reducing conditions showed that the apparent molecular weight (39,000) of this beta-subunit was extremely similar to that of the native beta-subunit, which is known to consist of 145 amino acid residues and carbohydrates. However, SDS-PAGE, under reducing conditions, resulted in two bands with apparent molecular weights of 22,000 and 18,000, indicating that it consisted of two peptide fragments connected with disulfide bridge(s). These two peptide fragments, separated and purified from the reduced and carboxymethylated protein, were subjected to amino acid and N-terminal sequence analyses. It was found that this beta-subunit consisted of two polypeptide chains composed of residues 1-47 disulfide-bridged to residues 48-145 of the beta-subunit, which may be produced by nicking of the beta-subunit at the one site (Gly47-Val48). This beta-subunit was termed a nicked beta-subunit of hCG (N-hCG beta). It was also found that N-hCG beta was present in urine as an alpha beta dimer, indicating that an intrachain nicking of this site in the beta-subunit does not inhibit alpha beta dimer formation.(ABSTRACT TRUNCATED AT 250 WORDS)     

Demonstration of a COOH-terminal extension on equine lutropin by means of a common acid-labile bond in equine lutropin and equine chorionic gonadotropin

The beta subunits of equine lutropin and equine chorionic gonadotropin were incubated in 0.013 N HCl for 30 min at 110 degrees C and separated into two fragments by reverse-phase high performance liquid chromatography. The amino acid and carbohydrate compositions of both fragments from each subunit were analyzed. The results demonstrated that equine lutropin-beta has a glycosylated COOH-terminal extension that differs only in carbohydrate composition from the COOH-terminal portion of equine chorionic gonadotropin-beta. This is the first demonstration of a glycosylated COOH-terminal extension in a pituitary glycoprotein hormone.     

Structural studies on equine glycoprotein hormones. Amino acid sequence of equine lutropin beta-subunit

The amino acid sequence was determined for equine lutropin beta (eLH beta). Large fragments were derived from reduced, carboxymethylated eLH beta by digestion with Staphylococcus aureus V8 protease, by cyanogen bromide cleavage, and by cleavage of acid-labile Asp-Pro bonds. The fragments were purified by gel filtration and high performance liquid chromatography (HPLC). The fragments were sequenced by automated Edman degradation to establish the primary structure of eLH beta. Some peptides were further digested with chymotrypsin and the resulting peptides purified by HPLC. In addition to sequencing by automated Edman degradation, these were also sequenced by the complementary 5-dimethylaminonaphthalene-1-sulfonyl-Edman procedure which enabled us to directly identify glycosylated amino acids. The eLH beta subunit is a glycoprotein of 149 amino acids containing both N- and O-linked oligosaccharides. It possesses a COOH-terminal extension similar to that seen in human chorionic gonadotropin. Carboxypeptidase Y digestions suggest that the COOH terminus is blocked by glycosylation. Interestingly, the amino acid sequence of eLH beta is identical to that of equine chorionic gonadotropin beta (Sugino, H., Bousfield, G. R., Moore, W. T., and Ward, D. N. (1987)J. Biol. Chem. 262, 8603-8609).     

Recognition of the beta-subunit of human chorionic gonadotropin and sub-determinants by target tissue receptors.

The beta-subunit of human chorionic gonadotropin, purified immunochemically to eliminate undissociated human chorionic gonadotropin, induced testosterone production by mouse Leydig cells at concentrations 400-fold higher than human chorionic gonadotropin. Steroidogenesis was also stimulated by a synthetic fragment of the beta-subunit of human chorionic gonadotropin conforming to the peptide sequence residues 39--71, whereas peptide sequence residues 39--56 and three C-terminal fragments (residues 115--145, 111--145 and 101--145) failed to cause steroidogenesis. These studies suggest the presence in the beta-subunit of human chorionic gonadotropin of determinants recognized by the tissue receptors, a part of these determinants residing between amino acid residues 57--71.     

Partial restoration of lutropin activity by an intersubunit disulfide bond: implications for structure/function studies

Gonadal function is controlled by lutropins and follitropins, heterodimeric cystine knot proteins that have nearly identical alpha-subunits. These heterodimeric proteins are stabilized by a portion of the hormone-specific beta-subunit termed the "seatbelt" that is wrapped around alpha-subunit loop 2 (alpha 2). Here we show that replacing human chorionic gonadotropin (hCG) alpha 2 residue Lys51 with cysteine or alanine nearly abolished its lutropin activity, an observation that implies that alpha Lys51 has a key role in hormone activity. The activity of the heterodimer containing alpha K51C, but not that containing alpha K51A, was increased substantially when beta-subunit seatbelt residue beta Asp99 was converted to cysteine. As had been reported by others, heterodimers containing alpha K51C and beta D99C were crosslinked by a disulfide. The finding that an intersubunit disulfide restored some of the activity lost by replacing alpha Lys51 suggests that this residue is not crucial for receptor binding or signaling and also that hCG and related hormones may be particularly sensitive to mutations that alter interactions between their subunits. We propose the unique structures of hCG and related family members may permit some subunit movement in the heterodimer, making it difficult to deduce key residues involved in receptor contacts simply by correlating the activities of hormone analogs with their amino acid sequences.     

Identification of epitopes associated with hCG and the beta hCG carboxyl terminus by monoclonal antibodies produced against a synthetic peptide

We have produced a library of monoclonal antibodies directed against a 37-amino acid synthetic polypeptide analogous to the carboxyl terminus of hCG. Five antibodies, designated FB01, FB02, FB03, FB04, and FB00, were developed and analyzed with respect to affinity and specificity for epitopes on human chorionic gonadotropin (hCG) and beta hCG by enzyme-linked immunoabsorbent and radioimmunoassays (RIA). All monoclonal antibodies demonstrated low affinity constants (approximately 10(-7) liters/mol) compared with those obtained by immunization with native beta hCG. One antibody, namely FB00, bound only to the synthetic peptide, whereas all other monoclonal antibodies recognized either free native beta hCG or both beta hCG and HCG. Antibodies produced against the synthetic peptide did not cross-react with other glycoprotein hormones such as LH, TSH, and FSH. Characterization of the monoclonal antibody-binding sites revealed the presence of at least four separate and distinct epitopes on the last 35 amino acids of beta hCG. Indeed, one epitope recognized by FB01 is located between residues 109 and 118, whereas another antigenic region recognized by FB04 appears to be present on the 109-121 portion of the molecule near or at position 118. One additional antigenic site was localized between residues 118 and 136. Finally, FB00 recognized an epitope located on the last 10 amino acids (136-145) of beta hCG. With the use of such antibodies, two- and three-site monoclonal RIA were developed and employed to detect free beta hCG and hCG in sera of patients with choriocarcinoma. These assays may be useful in the detection of beta hCG- and hCG-producing tumors and subsequent monitoring of patients in response to surgery and/or chemotherapy.     

New discoveries on the biology and detection of human chorionic gonadotropin

Human chorionic gonadotropin (hCG) is a glycoprotein hormone comprising 2 subunits, alpha and beta joined non covalently. While similar in structure to luteinizing hormone (LH), hCG exists in multiple hormonal and non-endocrine agents, rather than as a single molecule like LH and the other glycoprotein hormones. These are regular hCG, hyperglycosylated hCG and the free beta-subunit of hyperglycosylated hCG.     

The biological and clinical significance of nicks in human chorionic gonadotropin and its free beta-subunit.

Human chorionic gonadotropin (hCG) is a glycoprotein hormone composed of two dissimilar subunits, alpha and beta. Nicks or missing peptide linkages have been found in the beta 44-52 region of the beta-subunit of hCG, whether from pregnancy or trophoblast disease. This article reviews recent reports about the location of nicks in hCG, their origin and occurrence, their effects on the steroidogenic and receptor-binding activities of hCG, and on the immunological activities of hCG and its free beta-subunit. Taken together, the reports show: (1) nicks occur primarily between beta 47 and beta 48, and to a lesser extent between beta 44 and beta 45; (2) the extent of nicking in hCG samples varies widely, from undetectable to 100 percent of molecules; (3) nicks greatly reduce the steroidogenic activity of hCG in vitro (nicked molecules have less than 20 percent of the activity of the intact hormone); (4) nicks may occur at the trophoblast-myometrial interface or in the circulation by the action of human leucocyte elastase or similar leucocytic protease; (5) hCG testing kits using dimer-specific antibodies may not detect nicked molecules and may give different results from those using other antibodies; (6) hCG international reference preparations and the CR series of hCG standards are variably nicked (10 percent to 20 percent), complicating the problem of discordant hCG results in nick-sensitive assays; (7) results from commonly used immunoassays for measurement of the hCG free beta-subunit vary by as much as tenfold because some of the antibodies employed do not detect nick free beta-subunit.     

Non-cross-reactive monoclonal antibodies to human chorionic gonadotropin generated after immunization with a synthetic peptide

Noncross-reactive monoclonal antibodies specific for human chorionic gonadotropin (hCG) were obtained after pre-selection for submolecular specificity with a synthetic peptide immunogen. Mice were immunized with a synthetic peptide representing a segment unique to the beta-subunit of hCG (amino acid residues 109-145), conjugated to diphtheria toxoid. We then derived nine different hybridomas that secreted monoclonal antibodies reactive with both native hCG and isolated C-terminal peptide, after somatic cell hybridization of immune spleen cells with a nonsecretory myeloma cell line. None of the nine monoclonal antibodies, termed beta-hCG-CTPa1----a9, reacted with hLH, hFSH, or hTSH, although these pituitary hormones display extensive amino acid sequence homology with hCG. The noncross-reactive anti-beta-hCG monoclonal antibodies show apparent association constants on the order of 10(9) to 10(10) M-1. A sandwich-type enzyme-linked immunosorbent assay was set up with cut-off values of around 5 mIU/ml. These antibodies might have important implications for: a) improving the diagnosis and clinical management of pregnancy; b) monitoring the course of development of carcinomas which secrete the hormone, through in vitro assays or in vivo radioimmunodetection; c) evaluating the antibodies' therapeutic potential against such carcinomas; d) studying the biologic functions of the C-terminal segment of beta-hCG; and e) addressing the anti-fertility effect of antibodies raised against that segment.     

Location of major antigenic sites of the beta subunit of human chorionic gonadotropin.

Purified beta subunit of human chorionic gonadotropin (hCG) was partially reduced with beta-mercaptoethanol, carboxymethylated, and digested with chymotrypsin. The peptides were isolated by high-voltage electrophoresis and paper chromatography. Five major disulfide-containing peptides were isolated, and their location in the parent molecule was established by amino acid composition and amino- and carboxy-terminal analyses. All of these peptides inhibited the binding of 125I-labeled hCG by anti-beta hCG serum. The inhibitory effect of these peptides was lost when their disulfide bonds were reduced and alkylated. Synthetic carboxy-terminal peptides were not inhibitory. Based on these data it is concluded that a major antigenic site of hCG resides in the region of residues 21-23 with a disulfide bond connecting cysteine-23 or -26 with the cysteines at positions 72 or 110.     

Functional contributions of noncysteine residues within the cystine knots of human chorionic gonadotropin subunits

Human chorionic gonadotropin (hCG) is a heterodimeric member of a family of cystine knot-containing proteins that contain the consensus sequences Cys-X(1)-Gly-X(2)-Cys and Cys-X(3)-Cys. Previously, we characterized the contributions that cystine residues of the hCG subunit cystine knots make in folding, assembly, and bioactivity. Here, we determined the contributions that noncysteine residues make in hCG folding, secretion, and assembly. When the X(1), X(2), and X(3) residues of hCG-alpha and -beta were substituted by swapping their respective cystine knot motifs, the resulting chimeras appeared to fold correctly and were efficiently secreted. However, assembly of the chimeras with their wild type partner was almost completely abrogated. No single amino acid substitution completely accounted for the assembly inhibition, although the X(2) residue made the greatest individual contribution. Analysis by tryptic mapping, high performance liquid chromatography, and SDS-polyacrylamide gel electrophoresis revealed that substitution of the central Gly in the Cys-X(1)-Gly-X(2)-Cys sequence of either the alpha- or beta-subunit cystine knot resulted in non-native disulfide bond formation and subunit misfolding. This occurred even when the most conservative change possible (Gly --> Ala) was made. From these studies we conclude that all three "X" residues within the hCG cystine knots are collectively, but not individually, required for the formation of assembly-competent hCG subunits and that the invariant Gly residue is required for efficient cystine knot formation and subunit folding.     

The structures of the serine-linked sugar chains on human chorionic gonadotropin

The human chorionic gonadotropin beta-subunit tryptic COOH-terminal peptide (residues 123-145) which contains 3 serine-linked sugar chains was isolated. The sugar chains were cleaved by beta-elimination and then separated by gel filtration. The peaks were pooled and their compositions determined. The products of serial glycosidase digestion and periodate oxidation of the intact glycopeptide were also characterized. Of the serine-linked sugar chains, 13% were the hexasaccharide NeuAc alpha 2,3 Gal beta 1,3 (NeuAc alpha 2,3 Gal beta 1,4 GlcNAc beta 1,6) GalNAc, 34% the tetrasaccharide NeuAc alpha 2,3 Gal beta 1,3 (NeuAc alpha 2,6) GalNAc, 43% the trisaccharide NeuAc alpha 2,3 Gal beta 1,3 GalNAc and 10% the disaccharide NeuAc alpha 2,6 GalNAc.     

Intracellular immature subunits of human chorionic gonadotropin in first trimester placental cells: purification and characterization

As we previously reported [Sakakibara et al. (1986) Biochem. Biophys. Res. Commun. 137, 443-452; and Tominaga et al. (1989) J. Biochem. 105, 992-997], subunits of human chorionic gonadotropin (hCG) containing immature N-linked sugar chains (immature subunits), i.e., the 21 kDa form of alpha-subunit and the 23 and 19 kDa forms of beta-subunit, are present predominantly in first trimester placental cells. The molecular mass of intracellular hCG consisting of these subunits, based on gel filtration, was approximately 200 kDa, suggesting homo- or hetero-oligomerization of intracellular hCG. In the present study, we purified the 21 kDa form of alpha-subunit as well as the 23 and 19 kDa forms of beta-subunit from fresh normal first trimester placental tissues by gel filtration and reverse-phase high-performance liquid chromatography. Purified subunits were hydrolyzed (with a decrease in their molecular weighs) by endoglycosidase H and alpha-mannosidase but not by sialidase or sialidase followed by O-glycanase, indicating that those forms have presumably only high-mannose-type N-linked sugar chains but not O-linked sugar chains of the type present in mature beta-subunit. Fifteen cycles of Edman degradation of the purified forms of the subunits were performed. Only one phenylthiohydantoin amino acid, which was the same amino acid as in the urinary beta-subunit, was detected at each step for the mixture of 23 and 19 kDa forms of beta-subunit, indicating that the protein backbones of both forms are identical to each other as well as to the urinary beta-subunit.(ABSTRACT TRUNCATED AT 250 WORDS)     

The effect of administering equine chorionic gonadotropin (eCG) and human chorionic gonadotropin (hCG) post artificial insemination on fertility of lactating dairy cows

The objective was to evaluate the effect of equine chorionic gonadotropin (eCG) and hCG post artificial insemination (AI) on fertility of lactating dairy cows. In Experiment 1, cows were either treated with eCG on Day 22 post AI (400 IU; n = 80) or left untreated (n = 84). On Day 29, pregnant cows were either treated with hCG (2500 IU; n = 32) or left untreated (n = 36). Pregnancy and progesterone were evaluated on Days 29 and 45. In Experiment 2, cows (n = 28) were either treated with eCG on Day 22 (n = 13) or left untreated (n = 15) and either treated with hCG on Day 29 (n = 14) or left untreated (n = 14). Blood sampling and ultrasonography were conducted between Days 22 and 45. In Experiment 3, cows were either treated with eCG on Day 22 post AI (n = 229) or left untreated (n = 241). Pregnancy was evaluated on Days 36 and 85. In Experiment 1, eCG on Day 22 increased (P < 0.02) the number of pregnant cows on Day 29 (50.0 vs. 33.3%) and on Day 45, the increase was higher (P < 0.01) in cows with timed AI (41.2 vs. 6.5%) than in cows AI at detected estrus (50.0 vs. 37.8%). Pregnancy losses were reduced by eCG and hCG, but increased in cows that did not receive eCG but were given hCG (P < 0.01). Treatment with hCG tended (P < 0.06) to increase progesterone in control cows, but not in cows treated with eCG. In Experiment 2, hCG increased (P < 0.01) the number of accessory CLs on Day 35 (28.5 vs. 0.0%) and tended (P < 0.07) to increase progesterone. In Experiment 3, eCG increased the number of pregnant cows (P < 0.05) on Days 36 and 85, but only in cows with low body condition (eCG = 45.6 and 43.5%; Control = 22.9 and 22.9%). In conclusion, eCG at 22 days post insemination increased fertility, primarily in cows with low body condition and reduced pregnancy losses when given 7 days before hCG; hCG induced accessory CLs and slightly increased progesterone, but hCG given in the absence of a prior eCG treatment reduced fertility.     

GnRH agonist Lupron (leuprolide acetate) pre-treatments prevent ovulation in response to gonadotropin stimulation in the clouded leopard (Neofelis nebulosa)

In many species, controlling the ovary prior to induction of ovulation improves the success of ovarian response and artificial insemination (AI). We assessed the impact of suppression of estrus with the GnRH agonist, Lupron, on ovarian sensitivity to equine chorionic gonadotropin (eCG) and human chorionic gonadotropin (hCG) in the clouded leopard. Seven female clouded leopards were given two injections of Lupron (3.75 mg IM) 23 d apart, followed 44 d later by eCG and hCG. Daily fecal samples were collected from 60 d before Lupron to 60 d after hCG. Fecal metabolites of estrogen (E) and progesterone (P) were measured by radioimmunoassay. Lupron decreased (P < 0.05) the number of E peaks during Lupron treatment compared to pre-Lupron. All females had baseline E and six of seven (86%) had nadir P on day of eCG. Exogenous gonadotropins induced E elevations in all females. However, mean E in the gonadotropin-provoked estrus was decreased (P < 0.05) compared to pre-Lupron estrous periods. Only one of seven (14%) females ovulated after eCG/hCG. In conclusion, estrous cycle control with Lupron resulted in predictable ovarian suppression prior to gonadotropin stimulation but altered ovarian sensitivity by an as yet unknown mechanism so that ovulation was inhibited, even when using a proven exogenous gonadotropin protocol.     

Epitope analysis and molecular modeling reveal the topography of the C-terminal peptide of the beta-subunit of human chorionic gonadotropin.

Human chorionic gonadotropin (hCG) belongs to a family of heterodimeric glycoprotein hormones that share a common alpha-subunit and a hormone-specific beta-subunit. Among the gonadotropin beta-subunits, greater than 85% homology exists between lutropin (hLH)beta and hCGbeta in their first 114 amino acid residues. However, unlike hLHbeta, hCGbeta contains a 31-amino acid hydrophilic stretch at its carboxyl end (CTPbeta: C-terminal peptide). Although the crystal structure of deglycosylated hCG has been solved, the topography of CTPbeta remains unknown. In this study, we have attempted to define the topology of CTPbeta using mAb probes. We investigated three epitopes on hCGalpha, which are hidden in the hCGalphabeta dimer. However, these epitopes are not hidden in hLH, which has a similar subunit interface to that of hCG, but lacks CTPbeta. This suggested that these epitopes are not masked at the subunit interface of hLH or hCG. Hence, we hypothesized that, in the case of hCG, these epitopes are masked by the CTPbeta. Consistent with this view, several treatments of hCG that removed CTPbeta unmasked these epitopes and enhanced their reactivity with the corresponding mAbs. In order to localise the position of CTPbeta on the alpha-subunit, we used an epitope-mapping strategy [N. Venkatesh & G. S. Murthy (1997) J. Immunol. Methods 202, 173-182] based on differential susceptibility of epitopes to covalent modifications. This enabled us to predict the possible topography of CTPbeta. Further, we were also able to build a model of CTPbeta, completely independently of the epitope-mapping studies, using a homology-based modeling approach [S. Krishnaswamy, I. Lakshminarayanan & S. Bhattacharya (1995) Protein Sci. 4 (Suppl. 2), 86-97]. Results obtained from these two different approaches (epitope analysis and homology modeling) agree with each other and indicate that portions of CTPbeta are in contact with hCGalpha in the native hCG dimer.     

Modification of sialyl residues of gonadotropic hormones by reductive amination. Influence on biological activity and circulating half-life

The sialic acid residues of human chorionic gonadotropin, human lutropin and human follitropin were quantitatively modified by introduction of an amino compound. In radioreceptor assays, the modified chorionic gonadotropin, lutropin and follitropin saturated the receptors. However, in the low nanogram range, the gonadotropic binding was higher for the control compared to the modified sample.The hormonal activity of the chorionic gonadotropin was testedin vitro. The modified preparations were four- to thirteen-fold less stimulatory compared to the control but elicited the same maximal response. The biological activity of follitropin was determinedin vivo. In this case, the modified preparations were four- to five-fold less stimulatory than the control. Both the modified chorionic gonadotropin and follitropin preparations were found to act as agonists. Modification of the gonadotropin hormones did not significantly alter the immune recognition of these glycoproteins.The apparent circulating half-life in rats of the modified chorionic gonadotropin and follitropin was increased six- to nine-fold compared to that of native hormones; this might be a consequence of resistance of the modified sialyl residues to sialidases and the resultant slower exposure of terminal galactosyl residues; the plasma half-life of modified lutropin remained the same as that of the native hormone.Abbreviations hCG human chorionic gonadotropin - hLH human lutropin or luteinizing hormone - hFSF human follitropin or follicle stimulating hormone - mala methyl ester of alanine - hCG(ala, mala, etc.) human chorionic gonadotropin modified on sialicacid by reductive amination with alanine, methyl ester of alanine, etc. - IRP-HMG intact rat prostrate-human menopausal gonadotropin     

Luteinizing hormone of the sperm whale. Amino acid sequences of reduced and carboxymethylated beta-subunits

The amino acid sequence of reduced and carboxymethylated beta-subunit of the luteinizing hormone of the sperm-whale was established. The beta-subunit was found to consist of 118 amino acid residues, the N- and C-terminal amino acids being proline and leucine, respectively. The carbohydrate moiety is linked to the asparagine residue at position 13.