Heparin-binding motif mutations of human diamine oxidase allow the development of a first-in-class histamine-degrading biopharmaceutical

Background: Excessive plasma histamine concentrations cause symptoms in mast cell activation syndrome, mastocytosis, or anaphylaxis. Anti-histamines are often insufficiently efficacious. Human diamine oxidase (hDAO) can rapidly degrade histamine and therefore represents a promising new treatment strategy for conditions with pathological histamine concentrations. Methods: Positively charged amino acids of the heparin-binding motif of hDAO were replaced with polar serine or threonine residues. Binding to heparin and heparan sulfate, cellular internalization and clearance in rodents were examined. Results: Recombinant hDAO is rapidly cleared from the circulation in rats and mice. After mutation of the heparin-binding motif, binding to heparin and heparan sulfate was strongly reduced. The double mutant rhDAO-R568S/R571T showed minimal cellular uptake. The short α-distribution half-life of the wildtype protein was eliminated, and the clearance was significantly reduced in rodents. Conclusions: The successful decrease in plasma clearance of rhDAO by mutations of the heparin-binding motif with unchanged histamine-degrading activity represents the first step towards the development of rhDAO as a first-in-class biopharmaceutical to effectively treat diseases characterized by excessive histamine concentrations in plasma and tissues. Funding: Austrian Science Fund (FWF) Hertha Firnberg program grant T1135 (EG); Sigrid Juselius Foundation, Medicinska Understödsförening Liv och Hälsa rft (TAS and SeV).


Introduction
The biogenic amine histamine (2-[4-imidazolyl]ethylamine) is stored by basophils and mast cells (MCs) and is rapidly released after stimulation from intracellular storage vesicles. Following activation of the histamine receptors 1 and 2, histamine acts mainly on vascular endothelial, bronchial, and smooth Plasma of seven mastocytosis patients and three healthy volunteers was spiked with 6 nM rhDAO-WT, 545 nM histamine, and 20 µM of the potent DAO-inhibitor diminazene aceturate (DIMAZ; Sigma-Aldrich, D7770) and histamine concentrations were determined in duplicate with a histamine ELISA (Immunotech IM2562, Beckman Coulter, Brea, CA) as described previously (Boehm et al., 2019).

Site-directed mutagenesis to create HBM mutants
To generate rhDAO single and double HBM mutants, the charged amino acids arginine and lysine were replaced with polar threonine and serine by site-directed mutagenesis. Threonine and serine have been used to replace arginine residues in the HBM of fibronectin, resulting in significantly reduced binding affinities, while at the same time retaining the three-dimensional structure (Busby et al., 1995;Kapila et al., 2001). The triple mutant rhDAO-K570G/R571Q/K572T representing the HBM of guinea pig, dog, rat, mouse, and Chinese hamster DAO was also tested. All mutants were generated from the rhDAO expression plasmid described by Gludovacz et al., 2016. The HBM mutants and the respective 5′-phosphorylated primers are summarized in the Key resources table. The cloning procedure is described by Gludovacz et al., 2018. Human IgG Fc was fused to the N-termini of rhDAO-WT (rhFcDAO) and rhDAO-R568S/R571T (rhFcDAO-R568S/R571T) using an IgG1 hinge region as a linker. The Fc sequence was synthesized by Eurofins MWG Operon (Ebersberg, Germany), and plasmids were constructed by using the methods described by Gludovacz et al., 2016. Expression and purification of the HBM mutants Stable CHO-K1 cell lines expressing the different HBM mutants were generated and cultivated as described for the wildtype (Gludovacz et al., 2016). rhDAO-R571T and -R571T/K575T could not be successfully expressed. For batch cultivation, cells were seeded at a viable cell density of 0.2 × 10 6 cells/mL and incubated for 8 days. 10 µM CuSO 4 was added on days 0 and 4. One batch of rhDAO-WT (batch 2) and rhDAO-R568S/R571T (batch 3) was produced with the ExpiCHO expression system (Thermo Fisher Scientific, Waltham, MA) using the standard protocol over 10 days (see 'ExpiCHO Expression System User Guide,' MAN0014337). 10 µM CuSO 4 was added on days 0, 2, 4, 6, and 8.
Culture supernatants were ultra-and diafiltrated using the Labscale TFF System in combination with one to three Pellicon XL 50 Ultrafiltration Cassette-Biomax Polyethersulfone with a 100 kDa molecular mass cutoff (both Merck Millipore, Burlington, MA). The supernatants were concentrated 10-to 50-fold, and the culture media were replaced by 10 mM potassium phosphate buffer, pH 7.4 (Merck Millipore). The samples were loaded onto three 5 mL HiTrap Heparin HP columns connected in series at a flow rate of 2 mL/min using an Äkta Purifier or Start HPLC device (all GE Healthcare, Chicago, IL). Stepwise elution of rhDAO-WT, rhFcDAO, and rhDAO-K575T was performed with 0.25, 0.5, and 1 M KCl. The other HBM mutants were eluted with 0.125, 0.25, and 1 M KCl (Sigma-Aldrich) in 10 mM potassium phosphate buffer, pH 7.4. The eluates containing rhDAO variants were concentrated and desalted using the Labscale TFF System combined with one Pellicon XL 50 Ultrafiltration Cassette and 25 mM Tris-HCl buffer, pH 8.5 (AppliChem, Darmstadt, Germany). This material was loaded onto a 5 mL HiTrap CaptoQ column (GE Healthcare) at a flow rate of 3 mL/min. Stepwise elution was conducted with 0.315 and 1.5 M KCl in 25 mM Tris-HCl buffer, pH 8.5. Buffer exchange of the 0.315 M KCl eluate against 50 mM HEPES with 150 mM KCl, pH 7.5 was performed with the Labscale TFF System and one Pellicon XL 50 Ultrafiltration Cassette. Purified rhDAO was quantified using NanoDrop 1000 spectrophotometer (Thermo Fisher Scientific).

Determination of heparin binding using heparin-sepharose
The storage buffer of purified rhDAO-WT and the HBM mutants was replaced with either 10 mM potassium phosphate, pH 7.4, or 50 mM HEPES, pH 7.4, using Amicon Ultra centrifugal filter units (MWCO 100 kDa, Merck Millipore). They were loaded onto a HiTrap Heparin HP column (1 mL, GE Healthcare) at a flow rate of 0.2 mL/min using an Äkta purifier HPLC device. Elution was performed using a linear gradient of increasing amounts of 1 M KCl in 10 mM potassium phosphate, pH 7.4, or 1 M NaCl in 50 mM HEPES, pH 7.4.

Analysis of rhDAO-R568S/R571T binding to HMWH and HS
Binding of rhDAO-R568S/R571T to HMWH and HS compared to rhDAO-WT was analyzed with ITC using automated MicroCal PEAQ-ITC (Malvern Panalytical, Malvern, UK) and with BLI using Octet RED96e (FortéBio, Fremont, CA) as described previously . Sample concentrations and injection modes of all ITC experiments are listed in Figure 2-source data 1. For BLI comparison of binding of the wildtype protein and the R568S/R571T mutant to HS and HMWH, protein concentrations of 471 nM (HS) and 88 nM (HMWH) were used.
Suppliers, catalog numbers, and culture media compositions were recently published . The identity of all cell lines has been authenticated. Absence of mycoplasma has been confirmed on a regular basis.

Flow cytometry
The cells were seeded into 96-well plates. At confluency, the supernatant was replaced with 50 µL fresh culture medium supplemented with 30 nM Alexa488-labeled rhDAO-WT or rhDAO-R568S/ R571T and incubated for 60 min at 37 °C. These variants were fluorescently labeled by periodation of vicinal diol groups and subsequent conjugation with an Alexa Fluor 488-hydrazide . No DAO added served as a negative control. The cells were washed three times for 5 min with PBS containing 500 µg/mL heparin (sodium salt from intestinal mucosa, Sigma-Aldrich) at room temperature and detached with 50 µL 0.1% trypsin, 0.02% EDTA (Sigma-Aldrich). The reaction was stopped with 50 µL 1× trypsin inhibitor (Sigma-Aldrich, T6414) and addition of 100 µL PBS. 500 cells per cell sample were analyzed on a CytoFLEX S flow cytometer.
Raw data of fluorescence measurements were analyzed with the help of R version 3.6.3 (R Development Core Team, 2020) and RStudio 1.1.463 including the following package: userfriendlyscience version 0.7.2. Data of each tested cell line was loaded into R, log-transformed, and normalized between biological replicates by dividing the data by the median of the respective negative controls. The standard deviations of each cell line were adjusted between biological replicates by dividing the standard deviation of the negative controls of the first biological replicate by the standard deviation of the negative controls of the second biological replicate. This factor was then used to adjust the standard deviation of all samples. The medians of all samples were used for statistical testing. First, a Shapiro-Wilk test was used to test whether the medians were normally distributed. As this was the case for all samples and cell lines, a Bartlett test was used to see whether all samples of a cell line have equal variances. If this was the case, a one-way ANOVA with a Tukey's HSD test as a posthoc test was used to identify significant differences. If equality of variances could not be assumed, an ANOVA with Welch's correction and a Games-Howell posthoc test were applied. If a p-value of <0.05 was calculated, significant differences were assumed.
Background corrected means of the median values ± SEM: median fluorescence intensity values of all samples were calculated with Kaluza Flow Cytometry Analysis Software (Beckman Coulter, version 2.1). After correcting for background fluorescence, the means of the medians ± SEM are presented.

Fluorescence microscopy
The cells were seeded into 8-well µ-slides (ibidi, Martinsried, Germany, 80826) 3-4 days prior to the experiments. They were washed once with PBS and then covered with 130 µL of the respective culture medium containing 60-120 nM unlabeled or Alexa488-labeled rhDAO. Incubation was performed at 37 °C for 60 min. The cells were washed three times for 5 min with PBS containing 500 µg/mL heparin at room temperature. The cell fixation, permeabilization staining, and imaging procedures were recently published .
Determination of in vivo clearance and DAO antigen and activity rhDAO-WT and the different mutants were administered to rats and mice as described . DAO antigen concentrations and enzymatic activity were measured in duplicate as published Boehm et al., 2017). The activity curve in the linear range was used to calculate the slope corresponding to the oxidation rate.

Ethics statement
The experimental protocols for the treatment of rats and mice were approved by the local Animal Welfare Committee and the Federal Ministry of Science, Research and Economy (GZ 66.009/0152--WF/V/3b/2014) and conducted in full accordance with the ARRIVE guidelines (Kilkenny et al., 2010).

Prediction of mutation effects on protein stability and heparin-binding affinity in silico
The complex of wildtype hDAO and heparin hexasaccharide was constructed as described previously . Briefly, the missing side-chain atoms in the DAO dimer were first built with Pymol (The PyMOL Molecular Graphics System, version 2.0, Schrödinger, LLC, 2021). Using ClusPro, heparin tetrasaccharide probes were blindly docked into DAO dimer to reveal the heparin-binding site, which was later used for a restrained docking. Heparin hexasaccharide was then manually built by joining two probes in Pymol and subjected to restrained energy minimization using AmberTools GUI incorporated in Chimera (UCSF).
Three single (R568S, R571T, and K575T), three double (R568S/R571T, R568S/K575T, and R571T/ K575T), and one triple (K570G/R571Q/K572T) mutants were built with Modeller 9.2 GUI in BIOVIA Discovery Studio 2019 (Dassault Systemes). Twenty models for each mutant were built without global optimization to avoid driving the conformation too far from the initial one. The model with the lowest probability density function (PDF) total energy and discrete optimized protein energy (DOPE) score was selected as the representative structure. The 'Side-Chain Refinement' protocol was then used to optimize the mutated side chains without affecting the overall fold. For each mutant, the electrostatic surface was calculated with the APBS plugin (Baker et al., 2001) and visualized in Pymol.
To predict the effects of the mutations on DAO stability, the FoldX 4.0 (Centre for Genomic Regulation, Spain; Schymkowitz et al., 2005) command-line interface and 'PositionScan' command were used. Free energy contributions from individual mutations were summed to get total free energy changes.
Changes in free energy of binding to heparin hexasaccharide were predicted with the 'Calculate mutation energy (Binding)' protocol in Discovery Studio 2019 (Spassov and Yan, 2013) using the following parameters: pH-dependent electrostatics = true, pH = 7.4, preliminary minimization = false, forcefield = CHARMm Polar H. All other options were used with default values.

Results rhDAO rapidly degrades endogenous and exogenous histamine
The lowest published K m of rhDAO for histamine is 2.8 µM or 311 ng/mL (Elmore et al., 2002). Histamine concentrations inducing clinically relevant hypotension start at approximately 5 ng/mL, which is 62-fold below the K m . Is rhDAO able to rapidly degrade histamine using a pathophysiologically relevant histamine concentration range of 5-100 ng/mL? 6 nM (1 µg/mL) rhDAO-WT with and without a  Figure  2-source data 1. (C) Biolayer interferometry. Streptavidin sensors loaded with biotinylated HMWH or HS were incubated for 10 min with 88 nM or 471 nM rhDAO, respectively. Dissociation was measured for 15 min. The graphs represent one of three individual measurements and show the association and dissociation curves after subtraction of the negative control (no DAO added). The data of rhDAO-WT have already been published  but are added for better presentation.
The online version of this article includes the following source data for figure 2: Source data 1. All raw plots and integrated heat plots of the isothermal titration calorimetry (ITC) analyses of the heparin-binding motif (HBM) mutant are presented in Figure 2B. 100-fold molar excess of HMWH completely degraded 100 ng/mL histamine in less than 15 min using 1% HSA-PBS or EDTA plasma ( Figure 1A). Heparin did not influence DAO activity. Plasma of seven mastocytosis patients and three healthy volunteers with endogenous histamine levels of approximately 5 nM were also spiked with 545 nM exogenous histamine. rhDAO-WT degraded exogenous and endogenous histamine to harmless concentrations below 1 ng/mL. Histamine deamination was completely blocked by addition of the potent DAO inhibitor DIMAZ ( Figure 1B).

The proposed HBM is essential for binding to heparin and heparan sulfate
We have recently shown in rodents that rhDAO-WT is rapidly cleared from the circulation, precluding its use as a histamine-degrading biopharmaceutical . Mutations in the HBM of superoxide dismutase increased the plasma concentration 10-fold (Sandström et al., 1994). Based on the superoxide dismutase data and the complete blockade of rhDAO uptake into various cell lines using HMWH , we hypothesized that the proposed HBM of DAO is involved in the rapid clearance of rhDAO in vivo. The tested HBM mutants eluted from the heparin-sepharose at lower salt concentrations than the wildtype protein ( Figure 2 and Table 1). rhDAO-R568S/R571T demonstrated the weakest binding to heparin-sepharose and was therefore further analyzed using isothermal titration calorimetry (ITC) and biolayer interferometry (BLI). rhDAO-WT bound to HMWH with mean (SD) K D values of 634 (26) nM in ITC and 1.6 and 69 nM in BLI analyses with likely two DAO molecules binding to one HMWH molecule . The mean K D values for HS were 4 and 112 nM using BLI. The best curve fit was again generated with a ratio of two molecules DAO binding to one molecule HS . rhDAO-R568S/R571T, the rhDAO mutant with the lowest heparin affinity, did not show any binding to HMWH and only minimal binding to HS ( Figure 2B and C). All raw plots and integrated heat plots of the ITC analyses of the HBM mutant are presented in Figure 2-source data 1.

Mutations of the HBM significantly reduce cellular internalization of rhDAO
rhDAO-WT is rapidly internalized into various cell types . To test whether mutations of the HBM not only reduce binding of rhDAO to HMWH and HS, but also inhibit cellular uptake, the single mutant rhDAO-R568S and the two double mutants, rhDAO-R568S/R571T and rhDAO-R568S/K575T, were selected for cellular internalization experiments. SK-Hep1 cells showed a slightly reduced uptake of the single mutant rhDAO-R568S, but a strong decrease for both double mutants ( Figure 3A). The other cell lines were only incubated with rhDAO-WT and the double mutant rhDAO-R568S/R571T. The wildtype protein showed intracellular granular staining. The fluorescence signal using the mutant was comparable to the negative controls with no DAO added ( Figure 3A). In flow cytometric analyses, the reduction in fluorescence of the double mutant rhDAO-R568S/ R571T ranged from 67% in SK-Hep1 to 85% in PODO/TERT256 cells ( Figure 3B). The decrease in fluorescence intensity was statistically significant for all cell lines (p-value<0.05; Figure 3-source data 1). Source data 1. Statistical evaluation of flow cytometry data is summarized in Figure 3B.
Source data 2. Raw unedited western blots are presented in Figure 3C.

Clearance of HBM mutants is strongly decreased compared to wildtype DAO in rats and mice
After demonstrating reduced in vitro cellular uptake using the HBM mutants, pharmacokinetic parameters were determined in rats and mice. The data are summarized in Table 2. The double mutant rhDAO-R568S/ R571T generated the highest area under the curve (AUC), followed by rhDAO-R568S and rhDAO-R568S/ K575T ( Figure 4A). We also tested an rhFc-DAO wildtype fusion protein and the corresponding rhFc-DAO-R568S/R571T mutant confirming the improved pharmacokinetic parameters ( Figure 4B).
The most promising variant rhDAO-R568S/R571T was also tested in mice, and the data support the rat data with more than 15-fold increases in the AUC after intravenous or intraperitoneal administration ( Figure 4C and D and Table 2).
The rhDAO-R568S/R571T mutant with and without a 100-fold molar excess of HMWH degraded histamine as efficiently as wildtype DAO using 1 µg/mL or 6 nM enzyme concentration ( Figure 5A compared to WT in Figure 1A). No significant differences in enzymatic activity could be detected comparing rhDAO-WT and -R568S/R571T mutant at concentrations of 0.7, 2, and 6 nM using a different assay format ( Figure 5-figure supplement 2). After intravenous administration of rhDAO-R568S/R571T, rhDAO-R568S/K575T, and the Fc fusion protein Fc-R568S/R571T, the corresponding DAO activity could be readily measured for 28 hr ( Figure 5). Diamine oxidase concentrations and activities are highly correlated with R 2 values of >96% ( Figure 5-figure supplement 1). In silico analysis of heparin binding indicates a key role of the conserved R568 residue Comparison of the HBM in 87 DAO Mammalia sequences showed an overall sequence identity of 68% at positions R568 to K575, with 89, 49, and 67% conservation at positions R568, R571, and K575, respectively. In the alignment of 15 HBM sequences from old world monkeys, great apes, and humans, residues from R568 to P574 are 100% conserved. At the single variable position 575, 3 of the 15 sequences have a lysine (20%), but it is conservatively substituted by an arginine in 11 sequences (73%). The evolutionary conservation analysis is summarized in Supplementary file 1. The symmetric HBM on top of the DAO dimer is composed of the residues 568 RFKRKLPK 575 from both chains ( Figure 6A and Figure 6-figure supplement 1A), with 10 out of 16 residues (63%) positively charged. The experimental results demonstrated that R568 is critical for heparin binding. The replacement of R568 by a serine ( Figure 6B) was predicted to strongly decrease the binding affinity, while the variants R571T ( Figure 6C) and K575T (Figure 6-figure supplement 1B and C) were predicted to be associated with lower binding energy changes (Table 3), supporting the experimental binding data using heparin-sepharose ( Table 1). The squared correlation coefficient from regression analysis using affinity predictions and measured salt elution concentrations was 93% (p-value 0.0088) including wildtype and four HBM mutants but excluding the triple mutant ( Figure 6-figure supplement 2).
In both rhDAO chains, R568 forms ionic interactions and hydrogen bonds with the sulfate groups of the heparin hexasaccharide ( Figure 6D and Figure 6-figure supplement 1D). These interactions are absent in the R568S mutant ( Figure 6E and Figure 6-figure supplement 1E). The R571 residue has a role in the overall architecture of the HBM since its guanidinium group forms intra-chain hydrogen bonds ( Figure 6-figure supplement 1D), which are lost in the R571T mutant ( Figure 6-figure supplement 1E). The negligible effect of K575T change on binding affinity can be attributed to its deep location and long distance from the heparin hexasaccharide. The destabilizing effect (Table 3) likely results from the loss of favorable intramolecular interactions formed by K575 (Figure 6-figure  supplement 1B and C). Although the triple mutant K570G/R571Q/K572T (Figure 6-figure supplement 1F and G) showed the lowest predicted affinity of all mutants, binding to heparin-sepharose was stronger compared to R568S/R571T (Table 1). This discrepancy might be explained by the additive nature of the energy change calculating algorithm, which is functionally optimized for single and double point mutations and likely overestimates the effects of triple mutations (Spassov and Yan, 2013). Due to the surface position of R568, K570, R571, and K572, the effects of their mutations on the structural stability, even in the case of the triple mutant, are predicted to be lower (Table 3, Figure 6-figure supplement 1F and G).

Discussion
Plasma histamine concentrations of at least 500 ng/mL (4.5 µM) were measured in a systemic mastocytosis patient following gastrointestinal endoscopy (Desborough et al., 1990). We recently published data on the circulatory collapse in a mastocytosis patients with 70 ng/ mL plasma histamine (Boehm et al., 2019). Histamine concentrations in another systemic mastocytosis patient increased from 10 ng/mL to 35 ng/mL over a few hours. The patient developed severe clinical symptoms despite high-dose treatment with the histamine receptor 1 antagonist diphenhydramine and the histamine receptor 2 antagonist ranitidine (Boehm et al., 2019). Despite treatment with histamine receptor antagonists up to four times the approved dose and the use of additional medications, approximately 70% of mastocytosis patients define themselves as disabled in accordance with standard definitions of disability (Hermine et al., 2008). Finally, 25-33% of chronic urticaria patients are resistant to histamine receptor antagonists (van den Elzen et al., 2017). Why is the efficacy of histamine receptor blockers during severe anaphylaxis, MC activation, or chronic urticaria limited? Kaliner et al., 1982 andOwen et al., 1982 described that in healthy volunteers two combinations of histamine receptor antagonists, cimetidine with hydroxyzine and cimetidine with chlorpheniramine, respectively, were only able to sufficiently block symptom development until systemic histamine concentrations of approximately 6 ng/mL were reached. Single treatments with either histamine receptor 1 or 2 antagonist were minimally protective. The 6 ng/mL threshold is only three to four times higher compared to no treatment and 20-fold below the mean histamine concentration of 140 ng/mL reached during severe anaphylaxis after an insect sting challenge (van der Linden et al., 1992). Clinically relevant hypotension starts to develop at levels above 5 ng/mL histamine. We can only conclude that histamine receptor antagonists are easily overwhelmed during anaphylaxis, MC activation, and even chronic urticaria.
Herein we show for the first time that rhDAO can rapidly and completely degrade pathophysiologically relevant histamine concentrations of 100 ng/mL. Nevertheless, in rodents wildtype rhDAO showed a very fast α-distribution half-life, rendering it unsuitable for further development as a new and first-in-class biopharmaceutical. The fast clearance in vivo was independent of the asialoglycoprotein-and mannose-receptors, the two well-characterized protein clearance systems. Cellular internalization in vitro into various cell lines was dependent on the interaction with HS glycosaminoglycans and was blocked by excess heparin .
We therefore mutated the putative HBM (Novotny et al., 1994) and unequivocally demonstrated that this 2 × 8 linear amino acid stretch, which forms a distinct positively charged ring structure, mediates not only cellular internalization but also rapid clearance in rodents. Mutations in the HBM eliminated heparin binding. Importantly, DAO activity of wildtype and various mutants was indistinguishable. The wildtype DAO dimer sequence excluding the HBM contains 84 arginine and 42 lysines residues, many of which are surface-exposed, but none of these positively charged residues seems critically involved in heparin binding. BLAST analysis of the RFKRKLPK motif revealed a high sequence identity of R568 and K575 in all mammalian DAO sequences analyzed. Our experimental data suggest the essential role of the conserved R568 residue in the heparin binding of DAO, which is supported by the in silico modeling comparing the interactions of wildtype and mutant DAO with the docked heparin hexasaccharide. While the single mutation of this amino acid was sufficient to achieve a comparable decrease in binding to heparin-sepharose versus the double mutants, a strong reduction of cellular uptake was only accomplished with the double mutants. Neither ITC nor BLI detected any interaction of rhDAO-R568S/R571T with HMWH and HS. In silico modeling supports not only the biochemical and cellular in vitro but also the rats and mice in vivo data. The HBM mutant with the best in vivo pharmacokinetic profile, rhDAO-R568S/R571T, showed low increases in the Gibbs free energy concerning stability but demonstrated a strong decrease in affinity. The correlation coefficient comparing in silico-estimated affinity data changes with experimental heparin-sepharose in vitro binding data of wildtype and four HBM mutants is 0.93 with a p-value of less than 0.01. Incubation of various cell lines with rhDAO-R568S/R571T did not completely abrogate cellular uptake, but flow cytometry and western blot analyses showed that at least 15% of the internalization cannot be attributed to HSPG interaction. This is in accordance with our previous study, where a 100fold excess of HMWH over rhDAO-WT could not reduce the uptake below 5 and 25% in various cell lines . In the same study, we observed the presence of high-affinity-binding sites in HUVEC/TERT2 cells in addition to the low-affinity HSPG-binding sites. Similarly, cellular binding of amyloid protein precursor and fibroblast growth factor 2 is predominantly mediated by low-affinity HSPG interactions that serve as scaffolds or co-receptors that promote and/or stabilize the formation of complexes of proteins and high-affinity receptors (Duchesne et al., 2012;Reinhard et al., 2013;Thompson et al., 1994;Xu and Esko, 2014). Since the HS-binding sites can outnumber the amount of protein-specific receptors by 100-to 1000-fold (Duchesne et al., 2012), it is not surprising that only a low proportion of binding can be attributed to the latter. These high-affinity-binding sites might be responsible for the β-elimination of the heparin-binding mutants. Our data are in agreement with this hypothesis. While the clearance of more than 90% of the rhDAO-WT dose in the fast α-elimination phase of less than 5 min is fully abrogated by the double mutation of R568 and R571, the long β-elimination phase of about 6 hr is more or less unchanged. It is therefore likely that this second phase is determined by a different clearance mechanism independent of the heparin-binding domain.
It will be interesting to study the distribution of wildtype versus HBM mutant DAO variants after intravenous infusion in animal models because it is not clear which cells or organs remove 90% of the wildtype DAO protein within a few minutes. The liver has been considered mainly responsible for elimination of DAO, but in view of the dependence of DAO internalization on HS, endothelial cells throughout the circulatory system could rapidly remove DAO (D'Agostino et al., 1986). These cells abundantly express HSPGs and are able to bind and internalize DAO in vitro (Fuster and Wang, 2010;Gludovacz et al., 2020).
The half-life of the HBM mutant rhDAO-R568S/R571T is approximately 6 hr in rodents, which is certainly sufficient for the treatment of most human anaphylaxis events. The rodent HBM does not contain the central 570 KRK 572 sequence present in humans but instead 570 GQT 572 and therefore the halflife in humans could be longer, assuming that the 6 hr half-life is still at least partially determined by the same HS internalization process. The rodent HBM is only weakly heparin binding.
In conclusion, mutations in the proposed and now proven HBM converted rhDAO wildtype protein into a candidate for a first-in-class histamine receptor-independent biopharmaceutical for the rapid and complete elimination of excessive histamine. First clinical indications might be diseases where it has been known for decades that histamine plays an important pathophysiological role. These include anaphylaxis, MC activation syndrome, mastocytosis, and chronic urticaria. Nevertheless, the use of HBM mutants will also allow clinical proof-of-concept studies in other diseases, where the involvement of histamine and MCs is suspected, but where clinical studies with available histamine receptor antagonists were unsuccessful. No drug blocking histamine receptor 4 has been approved to date. This group of diseases includes, amongst others, asthma, atopic dermatitis, infusion reactions, different forms of pruritus, and inflammatory bowel disease. rhDAO with HBM mutations might overcome the current limitations of histamine receptor antagonists for the treatment of diseases that lead to lifethreatening histamine exposure but are resistant to conventional treatment modalities. sequences. We are grateful to the bioinformatics (J.V. Lehtonen), translational activities and structural biology (FINStruct) infrastructure support from Biocenter Finland and CSC IT Center for Science for computational infrastructure support at the Structural Bioinformatics Laboratory (SBL), Åbo Akademi University. SBL is part of the NordForsk Nordic POP (Patient Oriented Products), the Solutions for Health strategic area of Åbo Akademi University, and the InFlames Flagship program of the Academy of Finland on inflammation, cancer and infection, University of Turku and Åbo Akademi University. We are indebted to Sarah Ely for the final polish in the proper usage of the English language. The publication fees were covered by the BOKU Vienna Open Access Publishing Fund.

Additional information
Competing interests Elisabeth Gludovacz, Bernd Jilma, Nicole Borth, Thomas Boehm: is named as an inventor with The Medical University of Vienna and the University of Natural Resources and Life Sciences of a patent describing the rhDAO heparin-binding motif mutants presented herein (patent pending WO2020169577A1). The other authors declare that no competing interests exist. The funders had no role in study design, data collection and interpretation, or the decision to submit the work for publication.