A human monoclonal antibody against the distal carboxyl terminus of ADAMTS13 modulates its susceptibility to an inhibitor in thrombotic thrombocytopenic purpura

Summary

Background:

Immune thrombotic thrombocytopenic purpura (iTTP) is a potentially fatal thrombotic microangiopathy, resulting from a severe deficiency of plasma ADAMTS13 activity. Immunoglobulin (Ig) G-type autoantibodies are primarily responsible for the inhibition of plasma ADAMTS13 activity. However, the mechanism underlying autoantibody-mediated inhibition is not fully understood.

Objective:

The purpose of the present study is to determine the role of IgG autoantibodies against various carboxyl-terminal domains of ADAMTS13 in regulating ADAMTS13 activity and its inhibition.

Method:

Various human monoclonal antibodies isolated by phage display, recombinant protein expression and purification, and biochemical analyses were employed for the study.

Results:

Our results demonstrate for the first time that a human monoclonal antibody fragment, the single chain fragment of the variable region (scFv) isolated from a patient with acute iTTP that binds the distal carboxyl-terminus of ADAMTS13, is able to activate ADAMTS13 and increase the proteolytic cleavage of a FRETS-VWF73 substrate; moreover, binding of such a human monoclonal antibody against the carboxyl-terminus of ADAMTS13 to plasma ADAMTS13 appears to modulate inhibition by another human monoclonal antibody (i.e., scFv4–20), also isolated from an iTTP patient, that targets the spacer domain of ADAMTS13. These results provide new insights into our understanding of the pathogenesis of iTTP.

Introduction

Immune thrombotic thrombocytopenic purpura (iTTP), a potentially fatal blood disorder, is caused by immunoglobin (Ig) G-type autoantibodies against various domains of ADAMTS13, resulting in inhibition of plasma ADAMTS13 activity.^1–3 ADAMTS13 is a plasma metalloprotease that cleaves ultra large von Willebrand factor (VWF) released from endothelium, thus regulating normal hemostasis and eliminating unwanted thrombosis upon vascular injury.^4–6

ADAMTS13 is a multi-domain glycoprotein, consisting of a metalloprotease, a disintegrin domain, the first thrombospondin-1 repeat, a cysteine-rich, and a spacer domain (i.e., MDTCS).^{4, 5, 7} All of these N-terminal domains are necessary for efficient recognition and proteolytic cleavage of VWF under various conditions;^7–12 the more distal C-terminus of ADAMTS13 contains seven additional TSP1 repeats and two CUB domains.^{4, 5, 13} The function of these more distal C-terminal domains remains elusive.

Recombinant and plasma ADAMTS13 may exist as multiple different conformations in solution.^14–16 A conformational transition from a “closed” state, in which the C-terminal domains are in a close proximity to the N-terminal domains, to an “open” state is thought to be mediated by the removal of an autoregulatory element of ADAMTS13 ^{7, 14–24}. ADAMTS13 is thought to function optimally in its “open” conformation, which can be promoted by acidic pH, binding of the VWF-D4 domain, and certain mouse monoclonal antibodies that primarily target the C-terminal domains of ADAMTS13 ^{14–16, 18, 25–27}.

Mouse monoclonal antibodies raised against the distal C-terminus of ADAMTS13 were shown to affect the conformation of ADAMTS13, leading to the exposure of spacer domain¹⁴ and perhaps a cryptic epitope in the metalloprotease domain ²⁶. Furthermore, plasma ADAMTS13 in acute iTTP patients appeared to exist primarily in an “open” conformation, when ADAMTS13 activity is completely inhibited ^{19, 20, 23, 26}. There is no data to demonstrate the interactions among human monoclonal antibodies that bind distal and proximal domains of ADAMTS13.

Here, we describe the first evidence of a human monoclonal antibody against the distal C-terminal domain of ADAMTS13 from a patient with iTTP capable of enhancing ADAMTS13 activity and modulating the inhibition of ADAMTS13 activity by another human monoclonal antibody in a cooperative fashion. The findings suggest a novel mechanism underlying the pathogenesis of iTTP.

Methods

Preparation of recombinant Anti-ADAMTS13 scFv’s, recombinant ADAMTS13, and truncated variants of ADAMTS13.

The constructs encoding anti-ADAMTS13 monoclonal scFv3–3 and scFv4–20 were described previously.²⁸ Recombinant scFv3–3 was expressed in the pComb3X vector (Scripps Research Institute, La Jolla, CA). After induction with 0.5 mM isopropyl-β-D-1-thiogalactopyranoside (Thermo Fisher Scientific) at 30 °C for 4 hours, the cells were pelleted and lysed with 10 mg of lysozyme, followed by sonication. Recombinant scFv3–3 and scFv4–20 were purified by Ni2+-chelating affinity chromatography (GE Healthcare) ²⁹. Further purification was performed using Amicon® Ultra-4 Centrifugal Filter Units (Millipore Sigma) by obtaining the flow-through of a 50 kDa cutoff spin concentrator from pooled fractions of protein eluted with 80 mM imidazole and further concentrating with a 3 kDa cutoff spin concentrator; the concentrated purified protein was then further concentrated on the same 3 kDa concentrator via serial buffer exchange with 20 mM HEPES, pH 7.45, containing 150 mM NaCl. Purity was determined by SDS/PAGE with Coomassie blue staining, and concentration was determined by absorbance at 280 nm using a NanoDrop spectrophotometer (ThermoFisher Scientific). The final products were characterized for extant purity using SDS/PAGE with Coomassie blue staining and quantified by absorbance at 280 nm using NanoDrop spectrophotometer.

Recombinant ADAMTS13 and truncated variants were prepared according to the protocol described previously ^{7, 30, 31}. Cell lines used to generate recombinant ADAMTS13 and truncated variants were HEK-293 for full-length ADAMTS13 ³¹, COS-7 for DelCUB ⁷, and Drosophila S2 cell line (Invitrogen) for other constructs including delCUB2, MT7, MT6, MT5, T5C, T6C, T7C, T8C ³⁰.

Cleavage of FRETS-VWF73 by recombinant ADAMTS13 and truncated variants in the absence or presence of human monoclonal antibodies.

Proteolytic activity of recombinant full-length ADAMTS13 and truncated variants was determined by the cleavage of a fluorogenic substrate as previously described ^32–34. A 5′-maleimide fluorescein-labeled VWF73 fragment (FRETS-VWF73) at a final concentration of 1 µM was incubated with either pooled normal human plasma (NHP) (2.5 µL; George King Biotech), or recombinant full-length ADAMTS13 or a truncated variant (final concentration, 4–30 ng/mL). Reaction conditions were conducted in 5 mM Bis-Tris, pH 6.0, 25 mM CaCl₂, 0.005% Tween 20, and 1 mM Pefabloc in a 96-well opaque white Nunc plate (Thermo Scientific), or the same reaction buffer titrated instead to pH 7.45, at either 25 °C or 37 °C for one hour as described in the text.

For assays in the presence of scFv3–3 and/or scFv4–20, varying concentrations of anti-ADAMTS13 scFv’s were incubated with 2.5 µL NHP for 15 minutes at 37 °C and then allowed to cool to room temperature prior to addition of FRETS-VWF73 substrate. A standard curve was prepared by varying the concentration of NHP (0–5 µL) as 1 mL NHP has the equivalent of 1 unit of active ADAMTS13 activity. Cleavage of FRETS-VWF73 was measured on a SpectraMax Gemini XPS plate reader (Molecular Dynamics).

Results

Human monoclonal antibody scFv3–3 increased the proteolytic activity of plasma (or wild-type) ADAMTS13.

Previous studies suggest a murine monoclonal antibody that binds the distal C-terminal domains may allosterically regulate ADAMTS13 activity,¹⁴ likely by disrupting the interaction between the N-terminal domains and C-terminal domains as outlined in Fig. 1A. To determine if a human monoclonal antibody fragment targeting the distal C-terminal domain, isolated from an iTTP patient, would affect ADAMTS13 activity the same way, we titrated a purified recombinant human monoclonal antibody fragment scFv3–3 into normal human plasma (NHP) which presumably contains the wild-type or native ADAMTS13 in the FRETS-VWF73-based assay as described in the Methods. We demonstrated that under the standard FRETS assay conditions (i.e., pH 6.0 and 25 °C) that is widely used in the clinical diagnosis of TTP, purified scFv3–3 was able to increase the proteolytic activity of plasma ADAMTS13 by ~3.0 fold, with the half maximal concentration (EC₅₀) of 0.51 ± 0.08 nM (mean ± SEM) and the Hill coefficient for the titration of 1.95 ± 0.16 (mean ± SEM) (Table 1 & Fig. 1B). At more physiological conditions (i.e., pH 7.45 and 25 °C), little plasma ADAMTS13 activity was detected in the absence of scFv3–3. However, a nearly 20-fold increase of plasma ADAMTS13 activity when a saturated concentration of scFv3–3 (20 nM) was added, with the EC50 of 1.03 ± 0.03 nM (mean ± SEM) and the Hill coefficient of 1.7 ± 0.02 (mean ± SEM) (Table 1 & Fig. 1B).

Fig. 1. Human monoclonal antibody scFv3–3 isolated from an iTTP patient enhances proteolytic activity of plasma ADAMTS13.

A. Simplified cartoon representation of the potential interaction between scFv3–3 and ADAMTS13. As a reference point, the spacer domain is labeled with “S” that is adjacent to the Cys-rich domain on the left (blue rectangle) and the second thrombospondin-repeat (TSP1–2) on the right (light blue circle). The human monoclonal antibody scFv3–3 is represented in the indented white box and labeled “3–3”. The binding of scFv3–3 to the C-terminal domains of ADAMTS13 leading to an “open” conformation with the dissociation of C-terminal CUB domains (two light green boxes), which allows for binding of substrate “VWF73” (red violet oval). B. Titration of scFv3–3 to plasma ADAMTS13 (NHP) at pH 6.0 (blue dots and a fitted line) and pH 7.45 (red squares and a fitted line), at 25 °C. C. Titration of scFv3–3 to plasma ADAMTS13 at pH 6.0 (blue dots and fitted line) and pH 7.45 (red squares and a fitted line), at 37 °C. Each experiment was repeated more than three times, and the means ± standard errors of the means (SEM) are shown.

Table 1.

scFv3–3 enhances proteolytic activity of plasma ADAMTS13 under various pH and temperature conditions

Temp	25 °C		37 °C
pH	6.0	7.45	6.0	7.45
Max	3.13 ± 0.16*	2.82 ± 0.08	3.00 ± 0.04	2.84 ± 0.08
Min	1.04 ± 0.02	0.17 ± 0.01	1.52 ± 0.18	0.13 ± 0.02
EC50 (nM)	0.51 ± 0.08	1.03 ± 0.03	0.28 ± 0.02	1.20 ± 0.06
Hill Coeffi.	1.95 ± 0.16	1.70 ± 0.01	1.14 ± 0.09	3.12 ± 0.17

^*The data shown are mean ± standard error of the mean (SEM) of more than three independent experiments (n>3). Coeffi, coefficient.

Again, at low pH 6.0 and 37 °C, the maximal increase of ADAMTS13 activity at the saturated concentration of scFv3–3 (20 nM) was ~3.0 fold with the EC₅₀ of 0.28 ± 0.03 nM (mean ± SEM) (Fig. 1C) and the Hill coefficient of 1.14 ± 0.09 (mean ± SEM) (Table 1). At physiologic pH 7.45 and 37 °C, scFv3–3 at the saturated concentration (20 nM) increased plasma ADAMTS13 activity by ~46 fold, with the EC₅₀ of 1.2 ± 0.06 nM (mean ± SEM) and the Hill coefficient of 3.1 ± 0.17 (mean ± SEM) (Table 1 & Fig 1C). These results demonstrate that binding of scFv3–3 to the full-length ADAMTS13 dramatically enhances its proteolytic activity towards VWF substrate; and the rate-enhancing effect is much greater at physiological pH than at low pH regardless of temperature.

Human monoclonal antibody scFv3–3 had no effect on proteolytic activity of the C-terminal truncated ADAMTS13 variants.

Our previous study has demonstrated that scFv3–3 binds the region between the 5^th TSP1 repeat and 1^st CUB domain, although the exact binding epitope is yet to be identified.²⁸ To determine if the C-terminal domains of ADAMTS13 are required for the rate-enhancing effect of scFv3–3 or to rule out the possibility of contaminated proteases in the purified preparation that may cleave the VWF73 peptide non-specifically, we titrated scFv3–3 to the recombinant full-length ADAMTS13 or various C-terminal truncated ADAMTS13 variants as illustrated in Fig. 2A in the FRETS-based assay. The results showed that only the activity of the full-length recombinant ADAMTS13, but not the C-terminal truncated recombinant ADAMTS13 variants, increased in the presence of saturated concentration of scFv3–3 (Fig 2B). The fold of increase of proteolytic activity in the recombinant full-length ADAMTS13 was quite similar to that in plasma ADAMTS13 (Fig. 2B). No proteolytic cleavage was detected in the FRETS-VWF73 assay with the addition of a C-terminal fragment consisting of TSP1 5–8-CUB domains (i.e.T5C), suggesting no other contaminating proteases that catalyze VWF73 substrate cleavage. These results demonstrate that the rate-enhancing activity of scFv3–3 is mediated through its interaction with the distal C-terminal domains of ADAMTS13.

Fig. 2. scFv3–3 enhances the proteolytic activity of plasma-derived and recombinant full-length ADAMTS13 but not of the C-terminal truncated variants.

A. Schematic representation of a full-length recombinant ADAMTS13 construct (FL-A13) and various ADAMTS13 truncated variants as indicated. B. The relative proteolytic activity of recombinant FL-A13 and truncated recombinant ADAMTS13 variants in presence of a saturated concentration of scFv3–3 (90–180 nM). The data are the normalized activity against that of each recombinant protein in the absence of scFv3–3 from three independent experiments.

Stimulating human monoclonal antibody scFv3–3 modulates the inhibition of an inhibitory human monoclonal antibody scFv4–20.

Previous and present studies have demonstrated that binding of a monoclonal antibody to the distal C-terminal domains of ADAMTS13 may expose the spacer domain of ADAMTS13, resulting in enhanced proteolytic activity of ADAMTS13. However, it remains unknown if such a stimulating antibody would have an impact on how an inhibitory antibody works; additionally, it is also unknown if an inhibitory antibody and a stimulating antibody would function cooperatively to modulate ADAMTS13 activity. To address this important question, we titrated a human monoclonal antibody against the spacer domain of ADAMTS13 (i.e., scFv4–20), which was well characterized in our previous studies,^{28, 29, 35} to plasma ADAMTS13 in the FRETS-VWF73 assay. We first confirmed the inhibitory potency of the newly expressed and purified scFv4–20 as expected. The half-maximal inhibitory concentration (IC₅₀) of scFv4–20 was 0.65 ± 0.03 nM (mean ± SEM) (Fig. 3A & 3B), similar to what we previously reported ²⁹ with the Hill coefficient of 1.54 ± 0.09 (mean ± SEM) (Table 2). We then performed the inhibition assays for scFv4–20 in the presence of increasing concentrations of scFv3–3 (0, 0.7, 1.25, 2.5, and 5.0 nM) under the standard assay conditions (pH 6 and 25 °C). As shown, in the presence of increasing concentrations of scFv3–3, the inhibition curve shifted upwards with the calculated IC₅₀s (mean ± SEM) of scFv4–20 in the presence of 0, 1.25, 2.5, and 5 nM of scFv3–3 of 0.64 ± 0.03, 0.48 ± 0.01, 0.45 ± 0.08, and 0.39 ± 0.04 nM, respectively. The Hill coefficients (mean ± SEM) in the titrations in the presence of 0, 1.25, 2.5, and 5 nM of scFv3–3 were 1.54 ± 0.09, 2.14 ± 0.14, 2.18 ± 0.24, and 1.74 ± 0.18, respectively (Fig. 3C, 3D, and Table 2). These results suggest that scFv4–20 exhibits no cooperativity with scFv3–3 under the acidic assay condition, which is known to activate ADAMTS13 activity.¹⁴ However, when the titrations were performed at the physiologically relevant assay condition (pH 7.45 at 37 °C), the addition of increasing concentrations of scFv3–3 to a reaction that contains scFv4–20 dramatically shifted the titration curve upwards with increased steepness of the inhibition curve (Fig. 3E). In the presence of 0.7, 1.25, and 5.0 nM of scFv3–3, the IC₅₀ (mean ± SEM) for scFv4–20 to inhibit ADAMTS13 activity was 0.19 ± 0.01, 0.29 ± 0.02, and 0.61 ± 0.09 nM, respectively. However, when the concentration of scFv4–20 reached to the threshold of ~0.1 nM, strong positive cooperativity between scFv3–3 and scFV4–20 to mediate inhibition of ADAMTS13 activity was observed, as demonstrated by the increase of Hill coefficients from 1.70 to 5.67 as a function of increasing concentration of scFv3–3 from 0.7 to 5.0 nM, respectively (Fig 3E and Table 2). These results suggest that scFv3–3 is able to lessen antibody-mediated inhibition of ADAMTS13 when the concentration of inhibitor is relatively low, but at higher inhibitory antibody concentrations, robust inhibition is observed to occur in a strongly cooperative manner, and the effect of scFv3–3 on mitigating inhibition is significantly decreased.

Fig. 3. Synergistic effects of scFv3–3 and scFv4–20 on plasma ADAMTS13 activity under various conditions.

A. Schematic representation of the spacer domain of ADAMTS13 (orange square labeled S) bound by scFv4–20 (red octagon labeled 4–20) with concomitant exposure to a substrate (red violet oval labeled VWF73), leading to inhibition of the cleavage of a substrate by ADAMTS13. B. Dose-dependent inhibition of plasma ADAMTS13 activity (mean ± SEM) by a human monoclonal antibody scFv4–20 at pH 6.0 and temperature 25 °C. C. Cartoon representation of scFv3–3 (blue polygon labeled 3–3) bound to the C-terminus of ADAMTS13 and then exposed to scFv4–20 (red octagon labeled 4–20), which blocks the binding of a VWF73 substrate. D. Inhibition of plasma ADAMTS13 by scFv4–20 (0–10 nM) in the absence (blue line) or presence of scFv3–3 (1.25, 2.5, and 5.0 nM as indicated in the graph) at pH 6.0 and 25 °C. E. Inhibition of plasma ADAMTS13 activity by scFv4–10 (0–10 nM) in the absence (blue line) or presence of scFv3–3 (0.7, 1.25, and 5.0 nM as indicated in the graph) at pH 7.45 and 37 °C. Each experiment was repeated at least three times and the means ± SEM are shown.

Table 2.

scFv4–20-mediated inhibition of ADAMTS13 activity in the presence and absence of scFv3–3

Assay conditions	pH 6.0 and 25 °C	pH 7.45 and 37 °C
[scFv3–3] 0
IC50 (nM)	0.64 ± 0.03*	NA
Hill coeffi.	1.54 ± 0.09	NA
[scFv3–3] 1.25 nM
IC50 (nM)	0.48 ± 0.01	0.29 ± 0.02
Hill coeffi.	2.14 ± 0.14	2.26 ± 0.21
[scFv3–3] 5.0 nM
IC50 (nM)	0.39 ± 0.04	0.61 ± 0.09
Hill coeffi.	1.74 ± 0.18	5.67 ± 1.17

^*All data are presented as the mean ± SEM; NA, not determined.

Temp., Temperature; coeffi., coefficient.

Discussion

The present study for the first time provides biochemical and functional evidence to support the hypothesis that binding of a monoclonal autoantibody against the distal domains of ADAMTS13 isolated from a patient with iTTP is able to enhance ADAMTS13 activity and somehow modulates the inhibition profile of an inhibitory monoclonal autoantibody that primarily targets the spacer domain of ADAMTS13. These findings shed new light on the molecular mechanism underlying the pathogenesis of iTTP.

ADAMTS13 is known to exhibit in multiple intermediate conformations: the open, closed, and intermediate either as a monomeric protein ^{15, 36} or a possible oligomeric form in solution or plasma.³⁷ Lowering pH and binding of ADAMTS13 to a substrate such as VWF (D4 domain) or an antibody against the various C-terminal domains of ADAMTS13 may stabilize multiple intermediate conformations of ADAMTS13 with a different maximal rates of catalysis^{15, 22, 26} or convert ADAMTS13 from its “closed” and more stable conformation to an “open” but less stable conformation.^{14–16, 18, 27} The “open” ADAMTS13 species are found to be dramatically increased in patients with acute iTTP, then reduced or normalized during clinical remission or essentially undetectable in the healthy individual.³⁸

In acidic conditions, ADAMTS13 undergoes dramatic conformational changes to remove the allosteric inhibition by the distal C-terminal domains and expose the spacer domain,^{14, 39} which is primarily responsible for substrate and antibody binding.^{7, 29, 40, 41} In this case, the effect of scFv3–3 on ADAMTS13 function is further increased but to a lesser degree than that under physiologically relevant conditions. However, at a physiologically relevant pH 7.45, ADAMTS13 may adopt a more “closed” intermediate conformation. Therefore, the binding of scFv3–3 may drive the ADAMTS13 equilibrium towards an “open” conformation, allowing the exposure of spacer domain, possibly by increasing the binding of ADAMTS13 to its substrate, accelerating proteolysis. Alternatively, it appears based on recent data using mouse monoclonal antibodies that C-terminal antibodies may affect the turnover number (k_cat) of ADAMTS13 more so than the substrate binding affinity (K_M).²⁶ Further studies are needed to determine how scFv3–3 stimulates substrate cleavage by ADAMTS13.

We have observed evidence of potential positive cooperativity when scFv3–3 is titrated to ADAMTS13, particularly in physiologic conditions. When performed at pH 7.45 and 37 ºC, the Hill coefficient is greater than 3.0. This suggests that ADAMTS13 may function as an oligomer, similar to how hemoglobin works. In the case of hemoglobin, which can be potentially interpreted as an oligomeric “dimer-of-dimers” of two alpha and two beta subunits non-covalently bound together, binding of an oxygen molecule to one subunit changes the conformation of adjacent subunits, which increases the probability that another subunit will bind a different oxygen molecule. In the case of ADAMTS13, binding of one scFv3–3 to one ADAMTS13 molecule may increase the probability that another ADAMTS13 will cleave substrate in a similar manner. Alternatively, ADAMTS13 may function as a monomer capable of adapting multiple stable intermediate conformations which bind substrate and antibodies with different kinetics, which may also explain the positive cooperativity observed. It is known that ADAMTS13 is capable of forming oligomers (or aggregates) in solution; our ongoing work in the lab is exploring whether functional ADAMTS13 oligomers or aggregates form in plasma or if ADAMTS13 functions as a monomeric protease.

The effect of scFv3–3 on inhibition by anti-spacer antibody (scFv4–20) is only observed in the physiologic conditions, but results are dramatic. The scFv3–3 antibody is able to mitigate inhibition of ADAMTS13 in a dose-dependent manner under physiologic conditions, demonstrated by the fact that the IC₅₀ of scFv4–20 increases by approximately 3-fold as the concentration of scFv3–3 increases from 0.7 nM to 5.0 nM. However, evidence of a strong positive cooperativity of scFv4–20 inhibition is observed as the scFv3–3 concentration increases. This may reveal significant new mechanistic insights into ADAMTS13 function and the pathophysiology of iTTP. It is possible that binding of an inhibitory antibody to one monomer of ADAMTS13 affects the conformation of other monomers in a complex, and that when a threshold concentration of inhibitor is bound, the other ADAMTS13 monomers are easier to inhibit, so enzyme activity decreases dramatically.

In summary, the results let us hypothesize that in the presence of a low concentration of inhibitory autoantibodies in patients with acute iTTP, the antibodies against the distal carboxyl-terminus may facilitate proteolytic cleavage of VWF substrate and protect against ADAMTS13 inhibition; in the presence of a high concentration of inhibitory autoantibodies (e.g., above a certain threshold), the protective effect of the anti-carboxyl-terminal antibody is significantly lessened, thus the inhibitory autoantibodies appear to work cooperatively to prevent VWF from being cleaved. These results indicate that ADAMTS13 may function as a potential oligomer in human plasma. Further investigation of the elements that affect oligomerization of ADAMTS13, if it does occur, is crucial for our understanding of the mechanism of iTTP. Our findings may help redesign a more efficacious therapeutic strategy for such a potentially fatal disorder.

Essentials.

• Immune thrombotic thrombocytopenic purpura (iTTP) is caused by IgG-type autoantibodies against ADAMTS13.

• The mechanism underlying the autoantibody-mediated inhibition is not fully understood.

• Here, we demonstrate that a human monoclonal antibody targeting the distal carboxyl terminus of ADAMTS13 isolated from a patient with iTTP stimulates ADAMTS13 activity.

• Binding of this stimulatory human monoclonal antibody to ADAMTS13 modulates its inhibition by a different human monoclonal antibody under physiologic conditions.