| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Department of Biochemistry and Biophysics-CRISCEB, Second University of Naples, 80138 Naples, Italy
1Correspondence: Department of Biochemistry and Biophysics-CRISCEB, Second University of Naples, via Costantinopoli 16, 80138 Naples, Italy. E-mail: ragone@unina2.it or raffrag{at}tiscali.it
ABSTRACT
There is evidence that mild elevations of tHcy are associated with an increased risk for occlusive vascular disease, thrombosis, and stroke. It is hypothesized here that cellular toxicity could indirectly result from auto-oxidation of homocysteine to homocystine. Elevated levels of total plasma homocysteine could be the primary cause of increased vascular risk, causing endothelial damage through a mechanism similar to that of cystine precipitation, which is known to cause stone formation in cystinosis and cystinuria. In fact, only traces of homocysteine circulate in plasma as the free thiol; the remainder is present as oxidation products. Of these, the symmetric disulfide homocystine is scarcely soluble at neutral pH. Its saturation limit is so close to the concentration of homocysteine in normal plasma that a transient increase of homocysteine levels could lead to precipitation of homocystine microcrystals in the bloodstream. These could damage endothelial tissue, acting as a mechanic primer for subsequent prothrombotic blood vessel alterations.—Ragone, R. Homocystine solubility and vascular disease.
Key Words: homocysteine • cystine solubility • redox thiol status
EVIDENCE IS MOUNTING for adding high plasma concentrations (>15 µM) of total homocysteine (tHcy) to the list of risk factors for vascular disease (1
, 2)
according to the theory of McCully and Wilson (3)
. Numerous studies suggest that mild elevations of tHcy are associated with an increased risk for occlusive vascular disease, thrombosis, and stroke, and several mechanisms through which hyperhomocysteinemia may exert its thrombogenic action have been suggested (4)
. Nevertheless, the indirect hypothesis that cellular toxicity could result from auto-oxidation of homocysteine to homocystine does not seem to have been adequately considered. In fact, only trace amounts of tHcy circulate in plasma as the free thiol; the remainder is present as the symmetric disulfide homocystine (Hcy2), mixed disulfide homocysteine-cysteine (Hcy-Cys), and protein-bound homocysteine (Hcy-P) (2
, 5
, 6)
. It should be noted that at neutral pH the solubility of Hcy2 is very low (7
, 8)
, being close to and presumably even lower than that of cystine (Cys2). Therefore, it may be suggested that an occurrence similar to that of Cys2 precipitation, which is known to cause stone formation in cystinosis and cystinuria, could play a role in vascular disease.
HYPOTHESIS
Since in normal plasma tHcy is almost completely oxidized (2
, 5
, 6)
, it is suggested here that Hcy2 might precipitate in the bloodstream. Actually, tHcy concentration in plasma is not far from the solubility limit of Hcy2, which might be reached under some specific circumstances. Fast oxidation of high tHcy levels may lead to precipitation of Hcy2, if the amount of this symmetric disulfide exceeds its saturation concentration. Homocystine could therefore form insoluble crystals in vascular districts, all the more so because the vascular lining has a limited capacity to metabolize Hcy through both the alternate transmethylation and transsulfuration pathways. It is likely that circulating microcrystals, whenever transient and not necessarily leading to Hcy2 accumulation, could act as a mechanic primer for blood vessel damage. In response to this specific homocystine-induced injury, the vascular endothelium might undergo phenotypic and physiological alterations similar to those caused by a variety of agents (e.g., thrombin, cytokines, oxidized lipids, shear stress, and possibly infectious agents). In other words, damaged endothelial cells may become prothrombotic. To test this hypothesis, it is necessary to obtain information on the solubility of Hcy2 and conditions under which the concentration of tHcy in the blood could be close to saturation levels.
AQUEOUS SOLUBILITY OF HOMOCYSTINE
Data on the saturation concentration of Hcy2 (S) do not exist in the literature, presumably because of the low solubility of this compound. This rendered the results from absorption experiments in vivo uninterpretable (7
, 8)
. The only information available is on the racemic form (9)
, which is a 1:1 mixture of D- and L-Hcy2, but the value of 
740 µM at room temperature is probably different from the solubility of the L isomer present in the body. Preliminary saturation data in cold water (R. Ragone, unpublished observation) suggest that S is in the micromolar range (roughly 60 µM for both D- and L-Hcy2). Moreover, the solubility of Hcy2 may be estimated with a good approximation considering that from the organic chemistry point of view, Hcy2 is the next-higher homologue of Cys2. This means that compared with Cys2, its side chain carries two additional -CH2- groups (i.e., each disulfide moiety carries one additional hydrogen-saturated carbon). Hcy2, then, is likely less soluble than Cys2, because the water solubility of organic species usually decreases as the number of hydrogen-saturated groups increases.
From a comparative analysis of the saturation concentrations of several sulfur-containing substances (10)
at neutral pH and room temperature as a function of the number of hydrogen-saturated carbons (NC) present in their structure, it can be inferred that the solubility decrease produced by each additional -CH2- group is almost independent of the class and amounts to 0.54 ± 0.06 (mean±SD) logarithmic units (Fig. 1
). On this basis, the solubility of Hcy2 should be lower than that of Cys2 by about one order of magnitude, corresponding to a decrease of 2 x 0.54 = 1.08 logarithmic units. As shown in Fig. 1
, the solubility of Hcy2 can be extrapolated drawing a line with the same slope as those relative to the sulfur-containing substances from the point corresponding to the solubility of Cys2 (NC=4, logS = -3.34) (11)
down to NC = 6 (logS = -4.42 ± 0.12). This procedure estimates that the water solubility of Hcy2 at room temperature is close to our unrefined experimental datum (S=38 µM). The actual solubility of both D- and L-Hcy2 can then be reasonably placed at 50 µM.
|
BASAL AND TRANSIENT CONCENTRATIONS OF PLASMA HOMOCYSTEINE
The working ranges of tHcy under fasting conditions are 5–15 µM for normal plasma and 15–25, 25–50, and 50–500 µM for mild, intermediate, and severe hyperhomocysteinemia, respectively (12)
. On average, 99% of tHcy is oxidized to Hcy2 (5–10%), Hcy-Cys (5–10%), and Hcy-P (80–90%) (12)
. Thus, the majority of oxidized Hcy is protein bound. Results from studies of homocystinuric and cobalamin-deficient patients (13)
indicate that fasting levels of Hcy-P increase hyperbolically as a function of tHcy, approaching a maximal binding of 140 µM, with an equivalent reduction of protein-bound Cys. On the other hand, free oxidized species (Hcy2 and Hcy-Cys) increase exponentially as a function of tHcy, and the increase is most pronounced at > 140 µM tHcy, where the protein binding sites are saturated. It has recently been reported that protein multimerization and precipitation might be relevant for homocysteine-induced vascular damage via modification of lysyl residues by Hcy thiolactone (14)
, to which Hcy may be efficiently converted in endothelial cells (15)
.
It appears therefore that the chances for predominant Hcy2 formation from unbound Hcy excess are likely only at tHcy concentrations > 140 µM, all the more so because Hcy2 solubility might be even > 50 µM in vivo. However, these findings hold true only under stationary (fasting) conditions, where the concentration of tHcy represents the sum of the basal amounts of the various reduced and oxidized forms circulating in plasma. As such, it does not provide information on transient concentrations of these species subsequent to any homocysteine and/or methionine intake; i.e., the fasting concentration is measured after the dynamic sequence of related redox and disulfide exchange reactions (referred to as redox thiol status) involved in Hcy metabolic pathways has reached completion. It therefore has nothing to do with the time course of this system of reactions. In recent studies of the afterload kinetics of plasma Hcy in healthy subjects (16)
, it was found that free oxidized Hcy increased significantly, reaching a maximum after 30 min and declining slowly toward basal values. The response was slower for Hcy-P, which increased only later on, maximizing after 1–2 h and gradually approaching the fasting value. In healthy subjects, the concentration of reduced Hcy reached concentrations 20-fold above fasting values < 15 min after oral administration of methionine and/or homocysteine (16)
. In homocystinuric patients, it was increased up 200-fold the amount in healthy subjects (17)
. Overall, this results in marked changes in other aminothiol species (8
, 16
, 17)
, but even small fluctuations in total concentrations of plasma aminothiol compounds show the same effect (18)
.
This body of data suggests that the kinetics of Hcy needs to be accurately investigated before drawing a conclusion hinging on basal concentrations, because it cannot be excluded that the transient concentration of Hcy2 may be close to saturation levels in vivo. In the meantime, available kinetic data (16)
clarify that upon oxidation of Hcy, the plasma concentration of Hcy2 is expected to rise well before Hcy-P formation. During the complex sequence of events leading to basal levels of tHcy, formation of Hcy2 precedes that of Hcy-P because Hcy oxidizes to Hcy2 faster than to Hcy-P. Depending on the concentration of tHcy, Hcy2 could therefore transiently reach its saturation limit before proceeding toward the fasting condition, where its concentration is lower than its solubility and Hcy-P is the most populated oxidized species.
CLINICAL CONSIDERATIONS
To explain why existing evidence points to functional instead of mechanic damage to the endothelium by elevation of Hcy plasma levels, it can be hypothesized that insoluble Hcy2 crystals may form after sufficient methionine and/or homocysteine intake, act as a mechanic primer for tissue damage, and then dissolve during the time lag needed to reach basal conditions. However, current clinical evidence needs to be specifically addressed. The objection could be raised that in order for Hcy2 microcrystals to play a pathogenetic role, Hcy2 precipitates should have been observed in endothelial cells or other tissues. Instead, unlike cystinosis and cystinuria, neither the incidence of gall or renal stone formation in homocystinuric patients is increased nor have tissue deposits of Hcy2 precipitates or crystals been described. Moreover, it is recognized that even moderate elevations of tHcy plasma levels may significantly increase vascular risk. It is apparent, however, that this objection is based on a consideration of fasting (basal) values of various oxidized forms of Hcy. I have shown here that transient concentrations of Hcy are likely to play an important role, being notably higher than basal ones.
Another objection could be that one would expect to see vascular damage in cystinosis or cystinuria, by analogy with the hypothesis that limited Hcy2 solubility and precipitation may cause vascular damage in hyperhomocysteinemia. Instead, neither cystinosis nor cystinuria shows vascular alterations similar to those found in homocystinuria, although Cys2 stones are formed. It is important to note, however, that homocystinuria could be due to metabolic defects involving transmethylation and transsulfuration pathways, which produce high levels of Hcy in the blood, thus making some of the Hcy (but not enough to cause Hcy2 precipitation) to spill over into the urine. On the other hand, cystinuria develops due to a defect in the transport systems that normally reabsorb Cys2 from the glomerular filtrate, leading to high levels of Cys2 in urine. The same transporters are affected in the intestine, lowering absorption from dietary sources. The concentration of Cys2 in blood, therefore, is normal or even low to lead to formation of precipitates in the bloodstream.
Finally, formation of Hcy2 precipitates has not been observed in severe hyperhomocysteinemia, where even fasting levels of Hcy2 might approach the saturation concentration. In this case, the buffering action exerted by protein-bound Hcy may play an important role, protecting vulnerable structures against harmful levels of Hcy2. Nevertheless, the high frequency of vascular events occurring even in young homozygote individuals (1)
could be explained in that the saturation concentration of Hcy2 can be easily reached starting from a high basal level of tHcy, thereafter promoting prothrombotic alterations.
CONCLUSION
Although the aqueous solubility of L-Hcy2 seems low to measure accurately, its determination is being attempted in our laboratory and should be matter for a separate paper. In the meantime, preliminary saturation experiments on D- and L-Hcy2 solutions as well as analysis of solubilities of several organic sulfur compounds allow us to approximate what the concentration of Hcy2 might transiently reach in plasma. The existing literature suggests that levels of tHcy sufficiently high to achieve the saturation limit of Hcy2 may form immediately after methionine and/or homocysteine intake, being much higher than the basal value. Therefore, the possibility of this event being the primary cause of thrombogenic endothelial damage (through a mechanism similar to that leading to the formation of cystine stones) deserves careful attention.
Received for publication October 12, 2001. Accepted for publication November 5, 2001.
REFERENCES
This article has been cited by other articles:
|
|
 |
|
  A. B. Sobol, E. Bald, and J. Loba Fractions of Total Plasma Homocysteine in Patients with Ischemic Stroke Before the Age of 55 Years Angiology, March 1, 2005; 56(2): 201 - 209. [Abstract] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
) Di-alkyl sulfides. (
) Di-alkyl disulfides. (
) Thiophenes. (
) Thiols. (+) Cys and Hcy. (*) Cys2 and Hcy2. Molar solubilities (in logarithmic units) are plotted against the number of hydrogen-saturated carbons present in each compound. The line for di-alkyl sulfides was obtained by linear fitting [slope = (-0.53 ± 0.05) NC-1, intercept = 0.67 ± 0.22, r = 0.996]. Dashed lines were drawn to extrapolate Hcy and Hcy2 solubilities from experimental values for Cys and Cys2 (11)