To receive more information about up-to-date research on micronutrients, sign up for the free, semi-annual LPI Research Newsletter here.
Vitamin K is a fat-soluble vitamin. Originally identified for its role in the process of blood clot formation ("K" is derived from the German word "koagulation"), vitamin K is essential for the functioning of several proteins involved in physiological processes that encompass, but are not limited to, the regulation of blood clotting (coagulation) (1). Naturally occurring forms of vitamin K include a number of vitamers known as vitamin K1 and vitamin K2 (Figure 1). Vitamin K1 or phylloquinone is synthesized by plants and is the predominant form in the diet. Vitamin K2 includes a range of vitamin K forms collectively referred to as menaquinones. Most menaquinones are synthesized by human intestinal microbiota and found in fermented foods and in animal products. Menaquinones differ in length from 1 to 14 repeats of 5-carbon units in the side chain of the molecules. These forms of vitamin K are designated menaquinone-n (MK-n), where n stands for the number of 5-carbon units (MK-2 to MK-14) (2, 3). Widely used in animal husbandry, the synthetic compound known as menadione (vitamin K3) is a provitamin that needs to be converted to menaquinone-4 (MK-4) to be active (4).
Vitamin K functions as a cofactor for the enzyme, gamma-glutamylcarboxylase (GGCX), which catalyzes the carboxylation of the amino acid glutamic acid (Glu) to gamma-carboxyglutamic acid (Gla). Vitamin K-dependent gamma-carboxylation that occurs only on specific glutamic acid residues in identified vitamin K-dependent proteins (VKDP) is critical for their ability to bind calcium (5).
Although vitamin K is a fat-soluble vitamin, the body stores very small amounts that are rapidly depleted without regular dietary intake. Perhaps because of its limited ability to store vitamin K, the body recycles it through a process called the vitamin K-epoxide cycle (Figure 2). The vitamin K cycle allows a small amount of vitamin K to be reused many times for protein carboxylation, thus decreasing the dietary requirement. Briefly, vitamin K hydroquinone (reduced form) is oxidized to vitamin K epoxide (oxidized form). The reaction enables gamma-glutamylcarboxylase to carboxylate selective glutamic acid residues on vitamin K-dependent proteins. The recycling of vitamin K epoxide (oxidized form) to hydroquinone (reduced form) is carried out by two reactions that reduce vitamin K epoxide (KO) to vitamin K quinone and then to vitamin K hydroquinone (KH2; Figure 2). Additionally, the enzyme vitamin K oxidoreductase (VKOR) catalyzes the reduction of KO to vitamin K quinone and may be involved—as well as another yet-to-defined reductase—in the production of KH2 from vitamin K quinone (6, 7). The anticoagulant drug warfarin acts as a vitamin K antagonist by inhibiting VKOR activity, hence preventing vitamin K recycling (see Coagulation).
The ability to bind calcium ions (Ca2+) is required for the activation of the several vitamin K-dependent clotting factors, or proteins, in the coagulation (clotting) cascade. The term, coagulation cascade, refers to a series of events, each dependent on the other, that stop bleeding through clot formation. Vitamin K-dependent gamma-carboxylation of specific glutamic acid residues in those proteins makes it possible for them to bind calcium. Factors II (prothrombin), VII, IX, and X make up the core of the coagulation cascade. Protein Z appears to enhance the action of thrombin (the activated form of prothrombin) by promoting its association with phospholipids in cell membranes. Protein C and protein S are anticoagulant proteins that provide control and balance in the coagulation cascade; protein Z also has an anticoagulatory function. Control mechanisms for the coagulation cascade exist since uncontrolled clotting may be as life threatening as uncontrolled bleeding. Vitamin K-dependent coagulation factors are synthesized in the liver. Consequently, severe liver disease results in lower blood levels of vitamin K-dependent clotting factors and an increased risk for uncontrolled bleeding (hemorrhage) (8).
Oral anticoagulant therapy with vitamin K antagonists
Some people are at increased risk of forming clots, which could block the flow of blood in arteries of the heart, brain, or lungs, resulting in myocardial infarction (heart attack), stroke, or pulmonary embolism, respectively. Abnormal clotting is not related to excessive vitamin K intake, and there is no known toxicity associated with vitamin K1 or vitamin K2 (see Toxicity). Some oral anticoagulants, such as warfarin (Coumadin, Jantoven), inhibit coagulation by antagonizing the action of vitamin K. Warfarin prevents the recycling of vitamin K by blocking VKOR activity, thus creating a functional vitamin K deficiency (Figure 2). Inadequate gamma-carboxylation of vitamin K-dependent coagulation proteins interferes with the coagulation cascade, which inhibits blood clot formation. Large quantities of dietary or supplemental vitamin K can overcome the anticoagulant effect of vitamin K antagonists, thus patients taking these drugs are cautioned against consuming very large or highly variable quantities of vitamin K (see Drug interactions). Experts now advise a reasonably constant dietary intake of vitamin K that meets current dietary recommendations (90-120 mcg/day) for patients taking vitamin K antagonists like warfarin (9). Finally, because of the high variability in patients' response to vitamin K antagonists, it has been suggested that daily supplementation of low-dose phylloquinone may improve the stability of anticoagulation therapy. Yet, several meta-analyses recently highlighted the lack of sufficient evidence to support this option for those taking warfarin (10-12).
Skeletal formation and prevention of soft tissue calcification
Vitamin K-dependent gamma-carboxylation is essential to several bone-related proteins, including osteocalcin, anticoagulation factor protein S, matrix gamma-carboxylated glutamate (Gla) protein (MGP), Gla-rich protein (GRP), and periostin (originally called osteoblast-specific factor-2). Osteocalcin (also known as bone Gla protein) is synthesized by osteoblasts (bone-forming cells); the synthesis of osteocalcin is regulated by the active form of vitamin D, 1,25-dihydroxyvitamin D (calcitriol). The calcium-binding capacity of osteocalcin requires vitamin K-dependent gamma-carboxylation of three glutamic acid residues. Although its function in bone mineralization is not fully understood, osteocalcin is required for the growth and maturation of calcium hydroxyapatite crystals (see Osteoporosis) (13).
Protein S appears to play a role in the breakdown of bone mediated by osteoclasts. Individuals with inherited protein S deficiency suffer complications related to increased blood clotting, as well as osteonecrosis (14, 15). Protein S can bind and activate receptors of the TAM family that are involved in phagocytosis. Mutations in TAM receptors can result in visual impairment, defective spermatogenesis, autoimmune disorders, and platelet disorders (16).
MGP has been found in cartilage, bone, and soft tissue, including blood vessel walls, where it is synthesized and secreted by vascular smooth muscle cells. MGP has been involved in the prevention of calcification at various sites, including cartilage, vessel wall, skin elastic fibers, or trabecular meshwork in the human eye (see Vascular calcification) (17). Moreover, several VKDPs, including MGP, have been associated with calcification sites in arteries, skin, kidneys, and eyes in certain inherited conditions, such as pseudoxanthoma elasticum and beta-thalassemia (18, 19).
The vitamin K-dependent proteins, GRP and periostin, are also synthesized in bone tissue, but their roles in bone metabolism are still unclear (20, 21). Expressed in normal human skin and vascular tissues, GRP has been colocalized with abnormal mineral deposits in the extracellular matrix in calcified arteries and calcified skin lesions (22).
Expressed in most connective tissues, including skin and bone, periostin was initially associated with cell adhesion and migration. This VKDP also appears to promote angiogenesis (formation of new blood vessels) during cardiac valve degeneration and tumor growth (23, 24).
Current research suggests that reduced GGCX activity and/or lower vitamin K bioavailability may impair the activity of VKDPs and contribute to bone mineralization defects and abnormal soft tissue calcification (see Disease Prevention) (25).
Regulation of cellular functions
Growth arrest-specific gene 6 protein (Gas6) is a vitamin K-dependent protein that was identified in 1993. It has been found throughout the nervous system, as well in the heart, lungs, stomach, kidneys, and cartilage. Identified as a ligand of the TAM family of transmembrane tyrosine kinase receptors, Gas6 appears to be a cellular growth regulation factor with cell-signaling activities. Gas6 has been involved in diverse cellular functions, including phagocytosis, cell adhesion, cell proliferation, and protection against apoptosis (5). It may also play important roles in the developing and aging nervous system (reviewed in 26). Further, Gas6 appears to regulate platelet signaling and vascular hemostasis (27). Expressed in most tissues and involved in many cellular functions, Gas6 has also been linked to several pathological conditions, including clot formation (thrombogenesis), atherosclerosis, chronic inflammation, and cancer growth (28-30).
Overt vitamin K deficiency results in impaired blood clotting, usually demonstrated by laboratory tests that measure clotting time. Symptoms include easy bruising and bleeding that may be manifested as nosebleeds, bleeding gums, blood in the urine, blood in the stool, tarry black stools, or extremely heavy menstrual bleeding. In infants, vitamin K deficiency may result in life-threatening bleeding within the skull (intracranial hemorrhage) (8).
Vitamin K deficiency is uncommon in healthy adults for a number of reasons: (1) vitamin K is widespread in foods (see Food sources); (2) the vitamin K cycle conserves vitamin K (see Vitamin K oxidation-reduction cycle); and (3) bacteria that normally inhabit the large intestine synthesize menaquinones (vitamin K2), although it is unclear whether significant amounts are absorbed and utilized (see Food sources). Adults at risk for vitamin K deficiency include those taking vitamin K antagonists and individuals with significant liver damage or disease (8). Additionally, individuals with fat malabsorption disorders, including inflammatory bowel disease and cystic fibrosis, may be at increased risk of vitamin K deficiency (31-33).
Newborn babies who are exclusively breast-fed are at increased risk for vitamin K deficiency, because human milk is relatively low in vitamin K compared to formula. Newborn infants, in general, have low vitamin K status for the following reasons: (1) vitamin K transport across the placental barrier is limited; (2) liver storage of vitamin K is very low; (3) the vitamin K cycle may not be fully functional in newborns, especially premature infants; and (4) the vitamin K content of breast milk is low (5). Infants whose mothers are on anticonvulsant medication to prevent seizures are also at risk for vitamin K deficiency. Vitamin K deficiency in newborns may result in a bleeding disorder called vitamin K deficiency bleeding (VKDB) of the newborn (reviewed in 34). Because VKDB is life threatening and easily prevented, the American Academy of Pediatrics and a number of similar international organizations recommend that an intramuscular dose of phylloquinone (vitamin K1) be administered to all newborns (35).
Controversies around vitamin K administration to newborns
Vitamin K and childhood leukemia: In the early 1990s, two retrospective studies were published suggesting a possible association between phylloquinone injections in newborns and the development of childhood leukemia and other forms of childhood cancer. However, two large retrospective studies in the US and Sweden, which reviewed the medical records of 54,000 and 1.3 million children, respectively, found no evidence of a relationship between childhood cancers and phylloquinone injections at birth (36, 37). Moreover, a pooled analysis of six case-control studies, including 2,431 children diagnosed with childhood cancer and 6,338 cancer-free children, found no evidence that phylloquinone injections for newborns increased the risk of childhood leukemia (38). In a policy statement, the American Academy of Pediatrics recommended that routine vitamin K prophylaxis for newborns be continued because VKDB is life threatening and the risks of cancer are unproven and unlikely (35). In the last few years, physicians have reported a rise in late-onset cases of VKDB due to an increasing trend of parental omission or refusal of newborn vitamin K prophylaxis (39).
Lower doses of vitamin K1 for premature infants: The results of two studies of vitamin K levels in premature infants suggest that the standard initial dose of phylloquinone (vitamin K1) for full-term infants (1.0 mg) may be too high for premature infants (40, 41). These findings have led some experts to suggest the use of an initial phylloquinone dose of 0.3 mg/kg for infants with birth weights of less than 1,000 g (2 lbs, 3 oz), and an initial phylloquinone dose of 0.5 mg would probably prevent hemorrhagic disease in newborns (40).
In January 2001, the US Food and Nutrition Board (FNB) of the Institute of Medicine established the adequate intake (AI) level for vitamin K based on consumption levels in healthy individuals. The AI for infants was based on estimated intake of vitamin K from breast milk (42).
|Adequate Intake (AI) for Vitamin K|
|Life Stage||Age||Males (mcg/day)||Females (mcg/day)|
|Adults||19 years and older||120||90|
|Pregnancy||18 years and younger||-||75|
|Pregnancy||19 years and older||-||90|
|Breast-feeding||18 years and younger||-||75|
|Breast-feeding||19 years and older||-||90|
The discovery of vitamin K-dependent proteins in bone has led to research on the role of vitamin K in maintaining bone health.
Vitamin K and bone health: observational studies
Vitamin K1: Observational studies have found a relationship between phylloquinone (vitamin K1) and age-related bone loss (osteoporosis). The Nurses' Health Study followed more than 72,000 women for 10 years. In an analysis of this cohort, women whose phylloquinone intakes were lower than 109 micrograms/day (mcg/day) had a 30% higher risk for hip fracture compared to women with intakes equal to or above 109 mcg/day (43). Another prospective study in over 800 elderly men and women, followed in the Framingham Heart Study for seven years, found that participants with dietary vitamin K intakes in the highest quartile (median, 254 mcg/day) had a 65% lower risk of hip fracture than those with intakes in the lowest quartile (median, 56 mcg/day) (44). Osteoporotic fractures are often linked to a reduction in bone mineralization. Yet, the investigators found no association between dietary phylloquinone intake and bone mineral density (BMD) in the Framingham subjects (44). While other studies failed to observe associations between dietary phylloquinone intake and measures of bone strength, BMD, or fracture incidence (45, 46), the cross-sectional study of a cohort of 3,199 middle-aged women found that subjects in the highest quartile of dietary phylloquinone intake had significantly greater hip and lumbar spine BMD than those in the lowest quartile (162 mcg/day vs. 59 mcg/day) (47). Moreover, recent cross-sectional and case-control studies have reported associations between higher phylloquinone intakes and lower incidence of hip fracture (48, 49).
However, because green leafy vegetables are the primary dietary source of phylloquinone and because they are usually part of a balanced diet, high phylloquinone consumption may be just an indicator of healthy eating habits which may, rather than phylloquinone itself, account for all or part of the association reported in observational studies (50). The few studies that measured plasma phylloquinone generally found that higher circulating levels were associated with lower fracture risk (17, 51). For example, the incidence of vertebral fractures was inversely correlated with lumbar BMD and plasma phylloquinone in a four-year prospective study that included 379 Japanese women ages 30-88 years (51). Yet, observational studies are not designed to making causal inferences, and only randomized controlled trials can evaluate whether phylloquinone may have beneficial effects on bone health (see Vitamin K supplementation studies and osteoporosis).
Vitamin K2: There are few studies on associations between menaquinones (vitamin K2) and bone health, perhaps because of the limited number of dietary sources of menaquinone-4 (MK-4), the main form of vitamin K2 present in Western diets. The Japanese food natto, made of cooked soybeans fermented by bacillus subtilis natto, is rich in MK-7. In a prospective study that followed 944 Japanese women (ages 20-79 years), total hip BMD at baseline was positively associated with natto intake in postmenopausal women (52). During the three-year follow-up period, the rate of BMD loss at the femoral neck was significantly lower in women consuming natto (>200 mcg/day of MK-7) compared to non-consumers. No association was found between natto intake and BMD in premenopausal women (52).
Total hip and femoral neck BMD was also reportedly higher in nearly 2,000 Japanese men aged 65 years and older who regularly consumed at least of one pack per day of natto (≥350 mcg/day of MK-7) compared to those consuming less than one pack per week (<50mcg/day of MK-7) (53). Yet, increasing natto consumption also maximizes the intake of other dietary compounds (e.g., soy isoflavones) that have potential benefits for skeletal health; thus, there is need to find reliable measures of vitamin K status. To date, observational studies have failed to unequivocally support an association between circulating menaquinone (MK-7 and MK-4) levels and fracture risk (17, 54).
Biomarker of vitamin K status and bone health
Total circulating levels of the bone protein, osteocalcin, have been shown to be sensitive markers of bone formation. Several hormones and growth factors, including vitamin D but not vitamin K, regulate osteocalcin synthesis by osteoblasts. However, vitamin K is an essential cofactor for the gamma-carboxylation of three glutamic acid residues in osteocalcin. Undercarboxylation of osteocalcin in human bone and serum has been linked to poor vitamin K status. The degree of osteocalcin gamma-carboxylation is responsive to vitamin K nutritional interventions, and thus is used as a relative indicator of vitamin K status (13).
Circulating levels of undercarboxylated osteocalcin (ucOC) were found to be higher in postmenopausal women than premenopausal women and markedly higher in women over the age of 70. Also, high ratios of ucOC to total OC (ucOC/OC) appear to be predictive of hip fracture risk in elderly women (55, 56). Although vitamin K deficiency would seem the most likely cause of elevated blood ucOC/OC ratio, some investigators have documented an inverse relationship between biochemical measures of vitamin D nutritional status and ucOC levels, as well as a significant lowering of ucOC/OC ratio by vitamin D supplementation (57). It has been suggested that increased circulating ucOC/OC ratio could reflect a poor overall nutritional status that would include vitamin D inadequacy, which would explain the above-mentioned observations (58). However, in several randomized, placebo-controlled intervention studies conducted in young girls (58, 59) and postmenopausal women (60), vitamin D supplementation failed to decrease ucOC/OC ratios or show any additive effect on ucOC/OC lowering by supplemental vitamin K.
Vitamin K1 supplementation: The systematic review of five randomized clinical trials that assessed the effect of phylloquinone (vitamin K1) supplementation on hip BMD using doses ranging from 200 mcg/day to 5,000 mcg/day for durations of 12 to 36 months found little promising benefit for bone health (17). Although supplementation with phylloquinone decreased ucOC levels in all five studies, only one study reported an effect of supplemental phylloquinone on BMD (61). In this study, 150 postmenopausal women were randomized to receive a placebo, minerals (500 mg/day of calcium, 130 mg/day of magnesium, and 10 mg/day of zinc) plus vitamin D (320 IU/day), or minerals, vitamin D and phylloquinone (1,000 mg/day). The rate of BMD loss at the femoral neck, but not at the lumbar spine, was significantly lower in subjects with supplemental phylloquinone compared to the other two groups. Thus, evidence of a putative benefit of phylloquinone on bone health in older adults is considered weak. None of the studies were designed to assess the effect of phylloquinone on osteoporotic-related fractures. Further investigation may seek to evaluate whether phylloquinone supplementation could improve skeletal health in subjects at high-risk for vitamin K inadequacy (e.g., individuals with malabsorption syndromes or cystic fibrosis).
Vitamin K2 supplementation: Pharmacological doses of menaquinone-4 (MK-4; brand name, menatetrenone) are currently used in Japan in the treatment of osteoporosis. Accordingly, most intervention trials investigating the effect of high-dose MK-4 on bone loss have been conducted in Japanese postmenopausal women. A 2006 meta-analysis of seven randomized controlled trials associated MK-4 supplementation with increased BMD and reduced fracture incidence (62). All but one of the seven individual trials employed 45 mg of MK-4 daily; the other trial used 15 mg/day (62). This meta-analysis reported that MK-4 supplementation for more than six months significantly lowered risk for vertebral fractures by 60%, hip fractures by 77%, and nonvertebral fractures by 81%. However, the results of this meta-analysis were later downplayed because of the small size of the included studies and the fact that some of them were not placebo-controlled but instead used a concurrent or open-label treatment (e.g., with calcium and vitamin D). Additionally, the analysis did not include data from an unpublished study with a larger sample size that reported no MK-4 effect on fracture risk and would have altered the conclusion of the meta-analysis (63).
A more recent non-placebo controlled study randomized over 4,000 postmenopausal Japanese women to receive calcium alone or in combination with MK-4 (45 mg/day) for three years. At the end of an additional follow-up year (four years total), there were no differences between groups regarding the incidence of vertebral fractures, and only a small reduction in the incidence of new clinical fractures was seen in those taking the combined treatment compared to calcium alone (4.4% vs. 3.4%) but only in women at high-risk of fractures (64). Equivocal results have been reported in additional trials conducted in Europe and the US. A three-year placebo-controlled intervention trial in 325 healthy postmenopausal women found that supplemental MK-4 (45 mg/day) for three years improved measures of bone strength compared to placebo (65). Of note, this MK-4 dose used in most of the cited studies is about 500 times higher than the AI for vitamin K. Another one-year, randomized, double-blind, placebo-controlled trial in 365 healthy American postmenopausal women with vitamin K inadequacy (undercarboxylated osteocalcin ≥4%) found that neither supplemental high-dose phylloquinone (1,000 mcg/day) nor MK-4 (45 mg/day) had an effect on serum markers of bone turnover or on BMD (lumbar spine and hip) when compared to placebo (66). In this study, all the subjects also received daily, open-label calcium (630 mg) and vitamin D3 (400 IU).
Although a few observational studies have suggested a link between natto (rich in MK-7) consumption and bone health, a recent randomized, double-blind, placebo-controlled study in 334 healthy postmenopausal women (1 to 5 years postmenopause) found no effect of 360 mcg/day of MK-7 (in the form of natto capsules) on BMD at various sites after one year compared to baseline (67). Another comparable placebo-controlled trial in 244 menopausal women found that supplementation with180 mcg/day of MK-7 for three years significantly limited bone loss at the femoral neck but not at other sites (68). At present, the potential role for supplemental menaquinones on bone health still needs to be established in large, randomized, and well-controlled trials.
Vitamin K antagonists and bone health
Certain oral anticoagulants, such as warfarin, are known to be antagonists of vitamin K (see Coagulation). Few studies have examined chronic use of warfarin and risk of fracture in older women. One study reported no association between long-term warfarin treatment and fracture risk (69), while another one found a significantly higher risk of rib and vertebral fractures in warfarin users compared to nonusers (70). Additionally, a study in elderly patients with atrial fibrillation reported that long-term warfarin treatment was associated with a significantly higher risk of osteoporotic fracture in men but not in women (71). A meta-analysis of the results of 11 published studies found that oral anticoagulation therapy was associated with a very modest reduction in BMD at the wrist and no change in BMD at the hip or spine (72). The development of new anticoagulants that do not block vitamin K recycling may offer a safer alternative to the use of vitamin K antagonists (73).
An inverse relationship between vitamin K intake and mortality was reported in a US national survey (NHANES III) of 3,401 participants (74). Adequate vs. inadequate vitamin K intakes (based on sex-specific AI: 90 mcg/day for women and 120 mcg/day for men) were associated with a 22% lower risk of cardiovascular disease (CVD)-related mortality and a 15% lower risk of all-cause mortality. The report also indicated that, while over two-thirds of individuals with chronic kidney disease had vitamin K intakes below the AI, the risk of CVD mortality was 41% lower in those with adequate compared to suboptimal intakes (74). However, higher vitamin K intakes were not associated with lower CVD mortality in a prospective study that followed 7,216 older adults at risk for developing CVD (75). This study associated higher phylloquinone intakes, but not menaquinones, with lower risk of all-cause mortality.
In another recent prospective study, which followed 35,476 healthy Dutch men and women for a mean period of 12.1 years, the risk of incident stroke was not significantly associated with intakes of phylloquinone or menaquinones (76). Earlier observational studies offer limited support to an inverse relationship between phylloquinone intake and risk of CVD, despite high intakes being sometimes regarded as a marker of healthy dietary habits associated with low cardiovascular risk (reviewed in 77). A prospective cohort of 16,057 Dutch women (ages 49-70 years) followed for a mean period of 8.1 years found a 9% reduction in risk for coronary heart disease (CHD) per each incremental 10 mcg/day increase in menaquinone intake (78). In another earlier Dutch study that examined 4,807 healthy men and women 55 years and older, participants in the highest tertile of menaquinone intake (>32.7 mcg/day) had a 41% lower risk of incident CHD and a 26% lower risk of all-cause mortality than those in the lowest tertile (<21.6 mcg/day) (79). In addition, menaquinone intake was found to be inversely associated with aortic calcification, a major risk factor for CVD (79).
One of the hallmarks of cardiovascular disease is the presence of atherosclerotic plaques in arterial walls. Plaque rupture that causes blood clot formation (thrombogenesis) is the usual cause of a myocardial infarction (heart attack) or stroke. While calcification of the plaques occurs as the atherosclerosis progresses, it is unclear whether calcification increases plaque instability and could predict risk of rupture and thrombogenesis (80). However, calcification may be predictive of future cardiovascular events. A meta-analysis of 30 prospective cohort studies, including a total of 218,080 participants, found that the presence of vascular calcification was associated with an overall three- to four-fold increased risk of cardiovascular events and mortality (81). An early population-based study of postmenopausal women (ages, 60-79 years) observed that the younger women (60-69 years) with aortic calcifications had lower vitamin K intakes than those without aortic calcifications, but this was not true for older women (70-79 years) (82). A prospective cohort study in 807 men and women, 39-45 years of age, did not find any correlation between dietary phylloquinone intake and coronary artery calcification, as measured non invasively by computed tomography (83). Additionally, neither phylloquinone nor menaquinone intakes were associated with calcification of breast arteries in a cross-sectional study of 1,689 women ages 49-70 years (84). However, in another cross-sectional study, the upper vs. lowest quartile of menaquinone (MK-4 to MK-10) intake (median intakes, 48.5 mcg/day vs.18 mcg/day) was found to be associated with a 20% reduced prevalence of coronary artery calcification in 564 postmenopausal women (85).
Recent research has uncovered possible mechanisms by which vitamin K may inhibit mineralization (calcification) of vessels while promoting bone mineralization. The potential mechanisms, although not yet fully understood, implicate vitamin K-dependent proteins, including matrix Gla protein (MGP) and the newly described Gla-rich protein (GRP). Secreted by various cell types, such as vascular smooth muscle cells (VSMCs) in arterial vessel walls, MGP appears to be important for the prevention of calcification of soft tissues, including cartilage, vasculature, skin, and trabecular meshwork cells in the eye (17, 86). In MGP knockout mice, conversion of VSMCs into bone-like cells and extensive vessel calcification results in large vessel rupture and premature death. In humans, defective MGP gene has been linked to Keutel syndrome, a rare inherited condition characterized in particular by abnormal cartilage calcification and pulmonary artery stenosis (narrowing). Calcification prevention by MGP involved several mechanisms, including the binding to calcium crystals and the inhibition of proteins (bone morphogenic proteins; BMPs) known to promote ectopic bone formation (reviewed in 87).
Calcium-binding activity of MGP is regulated by two types of modifications (known as post-translational modifications since they take place after protein synthesis): the vitamin K-dependent carboxylation of up to five Glu residues and the phosphorylation of serine residues. A variation in the sequence (polymorphism) of the gene for MGP leading to a threonine-to-alanine transition in one of the five Gla domains of the protein may possibly prevent carboxylation and elicit a change in MGP ability to bind calcium. This polymorphism, known as MGPThr83Ala, has been associated with the progression of coronary artery calcification over a mean period of 10.6 years in a community-based prospective study that followed 605 middle-aged men and women (88). This association was only observed among participants without detectable calcification at baseline and not in those who had baseline calcification (88). Interestingly, MGPThr83Ala was also associated with higher risk of myocardial infarction and femoral artery calcification in carriers of the genotype (89).
Additionally, a small study initially found that, while undercarboxylated MGP (ucMGP) was absent from the innermost lining of the carotid arteries in healthy subjects, the majority of MGP in the carotid arterial lining of patients with atherosclerosis was undercarboxylated (90). In another study that examined the association between circulating MGP and incident cardiovascular events in 577 older men and women followed for a mean period of 5.6 years, the risk of cardiovascular disease (i.e., coronary artery disease, peripheral arterial disease, and cerebrovascular disease) was two- to three-fold greater in subjects in the highest vs. lowest tertile of plasma dephosphorylated and undercarboxylated MGP (dp-ucMGP) (91). The results of another prospective study suggested that circulating dp-ucMGP may be predictive of mortality risk in subjects with overt vascular disease (92). Indeed, the risk of cardiovascular-related and all-cause mortality was found to be nearly doubled in subjects with coronary artery disease or stroke in the highest vs. lowest quartile of dp-ucMGP concentrations (92).
Because suboptimal vitamin K nutritional status may limit carboxylation and result in biologically inactive ucMGP, it has been speculated that vitamin K supplementation may protect against vascular calcification. A three-year, double-blind, controlled trial investigated the potential effect of vitamin K on the progression of coronary calcification in 401 older, community-dwelling adults (ages, 60-80 years) free of CVD at baseline (93). The participants were randomized to receive a daily multivitamin plus calcium and vitamin D with or without 500 mg of phylloquinone. Using measurements of coronary artery calcification at baseline and follow up, it was found that phylloquinone supplementation was able to limit the progression of vascular calcification and reduce plasma dp-ucMGP compared to control (93, 94). Although circulating dp-ucMGP was correlated to various markers of vitamin K status, no association with measures of coronary artery calcification were found (94).
Further investigations are necessary to examine the role of other vitamin K-dependent proteins (e.g., GRP, periostin, Gas6) in human atherosclerotic plaque calcification and to evaluate the effect of supplemental vitamin K on the progression of vascular calcification and CVD risk.
Several cross-sectional studies have reported increased vascular calcium scores (a means to quantify vascular calcification) in chronic users of vitamin K antagonists compared to nonusers (reviewed in 95). Warfarin therapy has also been associated with higher circulating concentrations of dp-ucMGP in a recent prospective study that examined vascular calcification in subjects with CVD (92). Newly developed direct inhibitors of coagulation factors that do not interfere with VKDP activity may be more suitable than vitamin K antagonists, especially with regards to vascular calcification (95).
Nutrition surveys in Europe and the US have shown that mean dietary intakes of vitamin K (all forms) vary greatly among individuals and populations, with values ranging from 60 to 200 mcg/day (96).
Vitamin K1: Phylloquinone (vitamin K1) is the major dietary form of vitamin K in most diets. Green leafy vegetables and some plant oils (soybean, canola, olive, and cottonseed) are major contributors of dietary vitamin K. However, phylloquinone bioavailability from green vegetables is lower than in oil and supplements. Also, the phylloquinone content of green vegetables depends on their content in chlorophyll (green pigment), so that outer leaves have more phylloquinone than inner leaves. The efficiency of phylloquinone intestinal absorption varies among plant sources and is increased with the addition of a fat source to a meal. Finally, the hydrogenation of vegetable oils may decrease the absorption and biological effect of dietary phylloquinone (reviewed in 2, 9). If you wish to check foods for their nutrient content, including phylloquinone, search the USDA food composition database. A number of phylloquinone-rich foods are listed in the table below along, with their content in phylloquinone expressed in micrograms (mcg).
|Kale, raw||1 cup (chopped)||472|
|Swiss chard, raw||1 cup||299|
|Parsley, raw||¼ cup||246|
|Broccoli, cooked||1 cup (chopped)||220|
|Spinach, raw||1 cup||145|
|Watercress, raw||1 cup (chopped)||85|
|Leaf lettuce (green), raw||1 cup (shredded)||46|
|Soybean oil||1 Tablespoon||25|
|Canola oil||1 Tablespoon||10|
|Olive oil||1 Tablespoon||8|
|Cottonseed oil||1 Tablespoon||3|
Vitamin K2: Menaquinones (vitamin K2) are primarily of microbial origins and thus commonly found in fermented foods, such as cheese, curds, and natto (fermented soybeans). Another source of long-chain menaquinones (MK-7 to MK-13) is animal livers (9). Because of the limited availability of food composition tables for menaquinones, their contribution to total vitamin K intakes is difficult to estimate and likely to vary between populations with different food consumption practices (2). Bacteria that normally colonize the large intestine (colon) can synthesize menaquinones. It was initially thought that up to 50% of the human vitamin K requirement might be met by bacterial synthesis. However, all forms of vitamin K are absorbed in the small intestine via a mechanism requiring bile salts, while most of the menaquinone production takes place in the colon where bile salts are lacking. Current research suggests that the contribution of bacterial synthesis is much less than previously thought, although the exact contribution remains unclear (97). Among menaquinones, MK-4 is formed from menadione (synthetic vitamin K form) found in animal feeds or is converted in a tissue-specific way from dietary phylloquinone; thus, it is the only menaquinone that is not produced by bacteria (4). Longer-chain menaquinones are found in limited fermented food products. The Japanese fermented soybean-based, natto, is rich in MK-7 (998 mcg/100 g) and also contains MK-8 (84 mcg/100 g). Some cheeses contain MK-8 and MK-9 (2).
In the US, both phylloquinone and menaquinones are available without a prescription in multivitamin and other dietary supplements in doses that generally range from 25-100 mcg per tablet (98). While menaquinone-4 (MK-4) is marketed for osteoporosis treatment in Japan, the US Food and Drug Administration (FDA) does not currently allow any forms of vitamin K to be used for the prevention or treatment of osteoporosis.
Although allergic reaction is possible, there is no known toxicity associated with high doses (dietary or supplemental) of the phylloquinone (vitamin K1) or menaquinone (vitamin K2) forms of vitamin K (42). The same is not true for synthetic menadione (vitamin K3) and its derivatives. Menadione can interfere with the function of glutathione, one of the body's natural antioxidants, resulting in oxidative damage to cell membranes. Menadione given by injection has induced liver toxicity, jaundice, and hemolytic anemia (due to the rupture of red blood cells) in infants; therefore, menadione is no longer used for treatment of vitamin K deficiency (5). No tolerable upper intake level (UL) has been established for vitamin K (42).
Large doses of vitamin A and vitamin E have been found to antagonize vitamin K (8). Excess vitamin A appears to interfere with vitamin K absorption, whereas vitamin E may inhibit vitamin K-dependent carboxylase activity and interfere with the coagulation cascade (99). One study in adults with normal coagulation status found that supplementation with 1,000 IU/day of vitamin E for 12 weeks decreased gamma-carboxylation of prothrombin, a vitamin K-dependent protein (100). Individuals taking anticoagulatory drugs like warfarin and those who are vitamin K deficient should not take vitamin E supplements without close medical supervision because of the increased risk of hemorrhage (excessive bleeding) (101).
The anticoagulant effect of vitamin K antagonists (e.g., warfarin) may be compromised by very high dietary or supplemental vitamin K intake. Moreover, daily phylloquinone supplements of up to 100 mcg are considered safe for subjects taking warfarin, but therapeutic anticoagulant stability may be undermined by doses of MK-7 as low as 10-20 mcg (102). It is generally recommended that individuals using warfarin try to consume the AI for vitamin K (90-120 mcg) and avoid large fluctuations in vitamin K intake that might interfere with the adjustment of their anticoagulant dose (9). The prescription of anti-vitamin K anticoagulants, anticonvulsants (e.g., phenytoin), and anti-tuberculosis drugs (e.g., rifampin and isoniazid) to pregnant or breast-feeding women may place the newborn at increased risk of vitamin K deficiency (103).
Prolonged use of broad spectrum antibiotics, such as cephalosporins and salicylates, can interfere with vitamin K synthesis by intestinal bacteria and lower vitamin K absorption. The drug amiodarone, used in the management of certain cardiac arrhythmias (irregular heartbeat), including atrial fibrillation, can enhance the anticoagulant effect of warfarin and thus increase the risk of hemorrhage (104, 105). Further, the use of cholesterol-lowering medications (like cholestyramine and cholestipol), as well as orlistat, mineral oil, and the fat substitute, olestra, which interfere with fat absorption and may affect the absorption of fat-soluble vitamins, including vitamin K (106).
It is not clear whether the AI for vitamin K is enough to optimize the gamma-carboxylation of vitamin K-dependent proteins in bone (see Osteoporosis). To consume the amount of vitamin K associated with a decreased risk of hip fracture in the Framingham Heart Study (about 250 mcg/day) (44), an individual would need to eat a little more than ½ cup of chopped broccoli or a large salad of mixed greens every day. Though the dietary intake of vitamin K required for optimal function of all vitamin K-dependent proteins is not yet known, the Linus Pauling Institute recommends taking a multivitamin/mineral supplement and eating at least one cup of dark green leafy vegetables daily. Replacing dietary saturated fats like butter and cheese with monounsaturated fats found in olive oil and canola oil will increase dietary vitamin K intake and may decrease the risk of cardiovascular disease.
Older adults (> 50 years)
Because older adults are at increased risk of osteoporosis and hip fracture, the above recommendation for a multivitamin/mineral supplement and at least one cup of dark green leafy vegetables daily is especially relevant.
Written in May 2004 by:
Jane Higdon, Ph.D.
Linus Pauling Institute
Oregon State University
Updated in July 2014 by:
Barbara Delage, Ph.D.
Linus Pauling Institute
Oregon State University
Reviewed in August 2014 by:
Sarah L. Booth, Ph.D.
Director, Vitamin K Research Program
Jean Mayer USDA Human Nutrition Research Center on Aging
Copyright 2000-2014 Linus Pauling Institute
The Linus Pauling Institute Micronutrient Information Center provides scientific information on the health aspects of dietary factors and supplements, foods, and beverages for the general public. The information is made available with the understanding that the author and publisher are not providing medical, psychological, or nutritional counseling services on this site. The information should not be used in place of a consultation with a competent health care or nutrition professional.
The information on dietary factors and supplements, foods, and beverages contained on this Web site does not cover all possible uses, actions, precautions, side effects, and interactions. It is not intended as nutritional or medical advice for individual problems. Liability for individual actions or omissions based upon the contents of this site is expressly disclaimed.
Thank you for subscribing to the Linus Pauling Institute's Research Newsletter.
You should receive your first issue within a month. We appreciate your interest in our work.