Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2018 Jul 1.
Published in final edited form as: Toxicol Pathol. 2017 Jun 15;45(5):614–623. doi: 10.1177/0192623317714343

A black cohosh extract causes hematologic and biochemical changes consistent with a functional cobalamin deficiency in female B6C3F1/N mice

Michelle C Cora a, William Gwinn a, Ralph Wilson a, Debra King a, Suramya Waidyanatha a, Grace E Kissling b, Sukhdev S Brar a, Dorian Olivera c, Chad Blystone a, Greg Travlos a
PMCID: PMC5544593  NIHMSID: NIHMS877917  PMID: 28618975

Abstract

Available as a dietary supplement, black cohosh rhizome is most commonly marketed as a remedy for dysmenorrhea and menopausal symptoms. A previous subchronic toxicity study of black cohosh dried ethanolic extract (BCE) in female mice revealed a dose-dependent ineffective erythropoiesis with a macrocytosis consistent with the condition known as megaloblastic anemia. The purpose of this study was to investigate potential mechanisms by which BCE induces these particular hematological changes. B6C3F1/N female mice (32/group) were exposed by gavage to vehicle or 1000 mg/kg BCE for 92 days. Blood samples were analyzed for hematology, renal and hepatic clinical chemistry, serum folate and cobalamin, RBC folate, and plasma homocysteine and methylmalonic acid (MMA). Folate levels were measured in liver and kidney. Hematological changes included: decreased RBC count, increased MCV, and decreased reticulocyte, white blood cell, neutrophil and lymphocyte counts. Blood smear evaluation revealed increased Howell-Jolly bodies and occasional basophilic stippling in treated animals. Plasma homocysteine and MMA concentrations were increased in treated animals. Under the conditions of our study, BCE administration caused hematological and clinical chemistry changes consistent with a functional cobalamin, and possibly folate, deficiency. Further studies are needed to elucidate the mechanism by which BCE causes increases in homocysteine and MMA.

Keywords: megaloblastic, homocysteine, methylmalonic acid, cobalamin, folate

Introduction

Black Cohosh (Actaea racemosa; previously Cimicifuga racemosa) is an indigenous North American plant found in the woodlands of eastern United States up through southern Ontario. The rhizome of black cohosh has a history of being used as a remedy for rheumatism, neuralgias, dysmenorrhea, amenorrhea and painful labor (American Herbal Pharmacopoeia (AHP) 2002; Blumenthal 2003). In recent years, black cohosh has been marketed as a remedy for dysmenorrhea and for menopausal symptoms including hot flashes and night sweats (AHP 2002; Borrelli and Ernst 2008; Leach and Moore 2012). Black cohosh is most commonly sold as a dried ethanol extract in tablet or capsule form with a range of recommended daily doses, and is sometimes formulated in combination with other medicinal herbs. Black cohosh was nominated to the National Toxicology Program (NTP) for general toxicity testing by both the National Cancer Institute and National Institute of Environmental Health Sciences due to its widespread use and lack animal toxicity studies in the published literature. Because black cohosh is almost exclusively used by women, a 90-day (subchronic) study was conducted on female mice and rats only (i.e., not males) using a black cohosh extract (BCE) to evaluate general toxicity in addition to select developmental reproductive endpoints. Results of the mouse subchronic toxicity study revealed a dose-dependent decrease in red blood cell (RBC) count, hemoglobin concentration, and hematocrit with no change in the reticulocyte count and an increase in the mean corpuscular volume (MCV) (Mercado-Feliciano et al. 2012). Similar changes were observed in the rats. These particular changes indicated an ineffective erythropoiesis consistent with a condition known as megaloblastic anemia.

Despite its given name, megaloblastic anemia is a dyshematopoietic condition that affects all cell lines and is due to disruptions in DNA synthesis during hematopoiesis; however, abnormalities of the erythroid series are usually the most predominant feature. Within the erythroid series DNA disruptions lead to increased rates of erythroid precursor cell death (apoptosis), resulting in an ineffective erythropoiesis (Koury et al. 1997; Koury et al. 2000). Red blood cells surviving to maturity are larger in size (macrocytosis) due to asynchronous maturation. Megaloblastic anemias result from inherited (e.g., orotic aciduria) or drug-induced (e.g., pyrimidine antagonists, anti-viral) disorders of DNA synthesis, or are a consequence of a folate or cobalamin (vitamin B12) deficiency (Koury 2014; Means and Glader 2014). Deficiencies in folate and cobalamin cause a megaloblastic anemia because they are both essential for several reactions involved in DNA synthesis. Causes of folate and cobalamin deficiencies are many and include inadequate intake, changes in bacterial gut flora, altered enterobiliary secretions (e.g., decreased intrinsic factor, hypochlohydria), interference with their intestinal uptake (e.g., colchicine) or inhibition of enzymes involved in their normal metabolism (e.g., methotrexate) (Camel 2014; Hesdorffer and Longo 2015; Stabler 2013; Waxman et al. 1970). The purpose of this study was to investigate a functional cobalamin or folate deficiency as a potential explanation for the hematological changes observed in the NTP BCE subchronic mouse toxicity study. In order to determine if BCE induces a deficiency, several analyses were performed including serum folate and cobalamin levels, plasma homocysteine and methyl-malonic acid levels, and select tissue folate levels.

Material and Methods

BCE chemical characterization and dose formulation

Test article procurement and characterization and dose formulation development studies are described elsewhere (Mercado-Feliciano et al. 2012). Dose formulations of BCE were prepared in 0.5% aqueous methylcellulose at 100 mg/ml. Formulations were analyzed to determine BCE concentration with isoferulic acid as the marker compound. All BCE formulations were within 10% of the target concentration. Formulations were stored refrigerated (~ 5°C) and used within the 43-day stability period.

Animals and animal husbandry

Sixty-four, 4-week old female B6C3F1/N mice were obtained from Taconic Biosciences (Germantown, NY). Upon receipt at the Alion animal facility (Research Triangle Park, NC) the mice were quarantined for 12 or 13 days depending on study start. Mice were grouped housed up to 5 animals per cage. Animal rooms were maintained under a 12:12 light:dark light cycle. NTP-2000 pelleted feed (Zeigler Brothers, Inc., Gardners, PA) and tap water were provided ad libitum. Animal use was in accordance with the Guide for the Care and Use of Laboratory Animals (Institute for Lab Animal Research 2011) in a facility accredited by the Association for Assessment and Accreditation of Laboratory Animal Care. All procedures were approved by Alion Science and Technology Animal Care and Use Committee.

Study Design

Mice were exposed daily to vehicle (0.5% methylcellulose) (n=32) or 1000 mg/kg BCE (n=32) by gavage in a volume of 10 mL/kg for 92 days. The 1000 mg/kg dose was chosen, as it was the high dose used in the NTP subchronic mouse study and was selected to maximize the hematological effect for mechanistic evaluation. Animal weights were obtained weekly and dosing volume was adjusted according to the most recent body weight. The final dose was given the day prior to terminal sacrifice.

Within each treatment group, the mice were randomly divided into three cohorts with three different blood sample collection methods at study termination. The cohorts were as follows: animals for collection of blood into tubes containing ethylenediaminetetraacetic acid (EDTA) (n = 10/ group); animals for the collection of blood into tubes devoid of anticoagulant (n =10/ group); and animals for collection of blood into tubes containing heparin (n = 12/ group). To ensure blood collection within an appropriate time window for the processing of the blood samples (i.e., to minimize sources of pre-analytical variation), terminal blood collection was performed on two consecutive days. Before study start, the same numbers of mice from each group within the different cohorts were randomly selected for one of two consecutive study termination days (i.e., 32 mice were bled on each of the two days); treatment length was equal due to staggered start dates. At study termination, blood was collected from the retro-orbital sinus under isoflurane anesthesia and placed into the appropriate blood tubes.

Serum folate and cobalamin levels; kidney and liver clinical chemistry

Blood collected in tubes void of an anticoagulant were allowed to clot for at least 30 minutes and then centrifuged. The serum was then harvested and frozen at −80°C until analysis. Serum folate (as 5-methyltetrahydrofolate) and cobalamin (as cyanocobalamin) concentrations were measured using a MP biomedical SimulTRAC-SNB radioassay kit for vitamin B12 (57CO)/folate (125I) (MP biomedical, Santa Ana, CA) and performed according to the manufacture’s specifications using an Apex automatic gamma counter (ICN Micromedic Systems, Inc. Huntsville, AL). Preliminary analyses of murine sera were done prior to the study. Normal baseline murine folate and cobalamin serum levels are 10 to 100 times higher than humans as measured by radioimmunoassay (Bottiglieri et al. 2012; Cabelof et al. 2004; Leamon et al. 2008; Troen et al. 2008), thus analyses of serum dilutions were performed. Two levels of control material were also analyzed. The intra- and inter-assay coefficient of variation (CV) for folate was 7% and 7.6%, respectively, and for cobalamin 2.1% and 8.2%, respectively. Based on dilution studies and because the highest supplied standard was 2000 pg/ml and 20 ng/ml for cobalamin and folate, respectively, the serum samples from the BCE study were diluted to 1:21 for analysis. The following clinical chemistry parameters were analyzed on an Olympus AU400e chemistry analyzer, Olympus America, Inc. (Irvin, TX) using reagents obtained from Beckman Coulter (Brea, CA): creatinine, alanine aminotransferase, aspartate aminotransferase and alkaline phosphatase. All serum samples’ analyses were run in one batch on their respective analyzers.

Hematology and RBC folate levels

Blood collected into the EDTA tubes was used for the analysis of CBC and RBC folate concentrations. The following CBC parameters were analyzed using an ADVIA 2120i hematology analyzer (Siemens, Malvem, PA) within two hours of blood collection: RBC count, hemoglobin concentration, hematocrit, MCV, mean corpuscular hemoglobin (MCH), mean corpuscular hemoglobin concentration (MCHC), reticulocyte count, white blood cell (WBC) count and differential, and platelet count. Blood smears were prepared and stained with Wright-Giemsa stain. Blood smears were examined microscopically to document any abnormal erythrocyte or leukocyte morphologies (performed by M.C.C.). Since Howell-Jolly (HJ) body numbers were increased in the initial NTP subchronic study, HJ body numbers were evaluated semiquantitatively as follows: within the monolayer of the blood smears, the number of HJ bodies per ten, 50X oil-objective fields (HPF) was recorded for each animal.

After completion of the CBC analysis, the EDTA whole blood was assayed for RBC folate using a SimulTRAC-SNB radioassay kit for vitamin B12 (57CO)/folate (125I) (see above for more details). A RBC hemolysate was prepared according to the manufacturer’s instructions using a 0.2% ascorbic acid solution and then frozen at − 80°C until analysis. RBC folate (as 5-methyltetrahydrofolate) levels were measured according to manufacture’s specifications .

Total homocysteine and methyl-malonic acid (MMA) levels

Blood collected into the heparinized tubes was centrifuged and the plasma harvested. Total homocysteine was measured the same day as blood collection on an Olympus AU400e using a Diazyme Homocysteine 2 Reagent Enzymatic Assay kit (Poway, CA) and performed according to manufacturer’s specifications. Any remaining plasma was frozen at −80°C for MMA analysis.

Studies analyzing homocysteine levels in murine plasma using the Diazyme kit were performed prior to start of the BCE study. The intra- and inter-assay CVs were 2.3% and 1.8%, respectively. Results were within the reported linearity of the assay (up to 50 μmol/L) and similar to reported murine homocysteine plasma levels (Bottiglieri et al. 2012; Ernest et al. 2005; Troen et al. 2008).

For analysis of MMA, 25 μL of plasma was combined with the internal standard (MMA-d3), 25 μL of water and 250 μL of acetonitrile. The resulting supernatant was analyzed on a Waters Acquity Ultra Pressure Liquid Chromatograph (Milford, MA) equipped with a Sciex 4000 QTRAP mass spectrometer (Framingham, MA). The mass spectrometer was operated in negative electrospray ionization mode and the ions monitored were m/z 117→73 (MMA) and m/z 120→76 (MMA-d3). Chromatography was performed on a Merck SeQuant Zic-HILIC column (100 x 2.1 mm, 3.5 μm) (EMD Millipore, Billerica, MA) and isocratic conditions using 20% 100 mM ammonium acetate in water (pH 4.5) and 80% acetonitrile at a flow rate of 0.4 mL/min. The analytical method was qualified over the concentration range 20–2000 ng/mL . The method was linear (r ≥ 0.99), accurate (percent relative error ≤ ±20 % at the limit of quantitation and ≤ ±15 % for all other concentrations), and precise (percent relative standard deviation ≤ ±20 % at the limit of quantitation and ≤ ±15 % for all other concentrations). The limit of quantitation (LOQ) of the method was 11.0 ng/mL.

Dihydrofolate reductase assay

A dihydrofolate reductase (DHFR) assay kit (Sigma-Aldrich, Saint Louis, MS) was used to evaluate whether BCE inhibits DHFR enzyme. The assay was preformed according to manufacturer’s specifications. Briefly, the assay is based on the ability of DHFR enzyme to catalyze the reversible NADPH-dependent reduction of dihydrofolic acid to tetrahydrofolic acid by measuring the decrease in absorbance at 340nm with the use of a spectrophotometer (Beckman Coulter DU800; Brea, CA). Methotrexate was used as a positive control while the negative control was the complete reaction without the DHFR enzyme. Measurements were taken every 15 seconds over 30 minutes. Based on pilot studies to ascertain solubility, the BCE was tested at a concentration of 50mg/ml dissolved in 1% dimethyl sulfoxide (DMSO). Pilot studies showed that this concentration of DMSO did not inhibit DHFR enzyme activity.

Liver and kidney tissue folate levels

The left lateral liver lobe and the left kidney were collected from the cohort of animals that had blood collected into heparin tubes. The samples were cut into cubes, flash frozen and stored at −80°C until analysis. For analysis of liver and kidney tetrahydrofolate (THF) and 5-methyl-THF, homogenates were prepared with approximately 200 mg of each tissue and the equivalent buffer volume containing 50 mM phosphate buffer, pH 7, with 1% ascorbate and 0.1% β-mercaptopropanol. Fifty microliters of homogenate were transferred into boil-proof tubes (Axygen, Fisher Scientific) and combined with 50 μL of 5 – formyl THF (internal standard in buffer) and 250 μL buffer. The tubes were placed in a boiling water bath for ~ 1 min, then cooled on ice for ~ 5 min. The supernatant was passed through a1 0 kDa molecular weight cut-off membrane filter (Amicon, MilliporeSigma, Darmstadt, Germany) and the resulting supernatant was analyzed on a Waters Acquity Ultra Pressure Liquid Chromatograph (Milford, MA) equipped with a Sciex 4000 QTRAP mass spectrometer (Framingham, MA). The mass spectrometer was operated with positive electrospray ionization mode and the ions monitored were 460 → 313 for 5mTHF, 446 → 299 for THF, and 474 → 327 for internal standard. Chromatography was performed on a Waters Acquity UPLC HSS T3 column (2.1 × 100 mm, 1.8 μm) and isocratic conditions using 0.1% acetic acid in water and 0.1% acetic acid in acetonitrile at a flow rate of 0.3 mL/min. The analytical method was qualified over a concentration range of 5 to 5000 ng/mL for 5-methyl-THF and 50 to 50,000 ng/mL for THF.

Statistical methods

Prior to statistical analysis, data indicated as less than the limit of quantitation (LOQ) was replaced by one-half the LOQ. Because the testing for homocysteine and the CBC were run in two batches, batch-to-batch differences were tested by Mann-Whitney tests within each of the vehicle control group and the treated group. No differences were identified. All test data and the combined homocysteine and CBC data (from the two batches) were analyzed using one-sided Mann-Whitney tests to compare vehicle control to the treated group. Initial and final body weights and absolute and percent body weight gains were normally distributed with equal variances between the two groups, so two-sample t-tests were used to compare body weights and body weight gains. Statistical analyses were conducted using SAS 9.3 (SAS Institute, Cary, NC).

Results

Survival and body weights

There were no compound-related decreases in survival. Differences in body weight and body weight gains were statistically significant between the control and treated animals (P < .001). Average terminal body weight in the control mice (n = 30) was 30.5 g and for the treated mice (n = 29) was 26.8 g; the average terminal body weight of treated mice was 88% of that in the control group. The weight gained by the treated group was 66.3% of the weight gained by the control group.

CBC and blood smear evaluation

Hematology data are shown in Table 1. In the treated group, the RBC count (P < .001) and reticulocyte (P = .007) count were significantly lower, from the control group while the MCV and MCH were significantly higher in the treated group than in the control group (P < .001) and the MCHC was unchanged. Hemoglobin concentration and hematocrit values did not differ significantly between the treated groups and control groups. The WBC count (P < .001), neutrophil count (P = .004) and lymphocyte count (P < .001) were significantly lower in the treated group than in the control group.

Table 1.

Hematology parameters of B6C3F1/N female mice after gavage treatment with vehicle or 1000 mg/kg BCE for 92 days.

Parameter Control 1000 mg/kg Statistical Significance
Red Blood Cells, X 106/μL 9.46 (0.11) 8.53 (0.06) P < .001
Hemoglobin, g/dL 14.20 (0.16) 14.11 (0.09) NS
Hematocrit, % 43.2 (0.05) 43.0 (0.35) NS
MCV, fL 45.6 (0.2) 50.4 (0.23) P < .001
MCHC, g/dL 32.9 (0.1) 32.8 (0.1) NS
MCH, pg 15.0 (0.05) 16.6 (0.1) P < .001
Reticulocytes, X 103/μL 340 (20) 270 (10) P = .007
Platelets, X 103/μL 752.5 (37.2) 786.2 (25.7) NS
White blood cells, X 103/μL 3.53 (0.12) 2.29 (0.16) P < .001
Neutrophils, X 103/μL 0.56 (0.02) 0.44 (0.03) P = .004
Lymphocytes, X 103/μL 2.83 (0.13) 1.78 (0.13) P < .001
Open in a new tab

All data shown as mean (standard error). n = 10. MCV, mean corpuscular volume; MCHC, mean corpuscular hemoglobin concentration; MCH, mean corpuscular hemoglobin; NS, nonsignificant.

On blood smear evaluation, HJ bodies were more frequent (P < .0001) in the treated group, averaging more than twice as many per animal as in the control group (Figures 1 and 2). Rare to occasional (0–4 erythrocytes/10 HPFs) basophilic stippling was observed on the blood smears of the treated animals (Figure 3). A mild anisocytosis, as compared to concurrent controls, was noted on many of the treated animals’ blood smears. Leukocytes and platelets were observed to be morphologically unaffected.

Figure 1.

Figure 1

Open in a new tab

Mean number of Howell-Jolly (HJ) bodies observed per 10 high-power fields (HPF) on the peripheral blood smears of control mice and mice treated daily with 1000 mg/kg BCE for 92 d. Each bar represents mean ± SE. *P < .0001

Figure 2.

Figure 2

Open in a new tab

Howell-Jolly bodies were significantly increased in mice treated daily with 1000 mg/kg BCE for 92 d. Howell-Jolly bodies are present within 3 individual erythrocytes (arrows). Wright-Giemsa stain. Original magnification ×1000.

Figure 3.

Figure 3

Figure 3

Open in a new tab

Basophilic stippling was occasionally observed on the blood smears of mice treated daily with 1000 mg/kg BCE for 92 d. (A, B) Erythrocytes with basophilic stippling (arrows) observed on two different peripheral blood smears. Wright-Giemsa stain. Original magnification ×1000.

Folate and cobalamin levels and RBC folate levels

No significant differences were found between the control and treated group for serum folate and cobalamin concentrations and RBC folate concentrations (Table 2). Control levels of serum folate and cobalamin were similar to those reported in the literature (Bottiglieri et al. 2012; Cabelof et al. 2004; Leamon et al. 2008; Troen et al. 2008).

Table 2.

Clinical chemistry parameters of B6C3F1/N female mice after gavage treatment with vehicle or 1000 mg/kg BCE for 92 days.

Parameter Control 1000 mg/kg Statistical Significance
Folate, ng/mL 176.9 (21.1)* 222.8 (21.4)* NS
Cobalamin, pg/mL 27,018 (850)* 26,420 (672) NS
Red blood cell folate, ng/mL 1,688.8 (59.1) 1,729.2 (84) NS
Total Homocysteine, μmol/L 9.2 (0.4)† 12.0 (0.8) P = .008
Methylmalonic Acid, ng/mL 7.4 (1.3)† 23.9 (2.5) P < .001
Creatinine, mg/dL 0.1 (0)‡ 0.2 (0) NS
ALT, U/L 30 (6)* 21 (2) NS
AST, U/L 53 (13)‡ 40 (3) NS
ALP, U/L 79 (3)* 84 (2) NS
Open in a new tab

All data shown as mean (standard error). n = 10 except where footnoted. ALT, alanine aminotransferase; AST, aspartate aminotransferase; ALP, alkaline phosphatase; NS, nonsignificant.

*

n = 9.

†

n = 12.

‡

n = 8.

Total homocysteine and MMA levels, liver and kidney clinical chemistry

Results are listed in Table 2. Homocysteine (P = .008) and MMA (P < .001) were both significantly higher in the treated group with MMA levels 3.2 times higher in the treated group compared to the control. Homocysteine control levels were similar to those reported in the literature (Bottiglieri et al. 2012; Ernest et al. 2005; Troen et al. 2008). No significant differences were found between the control and treated group for serum liver and kidney biomarkers.

Liver and kidney tissue folate levels

Total folate tissue concentrations were calculated as the sum of the measured THF and 5-methyl-THF (Table 3). No significant differences were observed between control and treated groups’ levels of liver 5-methyl-THF, THF and total folate. A small but statistically significant (P = .02) higher level of kidney total folate was observed in the treated group; no significant difference was observed in kidney 5-methyl-THF and THF between the treated and control groups.

Table 3.

Liver and kidney tissue folate concentrations of B6C3F1/N female mice after gavage treatment with vehicle or 1000 mg/kg BCE for 92 d.

Control n = 12 1000 mg/kg n = 10 Statistical Significance
Liver
 5methyl-THF, ng/mg 1.9 (0.088) 1.86 (0.091) NS
 THF, ng/mg 21.08 (0.88) 19.11 (1.16) NS
 Total Folate,* ng/mg 22.98 (0.87) 20.97 (1.19) NS
Kidney
 5methyl-THF, ng/mg 1.84 (0.21) 2.39 (0.21) NS
 THF, ng/mg 5.09 (0.134) 5.46 (0.15) NS
 Total Folate,* ng/mg 6.92 (0.26) 7.85 (0.29) P = .02
Open in a new tab

All data shown as mean (standard error). THF, tetrathydrofolate; NS, nonsignificant.

*

Total folate = 5methyl-THF + THF

Dihydrofolate reductase assay

At 50 mg/ml, BCE did not inhibit DHFR; absorbance steadily decreased over 30 mins from a starting point of 40% down to 7% (data not shown). All controls performed as expected.

Discussion

Black cohosh extract is a widely available dietary supplement currently marketed for the alleviation of dysmenorrhea and menopausal symptoms, and is one of the top-selling herbal supplements (Smith et al. 2015). The current study was undertaken in an effort to begin to ascertain the mechanism by which BCE induced the hematological changes observed in the mice of the NTP subchronic toxicity study. This study confirmed the previous findings from the NTP study, in that exposure to BCE caused a decrease in the RBC count with an increase in MCV (macrocytosis) and MCH, as well as increases in HJ bodies in the peripheral blood. In addition, while folate and cobalamin levels were unchanged, two biomarkers that are used for the diagnosis or confirmation of folate and cobalamin deficiencies – homocysteine and MMA – were both elevated. The combination of these findings suggests a BCE-induced functional deficiency of cobalamin, and possibly folate.

Deficiencies in folate and cobalamin ultimately result in the development of a condition known as megaloblastic anemia, which, despite its name, is a disorder that can affect all three hematopoietic cell lines. Megaloblastic anemias are a result of deranged DNA synthesis. Folate, in the form of tetrahydrofolate (THF) coenzymes, is involved in several interrelated metabolic cycles (Figure 4; Carmel 2014; Shane 2013) related to DNA and RNA synthesis. These cycles include those required for the synthesis of thymidylate and purines (dGTP and dATP), precursors for DNA and RNA synthesis, and for the synthesis of methionine from homocysteine; the synthesis of methionine from homocysteine requires cobalamin as a cofactor. In addition, folate, through the methionine cycle, is also indirectly involved in the methylation of cytosines in DNA, among others (Shane 2013; Stover 2009). With cobalamin deficiency, 5-methyl-THF accumulates, while the other forms of THF needed for purine and thymidylate synthesis decrease. Thus, with either folate or cobalamin deficiency, DNA synthesis is ultimately impaired, which, in the case of the erythron, leads to cell death of many of the immature erythroid cells (Koury et al 1997). The decreased numbers of erythroid cells surviving to maturity leads to an anemia (i.e., ineffective erythropoiesis), while the surviving cells give rise to larger than normal erythrocytes, which is reflected in an increase in the MCV (Koury et al 1997). The increased cell size is due to an asynchronous maturation: during the delays in DNA replication and cell division that occur with these deficiencies, the erythroid precursors continue to synthesize proteins (mostly in the form of hemoglobin) resulting in an enlarged cell (Koury 2014; Means and Glader 2014). Additionally, in humans, giant band cells and metamyelocytes can be seen in the bone marrow and hypersegmented neutrophils observed in the peripheral blood (Carmel 2014).

Figure 4.

Figure 4

Open in a new tab

Principal components of folate and cobalamin metabolism as it relates to DNA synthesis, methylations, and the production of methionine from homocysteine. DHFR = dihydrofolate reductase, THF = tetrahydrofolate, TS = Thymidylate synthase, dUMP = deoxyuridine monophosphate, dUMT = deoxythymidine monophosphate, MTHFR = methylene tetrahydrofolate reductase, MS = methionine synthase, SAM = S-adenosylmethionine (also AdoMet), SAH = S717 adenosylhomocysteine (also AdoHcy).

In humans with deficiencies in folate or cobalamin the characteristic erythroid changes generally occur slowly over time as the deficiency worsens, with anemia (defined as a decrease in hemoglobin concentration) occurring as a late event. (Carmel 2014). There is a gradual rise in the MCH and then the MCV that will eventually rise outside of the normal reference interval. Weeks to months later the RBC counts and hemoglobin levels will begin to fall with the RBC count decreases usually preceding the fall in hemoglobin concentrations. As such, it is not unusual to observe a macrocytosis in the face of normal hemoglobin concentrations in less advanced deficiencies (Carmel 1979; Carmel 2014; Hall 1981; Herbert 1962). Additionally, decreases in WBCs and platelets may also occur culminating in a pancytopenia.

After approximately three months of daily treatment with BCE, the treated mice weighed on average 12% less than the control mice. These findings are consistent with other studies reporting on the effects of cobalamin or folate deficiencies in mice (Bills et al. 1992; Salojin et al. 2011; Troen et al. 2008). Treated mice had increases in the MCH and MCV with lower RBC counts and reticulocytes counts than control mice but the hemoglobin concentration did not differ, suggesting an earlier manifestation of the disorder. The WBC count, neutrophil count and lymphocyte count were also lower and considered treatment-related. On blood smear evaluation, the treated mice had more HJ bodies as compared to the control mice (Figures 1 and 2). In addition, rare to occasional erythrocytic basophilic stippling was observed on the blood smears of the treated mice (Figure 3); basophilic stippling represents ribosome aggregates and is associated with abnormal alterations in erythropoiesis (dyserythropoiesis). Howell-Jolly bodies, also known as micronuclei, are whole or fragmented chromosomes that lag behind during production and maturation of red blood cells. Howell-Jolly bodies are used as biomarkers of chromosomal damage (Everson et al 1988; Fenech 2012) and increases in HJ bodies are a non-specific feature of dyserythropoisesis. The increased frequency of HJ bodies observed on the peripheral blood smears of treated mice in this study supported the increased micronucleated RBCs (P < 0.0001; as measured by flow cytometry) observed in the treated mice of the BCE subchronic toxicity study (Mercado-Feliciano et al. 2012). Increased micronucleus formation is significantly correlated with deficiencies in both folate and cobalamin (Fenech 2012). Hypersegemented neutrophils, typical of folate and cobalamin deficiencies in humans, were not observed in this study. However, most murine neutrophil nuclei are twisted or ring-like with areas that may be folded onto themselves and usually, but not always, lack the distinct filamentous chromatin (as is typically seen in human neutrophils) that separates nuclear segments (Zhou et al. 2015). Thus, evaluating for neutrophilic hypersegementation in cases of dysmyelopoeisis (or myelodysplastic syndromes) in murine neutrophils can be challenging (Zhou et al. 2015).

While bone marrow smear evaluation was not done in this study, bone marrow smears were evaluated as part of a 3-month interim study of the NTP BCE mouse 2-year bioassay (report in preparation). Preliminary data indicate dysplastic changes in the metarubricytes, as well as a macrocytosis and decreased RBC count on the CBC. Dysplastic findings (identified by MCC) included low incidences of multilobulated nuclei, micronuclei, and nuclear to cytoplasmic asynchrony of the erythroid series as compared to concurrent controls.

Homocysteine and MMA are metabolites measured for the diagnosis, confirmation and differentiation of folate and cobalamin deficiencies (Savage et al. 1994; Klee 2000). In mammals, only two cobalamin-dependent enzyme reactions are known to exist (Shane 2013). As previously discussed, cobalamin (as methyl-cobalamin) is a cofactor for the methylation of homocysteine to methionine in which 5-methyl-THF serves as a methyl donor (Figure 4); therefore, deficiencies in either folate or cobalamin result in an accumulation of serum homocysteine. The other cobalamin-dependent enzyme reaction involves the conversion of methylmalonyl-coenzyme A (CoA) to succinyl-CoA where cobalamin (as adenosyl-cobalamin) serves as a cofactor for the enzyme methylmalonyl-CoA mutase (Figure 5). A cobalamin deficiency will lead to the accumulation of methylmalonyl-CoA and its hydrolysis product of MMA. Hence, increases in homocysteine may indicate a folate or cobalamin deficiency, while increases in MMA are specific only for the diagnosis of a cobalamin deficiency.

Figure 5.

Figure 5

Open in a new tab

The role of cobalamin in the metabolism of methymalonic acid (MMA). MCM = methymalonyl-CoA mutase, MCH = methylmalonyl-CoA hydrolase.

In the treated mice the serum folate and cobalamin concentrations were no different than the control animals; however, both plasma MMA and total homocysteine were elevated. In humans, homocysteine and MMA usually increase in the early stages of deficiency before measurable decreases in serum folate or cobalamin and, therefore, are considered more sensitive markers for deficiency than serum levels of the vitamins themselves (Holleland et al. 1999; Kang et al. 1987; Refsum et al. 2004) Thus, although cobalamin levels were unchanged, the findings of a macrocytosis with a decreased RBC count, increased HJ bodies, morphological changes consistent with a dyserythropoiesis, decreased leukocyte counts and elevations in homocysteine and MMA strongly support a perturbation of cobalamin metabolism. Because homocysteine does not differentiate between a folate or cobalamin deficiency, the possibility of a concurrent functional deficiency of folate cannot be ruled out. In humans, increased levels of homocysteine and MMA are also associated with renal failure; however, creatinine levels were unchanged in the treated mice and, in the previous BCE subchronic study, no renal histologic lesions were observed (Mercado-Feliciano et al. 2012).

In addition to serum folate levels, liver and kidney tissue folate levels were measured. A small but statistically significant increase in total kidney folate (THF + 5-methyl-THF) was observed while the individual levels of kidney THF and 5-methyl-THF were unchanged. In the limited reports of mouse kidney folate levels as they relate to folate deficiencies and in which there were similar hematological and biochemical findings (e.g., increased homocysteine) as the current study, kidney folate levels were decreased (Bills et al. 1992; Salojin et al. 2011); mouse studies regarding tissue vitamin levels in cases of cobalamin deficiency were not found in the literature. In light of these reported findings as well as those of the current study, the significance of the small increase in total kidney folate is uncertain and may represent biological variability.

There are many well-known causes of a functional folate or cobalamin deficiency. Small intestinal bacterial overgrowth can lead to cobalamin deficiency because the bacteria take up cobalamin and may compete with intrinsic factor (a cobalamin-binding protein) for it (Carmel 2014). Many compounds interfere with folate and cobalamin absorption or distribution. These include, ethanol (folate), anticonvulsant agents (folate; e.g., phenytoin, phenobarbital), aminosalicylic acid (folate and cobalamin), metformin (cobalamin), colchicine (cobalamin), and proton-pump inhibitors (cobalamin) (Carmel 2014; Hesdorffer and Lango 2015; Waxman et al. 1970). Other compounds specifically interfere with the metabolism of folate by way of enzyme inhibition, most commonly of the enzyme DHFR (Figure 4). Methotrexate and pemetrexed are two well-known inhibitors of DHFR. To ascertain whether BCE inhibits DHFR, BCE was tested with a DHFR assay. Results of this assay did not support BCE, at a concentration of 50mg/ml, as an inhibitor of DHFR.

The BCE lot used in this study was the same as that of the subchronic toxicity study. On chromatographic profiles, the BCE lot was similar to that of Remifemin® tablets (Mercado-Feliciano et al. 2012). Remifemin® tablets are a commercial product containing black cohosh extract and are readily available to the general public. Most clinical research on black cohosh has been conducted using Remifemin® (AHP 2002).

While this study elucidates changes in mice, direct extrapolation of our results to humans is difficult given that there are species differences between folate and cobalamin requirements, storage and metabolism. In addition, BCE preparations are variable natural mixtures of compounds and it is not known which of these compounds are active. The mice in the current study were dosed relatively high at 1000 mg/kg; however, hematological effects were observed as low as 62.5 mg/kg (low-dose group) in the subchronic study (Mercado-Feliciano et al. 2012). In addition, similar CBC changes were observed in the NTP subchronic rat BCE study (Mercado-Feliciano et al. 2012). The typical recommended daily dose of BCE for humans is 40 mg daily, which is approximately 0.5mg/kg for a 70kg person. The potential that low doses of BCE may cause or contribute to a functional cobalamin deficiency in humans warrants investigation, especially in light of the fact that cobalamin deficiency is encountered with some frequency in older human patients and is most commonly caused by a food-cobalamin malabsorption (Andrès et al. 2003; Andrès et al. 2007).

In conclusion, administration of BCE to female B6C3F1/N mice caused hematological and biochemical changes consistent with a functional cobalamin deficiency. Further, the possibility of a functional folate deficiency could not be ruled out. While serum folate and cobalamin levels were no different than controls, the hematological changes and increases in homocysteine and MMA support a perturbation of cobalamin metabolism. Further studies are needed to fully characterize bone marrow changes as they relate to BCE treatment in mice and to elucidate the mechanisms by which administration of BCE to mice causes increases in homocysteine and MMA.

Acknowledgments

This research was supported by the NIH, National Institute of Environmental Health Sciences. The authors would like to thank Drs. Cynthia Rider and Arun Pandiri for their critical review of this article. The authors also thank Laura Betz for help with the statistical analysis, David Sabio for his creation of some of the figures and Veronica Godfrey Robinson for her assistance with the chemistry work. The study was conducted at Alion Life Sciences Laboratory, Research Triangle Park, NC. Dose formulation work was done at the facilities of Battelle Memorial Institute, Columbus, OH. Mass spectrometry and LS were performed at RTI International, Research Triangle Park, NC.

References

  1. American Herbal Pharmacopoeia (AHP) Black cohosh rhizome, Actaea racemosa L. syn. Cimicifuga racemosa (L.) Nutt., standards of analysis, quality control, and therapeutics. In: Upton R, editor. American Herbal Pharmacopoeia and Therapeutic Compendium. American Herbal Pharmacopoeia; Santa Cruz, California: 2002. [Google Scholar]
  2. Andrès E, Perrin AE, Demangeat C, Kurtz JE, Vinzio S, Grunenberger F, Goichot B, Schlienger JL. The syndrome of food-cobalamin malabsorption revisited in a department of internal medicine. A monocentric cohort study of 80 patients. Eur J Intern Med. 2003;14:221–26. doi: 10.1016/s0953-6205(03)00074-8. [DOI] [PubMed] [Google Scholar]
  3. Andrès E, Vidal-Alaball J, Federici L, Loukili NH, Zimmer J, Kaltenbach G. Clinical aspects of cobalamin deficiency in elderly patients. Epidemiology, causes, clinical manifestations, and treatment with special focus on oral cobalamin therapy. Eur J Intern Med. 2007;18:456–62. doi: 10.1016/j.ejim.2007.02.013. [DOI] [PubMed] [Google Scholar]
  4. Bills ND, Koury MJ, Clifford AJ, Dessypris EN. Ineffective hematopoiesis in folate-deficient mice. Blood. 1992;79:2273–80. [PubMed] [Google Scholar]
  5. Blumenthal M. The ABC Clinical Guide to Health. American Botanical Council; Austin, TX: 2003. Black Cohosh; pp. 13–20. [Google Scholar]
  6. Borrelli F, Ernst E. Black cohosh (Cimicifuga racemosa) for menopausal symptoms: a systemic review of its efficacy. Pharmacol Res. 2008;58:8–14. doi: 10.1016/j.phrs.2008.05.008. [DOI] [PubMed] [Google Scholar]
  7. Bottiglieri T, Arning E, Wasek B, Nunbhakdi-Craig V, Sontag JM, Sontag E. Acute administration of L-Dopa induces changes in methylation metabolites, reduced protein phosphatase 2A methylation, and hyperphosphorylation of tau protein in mouse brain. J Neurosci. 2012;32:9173–81. doi: 10.1523/JNEUROSCI.0125-12.2012. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Cabelof DC, Raffoul JJ, Nakamura J, Kapoor D, Aballa H, Heyari AR. Imbalanced base excision repair in response to folate deficiency is accelerated by polymerase β haploinsufficiency. J Biol Chem. 2004;279:36504–513. doi: 10.1074/jbc.M405185200. [DOI] [PubMed] [Google Scholar]
  9. Carmel R. Macrocytosis, mild anemia, and delay in the diagnosis of pernicious anemia. Arch Intern Med. 1979;139:47–50. [PubMed] [Google Scholar]
  10. Carmel R. Megaloblastic anemias: disorders of impaired DNA synthesis. In: Greer JP, Arber DA, Glader B, List AF, Means RT, Paraaskevas F, Rodgers GM, editors. Wintrobe’s Clinical Hematology. Lippincott Williams and Wilkins; Philadelphia, PA: 2014. pp. 927–53. [Google Scholar]
  11. Ernest S, Hosack A, O’Brien WE, Rosenblatt DS, Nadeau JH. Homocysteine levels in A/J and C57BL/6J mice: genetic, diet, gender and parental levels. Physiol Genomics. 2005;21:404–10. doi: 10.1152/physiolgenomics.00199.2004. [DOI] [PubMed] [Google Scholar]
  12. Everson RB, Wehr CM, Erexson GL, MacGregor JT. Association of marginal folate depletion with increased human chromosomal damage in vivo: demonstration by analysis of micronucleated erythrocytes. J Natl Cancer Inst. 1988;80:525–29. doi: 10.1093/jnci/80.7.525. [DOI] [PubMed] [Google Scholar]
  13. Fenech M. Folate (vitamin B9) and vitamin B12 and their function in the maintenance of nuclear and mitochondrial genome integrity. Mutat Res. 2012;733:21–33. doi: 10.1016/j.mrfmmm.2011.11.003. [DOI] [PubMed] [Google Scholar]
  14. Hall CA. Vitamin B12 deficiency and early rise in mean corpuscular volume. J Amer Med Assoc. 1981;245:1144–46. [PubMed] [Google Scholar]
  15. Herbert V. Experimental nutritional folate deficiency in man. Trans Assoc Am Phys. 1962;75:307–20. [PubMed] [Google Scholar]
  16. Hesdorffer CS, Longo MD. Drug-induced megaloblastic anemia. N Engl J Med. 2015;373:1649–58. doi: 10.1056/NEJMra1508861. [DOI] [PubMed] [Google Scholar]
  17. Holleland G, Schneede J, Ueland PM, Lund PK, Refsum H, Sandberg S. Cobalamin deficiency in general practice. Assessment of the diagnostic utility and cost-benefit analysis of methylmalonic acid determination in relation to current diagnostic strategies. Clin Chem. 1999;45:189–98. [PubMed] [Google Scholar]
  18. Institute for Laboratory Animal Research. Guide for the care and use of laboratory animals. National Academies Press; Washington, DC: 2011. [Google Scholar]
  19. Kang S-S, Wong PWK, Norusis M. Homocysteinemia due to folate deficiency. Metabolism. 1987;36:458–62. doi: 10.1016/0026-0495(87)90043-6. [DOI] [PubMed] [Google Scholar]
  20. Klee GG. Cobalamin and folate evaluation: measurement of methymalonic acid and homocysteine vs vitamin B12 and folate. Clin Chem. 2000;46:1277–83. [PubMed] [Google Scholar]
  21. Koury MJ. Abnormal erythropoiesis and the pathophysiology of chronic anemia. Blood Reviews. 2014;28:49–66. doi: 10.1016/j.blre.2014.01.002. [DOI] [PubMed] [Google Scholar]
  22. Koury MJ, Horne DW, Brown ZA, Pietenpol JA, Blount BC, Ames BN, Hard R, Koury ST. Apoptosis of late-stage erythroblasts in megaloblastic anemia: association with DNA damage and macrocyte production. Blood. 1997;89:4617–23. [PubMed] [Google Scholar]
  23. Koury JM, Price JO, Hicks GG. Apoptosis in megaloblastic anemia occurs during DNA synthesis by p53-independent, nucleoside-reversible mechanism. Blood. 2000;96:3249–55. [PubMed] [Google Scholar]
  24. Leach MJ, Moore V. Black cohosh (Cimicifuga spp.) for menopausal symptoms. Cochrane Database of Systematic Reviews. 2012;9:CD007244. doi: 10.1002/14651858.CD007244.pub2. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Leamon CP, Reddy JA, Dorton R, Bloomfield A, Emsweller K, Parker N, Westrick E. Impact of high and low folate diets on tissue folate receptor levels and antitumor responses toward folate-drug conjugates. J Pharmacol Exp Ther. 2008;327:918–25. doi: 10.1124/jpet.108.143206. [DOI] [PubMed] [Google Scholar]
  26. Means RT, Jr, Glader B. Disorders of red cells. In: Greer JP, Arber DA, Glader B, List AF, Means RT, Paraaskevas F, Rodgers GM, editors. Wintrobe’s Clinical Hematology. Lippincott Williams and Wilkins; Philadelphia, PA: 2014. pp. 927–53. [Google Scholar]
  27. Mercado-Feliciano M, Cora MC, Witt KL, Granville CA, Hejtmancik MR, Fomby L, Knostman KA, Ryan MJ, Newbold R, Smith C, Foster PM, Vallant MK, Stout MD. An ethanolic extract of black cohosh causes hematological changes but not estrogenic effects in female rodents. Toxicol Appl Pharm. 2012;263:138–47. doi: 10.1016/j.taap.2012.05.022. [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Refsum H, Smith AD, Ueland PM, Nexo E, Clark R, McPartlin J, Johnston C, Engbaek F, Schneede J, McPartlin C, Scott JM. Facts and recommendations about total homocysteine determinations: an expert opinion. Clin Chem. 2004;50:3–32. doi: 10.1373/clinchem.2003.021634. [DOI] [PubMed] [Google Scholar]
  29. Salojin KV, Cabrera RM, Sun W, Chang WC, Lin C, Duncan L, Platt KA, Read R, Vogel P, Liu Q, Finnell RH, Oravecz T. A mouse model of hereditary folate malabsorption: deletion of the PCFT gene leads to systemic folate deficiency. Blood. 2011;117:4895–4904. doi: 10.1182/blood-2010-04-279653. [DOI] [PubMed] [Google Scholar]
  30. Savage DG, Lindenbaum J, Stabler SP, Allen RH. Sensitivity of methymalonic acid and total homocysteine determination for diagnosing cobalamin and folate deficiencies. Am J Med. 1994;96:239–46. doi: 10.1016/0002-9343(94)90149-x. [DOI] [PubMed] [Google Scholar]
  31. Shane B. Folate and vitamin B12 function. In: Lennarz WJ, Lane MD, editors. Encyclopedia of Biological Chemistry. Elsevier; San Diego, CA: 2013. pp. 324–328. [Google Scholar]
  32. Smith T, Lynch ME, Johnson J, Kawa K, Bauman H, Blumenthal M. Herbal dietary supplement sales in US increases 6.8% in 2014. HerbalGram. 2015;107:52–9. [Google Scholar]
  33. Stabler SP. Vitamin B12 deficiency. Engl J Med. 2013;368:140–60. doi: 10.1056/NEJMc1304350. [DOI] [PubMed] [Google Scholar]
  34. Stover PJ. One-carbon metabolism-genome interactions in folate-associated pathologies. J Nutr. 2009;139:2402–05. doi: 10.3945/jn.109.113670. [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Troen AM, Shea-Budgell M, Shukitt-Hale B, Smith DE, Selhub J, Rosenberg IH. B-vitamin deficiency causes hyperhomocysteinemia and vascular cognitive impairment in mice. P Natl Acad Sci USA. 2008;105:12474–479. doi: 10.1073/pnas.0805350105. [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Waxman S, Corcino JJ, Herbert V. Drugs, Toxins and dietary amino acids affecting vitamin B12 or folic acid absorption or utilization. Am J Med. 1970;48:599–608. doi: 10.1016/0002-9343(70)90010-0. [DOI] [PubMed] [Google Scholar]
  37. Zhou T, Kinney MC, Scott LM, Zinkel SS, Rebel VI. Revisiting the case for genetically engineered mouse models in human myelodysplastic syndrome research. Blood. 2015;126:1057–1068. doi: 10.1182/blood-2015-01-624239. [DOI] [PMC free article] [PubMed] [Google Scholar]

RESOURCES