Large-scale, high-throughput screening for coagulation and hematologic phenotypes in mice*

Luanne L. Peters1, Eleanor M. Cheever2, Heather R. Ellis1, Phyllis A. Magnani1, Karen L. Svenson1, Randy Von Smith1 and Molly A. Bogue1

1 The Jackson Laboratory, Bar Harbor, Maine 04609
2 Dade Behring, Inc., Deerfield, Illinois


    ABSTRACT
 TOP
 ABSTRACT
 INTRODUCTION
 METHODS
 RESULTS
 DISCUSSION
 References
 
The Mouse Phenome Project is an international effort to systematically gather phenotypic data for a defined set of inbred mouse strains. For such large-scale projects the development of high-throughput screening protocols that allow multiple tests to be performed on a single mouse is essential. Here we report hematologic and coagulation data for more than 30 inbred strains. Complete blood counts were performed using an Advia 120 analyzer. For coagulation testing, we successfully adapted the Dade Behring BCS automated coagulation analyzer for use in mice by lowering sample and reagent volume requirements. Seven automated assay procedures were developed. Small sample volume requirements make it possible to perform multiple tests on a single animal without euthanasia, while reductions in reagent volume requirements reduce costs. The data show that considerable variation in many basic hematological and coagulation parameters exists among the inbred strains. These data, freely available on the World Wide Web, allow investigators to knowledgeably select the most appropriate strain(s) to meet their individual study designs and goals.

hemostasis; hematology; mouse phenome; NHLBI Program for Genomic Applications


    INTRODUCTION
 TOP
 ABSTRACT
 INTRODUCTION
 METHODS
 RESULTS
 DISCUSSION
 References
 
THE ABUNDANCE OF SEQUENCE information available from the human genome project and from efforts to sequence an ever-growing number of model organisms has left science firmly ensconced in the post-genome era. The challenge now is to assign function to the sequence data.

Although there are many avenues by which to assign function to genes (e.g., bioinformatic and gene-driven strategies such as knockouts), phenotype-driven approaches followed by positional cloning will clearly play a critical role in the functional annotation of the mammalian genome in the coming years. Attesting to this is the number of large-scale mouse mutagenesis centers now operating, eight in the United States alone (1). Such centers draw on the precedence set by classic genetic studies of spontaneous mouse mutants; positional cloning of the gene defect in combination with extensive phenotypic data provides valuable clues to gene function. To cite just two recent examples, the study of the spontaneous mouse mutation obese (ob) led to the discovery of leptin and has revolutionized our concept of obesity (5), while studies of iron-deficient mutants (e.g., microcytic anemia, mk; sex-linked anemia, sla) have unveiled long-sought iron transport mechanisms (3, 11).

In addition to mutagenesis, a second largely untapped source of phenotypic variation can be found among the inbred mouse strains (6). Each strain has a fixed, homozygous genetic background that can be sampled repeatedly, yet there is great sequence variation between strains, attributable to their diverse origins. Extensive phenotyping of the common inbred strains serves the scientific community by allowing investigators from all disciplines to choose appropriate strains for study. A clear example can be found in the analysis of complex traits; a body of phenotyping data across multiple inbred strains allows easy identification of the most suitable parental strains for setting up an appropriate cross to identify quantitative trait loci (QTL) influencing a phenotype of interest. With "in silico" identification and refinement of QTL intervals now a reality, a repository of phenotypic data on the common inbred strains becomes an invaluable resource for the dissection of complex traits. Moreover, as concordance between QTL in the mouse and human has been demonstrated for a variety of diseases, complex trait analysis in the mouse has significant biomedical relevance (7, 9).

In recognition of the need to provide extensive, systematic phenotyping of the common inbred strains, an international committee of scientists established the Mouse Phenome Project in 1999 (6). Investigators from laboratories worldwide are contributing data across a broad spectrum of physiological, neurological, behavioral, and genetic disciplines. All data can be freely accessed and downloaded from the Mouse Phenome Database (MPD) (http://www.jax.org/phenome), which is maintained by The Jackson Laboratory (http://www.jax.org) and linked to the Mouse Genome Informatics (MGI) Database (http://www.informatics.jax.org/).

As part of The Jackson Laboratory Program for Genomic Applications (PGA) sponsored by the National Heart, Lung, and Blood Institute, we are contributing heart-, lung-, and blood-related data to the MPD. This data can be accessed directly from the Jax PGA website (http://pga.jax.org).

In this report, we present our strategies to screen for blood phenotypes in mice. We 1) describe our modifications of the Dade Behring Blood Coagulation System (BCS) analyzer that allow automated high-throughput, repetitive coagulation testing in mice, 2) present data that validate our modified mouse coagulation testing protocols using the BCS, and 3) present extensive inbred strain survey data for hematologic and coagulation parameters. The data show significant variation in many basic hematological and coagulation parameters among the inbred strains, providing the groundwork for investigators to knowledgeably select strains for QTL studies, sensitized mutagenesis screens, physiological analyses, and drug development.


    METHODS
 TOP
 ABSTRACT
 INTRODUCTION
 METHODS
 RESULTS
 DISCUSSION
 References
 
Mice
All mice were obtained from The Jackson Laboratory (Bar Harbor, ME). They were housed in humidity- and temperature-controlled rooms with a 12-h light cycle. They had free access to acidified water and chow (NIH 5K52). All protocols were approved by The Jackson Laboratory Animal Care and Use Committee. The Jackson Laboratory is fully accredited by the American Association for Accreditation of Laboratory Animal Care (AAALAC). For the strain characterization project, prothrombin time (PT), activated partial thromboplastin time (APTT), fibrinogen (Fib), and fibrinogen low range (Fib low, <125 mg/dl) assays were performed on 10-wk-old mice (the standard for the Mouse Phenome Project). Testing to validate newly developed assays [antithrombin III (ATIII), factor VIII (FVIII), and protein C (ProC)] was performed on 6-wk-old C57BL/6J (B6) mice unless otherwise noted.

Coagulation Studies
Whole blood (~275 µl) was drawn from the retro-orbital sinus through an uncoated microhematocrit tube (Fisher Scientific, Pittsburgh, PA) directly into an Eppendorf tube containing 30 µl of 3.8% sodium citrate in murine (340 mosM) phosphate-buffered saline (PBS; 10 mM NaCl, 155 mM KCl, 10 mM glucose, 1 mM MgCl2, 2.5 mM KHPO4, pH 7.4). The hematocrit values for the strains analyzed ranged from a mean of 36.5% (MOLF/Ei) to 46.5% (C57BR/cdJ). Hence, the standard ratio of whole blood to anticoagulant was used (9:1 vol/vol whole blood:3.8% citrate) in all cases. For each mouse, the collection process was completed within 5–20 s. All plasma samples were collected by retro-orbital bleeding with the exception of a subset of ATIII samples, which were collected by cardiac puncture. Cardiac puncture was performed under general anesthesia (Avertin), and blood (450 µl) was withdrawn using a 25-gauge needle containing 50 µl of 3.8% sodium citrate. All samples were centrifuged at 15,000 rpm for 10 min to separate plasma. The plasma was centrifuged again at 15,000 rpm for 10 min to remove any remaining cellular debris. We found that the second centrifugation step was critical in establishing reliable baselines, as nearly all BCS measurements depend upon changes in turbidity to detect the appropriate endpoint. Centrifugations were carried out at room temperature. The samples were analyzed immediately using the Dade Behring (Marburg, Germany) BCS analyzer. As no mouse-specific coagulation testing reagents are available, reagents and standards designed for use in human clinical testing were used in all procedures. Dade Behring reagents were used for the PT, APTT, Fib, Fib low, FVIII, and ATIII assays. Some Dade Behring assays have a choice of reagents. In this study, thromboplastin C plus (from rabbit brain) was used as the substrate in the PT assay, actin FS (1.0 x 10-4 M ellagic acid, the Dade Behring activator most sensitive to factor deficiencies) was used in the APTT assay, and thrombin (bovine) was used in the fibrinogen assay. ProC reagents were obtained from Diapharma (West Chester, OH). All lyophilized reagents were reconstituted according to the manufacturer’s instructions. Owren’s veronal buffer was used as diluent in the Fib and FVIII assays, and NaCl was used in the ATIII assay.

Principles of BCS assay protocols are available from Dade Behring package inserts, which can be found on Dade Behring’s web site (http://www.dadebehring.com) or by calling technical assistance (800-242-3233). The modified BCS assay procedures that we developed are described below (see RESULTS). Table 1 provides a summary of the settings used for each assay.


View this table:
[in this window]
[in a new window]
 
Table 1. Summary of BCS mouse assay procedures

 

    RESULTS
 TOP
 ABSTRACT
 INTRODUCTION
 METHODS
 RESULTS
 DISCUSSION
 References
 
Modification of the Dade Behring BCS Protocols
Figure 1A shows the standard human assay procedure for determination of the APTT. The assay requires 50 µl each of citrated plasma sample, actin FS reagent, and calcium chloride. These reagents are added to the appropriate chamber via inlet ports on the BCS sample rotor (Fig. 1, B and C). Each "spoke" on the BCS rotor is essentially a test tube. Samples/reagents are added, incubated as needed, and mixed once the rotor commences spinning. The final value is obtained by changes in turbidity as clotting occurs and is measured in the outer cuvette while the rotor is spinning (Fig. 1C).



View larger version (100K):
[in this window]
[in a new window]
 
Fig. 1. A: display of the standard BCS assay procedure for APTT, showing the pipetting cycles and the sample/reagent volumes (µl). Volumes shown to the left are those used in the standard human (H) assay and those on the right are those utilized in the modified mouse (M) protocol. Similar program changes were made to all other assay procedures (see text and Table 1) to reduce sample and reagent volume requirements for testing of mice. B: the BCS sample rotor where all reactions, incubations, and measurements take place. C: close-up view of a portion of the BCS sample rotor showing sample and reagent ports and the cuvette where measurements are taken as the rotor is spinning. Endpoints are based on changes in turbidity (PT, APTT, Fib, FVIII) or color (ATIII, ProC). PT, prothrombin time; APTT, activated partial thromboplastin time; Fib, fibrinogen; ATIII, antithrombin III; FVIII, factor VIII; and ProC, protein C.

 
We duplicated the standard human assay procedure file and changed the sample, reagent, and transfer volumes as shown (Fig. 1A), creating a new "mouse" protocol. Similarly, we halved the sample and reagent volumes for PT, Fib, Fib low, ATIII, FVIII, and ProC assays (not shown). For each assay, we also decreased the lag time, allowing the instrument to start the actual measurement earlier and establish a reliable baseline. This is important for measurements in mice, as clotting times in mice tend to be faster than in humans. Calibration curves were generated using standard human plasma (SHP, Dade Behring) as reference.

To determine whether the new mouse protocols were valid, we compared the values obtained for two commercially available controls, Control N (normal) and Control P (pathological) (Dade Behring), using the standard human protocols and our modified mouse protocols. Although there were some statistically significant differences, the values obtained using the modified mouse protocols for PT, APTT, Fib, Fib low, and ATIII were consistently within the established range of limits provided for each lot of the commercial controls and were very close to those obtained using the standard human assay procedures (Table 2). Hence, the mouse protocols give reproducible and verifiable results. Most significantly, the PT, APTT, and Fib strain characterization data (see below) are consistent with previously reported results (10).


View this table:
[in this window]
[in a new window]
 
Table 2. Comparison of standard human and modified mouse BCS protocols

 
ATIII levels in adult mice cannot be compared with previous results, as none have been reported. We assessed ATIII levels in 6-wk-old B6 mouse plasma using our modified BCS mouse assay with SHP as the reference plasma. In male mice levels of 171 ± 15% normal (mean ± SD; n = 98) were obtained, while females averaged 146 ± 20% normal (n = 70). As ATIII would be expected to be particularly sensitive to tissue damage upon retro-orbital bleeding, we compared results from samples obtained from adult male NZW/LacJ mice at 3–4 mo of age by eye-bleeding and by cardiac puncture. Retro-orbital and cardiac ATIII levels in these mice were 123 ± 5.0% normal (n = 5) and 88.9 ± 10.2% normal (n = 5, P < 0.001), respectively. Clearly, tissue trauma associated with eye-bleeding even by highly experienced technical personnel elevates ATIII levels. A consistent method of sample procurement is therefore critical.

We conclude from these studies that the modified assay protocols are valid for use as standard procedures for automated analyses in mice.

Other Tests
FVIII.
Using human FVIII-deficient plasma and SHP as the reference, mouse FVIII levels measured in a coagulometric assay are too high to read on the standard curve. The BCS instrument allows for automatic dilution of samples. We obtained FVIII levels for Control N of 87.8 ± 3.8% (n = 18), within the expected range, using the mouse protocol at a dilution of 1:8. In 6-wk-old B6 mice, FVIII values of 266 ± 58% normal (n = 20) were obtained using the mouse BCS protocol. These results are in keeping with previously reported mouse FVIII levels (10). Male levels are higher than female (308 ± 44% vs. 224 ± 40%). To further validate our FVIII protocol, we measured FVIII in plasma samples from hybrid (B6, 129) von Willebrand factor (vWF) null mice (kindly provided by Dr. Denisa Wagner, Harvard Medical School, Boston, MA), which have a severe deficiency of FVIII (20% of wild-type levels) (2). We obtained values of 54.4 ± 3.6% normal (n = 5) in these mice, or 20% of the levels we obtained in normal B6 mice. From the foregoing, we conclude that our modified mouse FVIII protocol is valid for automated testing in mice. The mouse FVIII assay protocol serves as a prototype for the development of automated assays for other coagulation factors in mice.

Protein C.
A chromogenic ProC assay kit from Diapharma was used by Jalbert et al. (4) in an ELISA format to measure ProC levels in mice using pooled normal mouse plasma as reference. Notably, they detected no ProC in homozygous ProC null mice, and 63.5% of normal in heterozygous ProC-deficient mice, providing validity to the assay. The assay depends upon activating endogenous ProC with an enzyme from snake venom (Protac C). Subsequent hydrolysis of a chromogenic substrate (S-2366) serves as the endpoint to determine ProC activity. Diapharma provided BCS settings for testing of human plasma. With the standard human protocol and SHP to generate the standard curve, all mouse values obtained were low (range 0.6–9.1% normal), indicating that mouse ProC levels as measured in this assay are much lower than in humans. Therefore, we established a standard curve using standard mouse plasma (SMP) pooled from adult B6 mice (10 adult females, 10 males) as reference. A reference value of 100% was assigned to the SMP. Assays of normal 6-wk-old B6 mouse samples gave an average ProC level of 98.2 ± 12% in males and 90.8 ± 12.2% in females.

An obvious concern is the effectiveness and specificity of Protac C in activating mouse ProC. In our assay, the Protac C reagent is reconstituted as recommended by the manufacturer to a final concentration of 0.17 U/ml (Table 1). Notably, increasing the Protac C concentration to as high as 1.2 U/ml did not increase the amount of detectable ProC in mouse plasma samples, and substitution of the Protac C with PBS resulted in readings of zero. In addition, incremental increases in the amount of substrate (up to 7.5x the manufacturer’s recommended concentration of 0.8 µg/µl) while maintaining the Protac C at 0.17 U/ml increased the amount of ProC detected up to an asymptote (data not shown). Together, these data suggest that Protac C is activating mouse ProC and that the amount of activated ProC is not limiting under our assay conditions. Moreover, the previously reported "genetic" depletion studies using ProC knockout mice (4) described above attest to the specificity of Protac C. We conclude that the BCS ProC assay is specific and valid for comparative studies in mice with the caveat that levels obtained are relative to an unassayed pooled mouse reference sample.

Stability and Reproducibility of the Mouse BCS Assays
The availability of assayed commercial controls and standards allows one to easily monitor assay performance over time. The mouse BCS procedures are very stable and reproducible. Control N and Control P are tested each day samples are analyzed. Their values have been consistently within the established ranges for each of the mouse assays for as long the assays have been performed (2 yr for PT, APTT, and Fib, 1 yr for ATIII, and 6 mo for ProC and FVIII). In addition, our program includes running seasonal controls each quarter for all tests performed. Table 3 shows results for three consecutive quarters for each assay. Although there are some statistically significant differences (likely representing normal seasonal variations in mice), the results overall are remarkably consistent. FVIII assays have not yet been performed for two consecutive quarters. However, results in B6 mice serially bled at 6, 8, and 10 wk of age show no significant differences (Table 3).


View this table:
[in this window]
[in a new window]
 
Table 3. Stability/reproducibility of mouse BCS assays over time

 
Strain Characterization Survey
The Mouse Phenome Project is an international effort to provide systematic phenotyping data across a wide variety of disciplines on the commonly used and genetically diverse inbred mouse strains. We are characterizing coagulation and hematology phenotypes in 42 inbred strains, including several wild-derived strains (e.g., CAST/Ei, SPRET/Ei, CZECHII/Ei) and the progenitor strains for the two largest and most commonly used recombinant inbred (RI) lines, DBA/2J, C57BL6/J, and A/J (Table 4). We attempted to test at least 10 males and 10 females of each strain; for some strains we have data for as many as 23 mice per sex.


View this table:
[in this window]
[in a new window]
 
Table 4. List of 42 inbred strains being phenotyped by The Jackson Laboratory Program for Genomic Applications as part of the international Mouse Phenome Project

 
Coagulation (PT, APTT, Fib) and hematologic data for more than 30 inbred strains are currently available. As seen in Fig. 2, there is significant interstrain and sex variation in coagulation parameters in mice. These differences can be exploited for physiological studies, QTL analyses, and sensitized mutagenesis screens. Data for three hematologic parameters are shown in Fig. 3. We have noted repeatedly that mouse blood clots easily. We solved this problem by collecting blood using EDTA-coated microhematocrit tubes, adding an additional 20 µl of 20% EDTA, and mixing (see METHODS). All hematology data posted on the MPD were obtained in this way. Note the marked variation in total white blood cell and platelet counts among the inbred strains (4-fold and 2.5-fold differences, respectively) (Fig. 3, A and B). There is also significant variation in the percent circulating neutrophils, with as much as 4.9-fold differences among strains (Fig. 3C). Plots and data for other hematologic parameters are also available on the MPD website (red blood cell count, measured hemoglobin, hematocrit, mean cell volume, mean cell hemoglobin content, mean cell hemoglobin concentration, red cell hemoglobin concentration mean, red cell distribution width, hemoglobin concentration distribution width, % lymphocyte, % monocyte, % eosinophil, % large unstained cell, % basophil, % reticulocyte, calculated hemoglobin, and mean platelet volume).



View larger version (41K):
[in this window]
[in a new window]
 
Fig. 2. Coagulation measurements for over 30 inbred strains of mice. These plots were generated on the MPD website and downloaded for publication. Each plot shows the result of one measurement for all mouse strains tested. Strain names are shown across the x-axis, and the measurement of interest is shown on the y-axis. For each strain that was tested, one or two bars are shown (gray for male, black for female). The middle tick in each bar represents the mean; outer ticks represent statistical error range (in all cases shown the default is ±1 standard error). The overall mean for all strains is indicated by the bold dotted line, and ±1 standard deviation is shown by the thin dotted line. These plots display strains in numeric order based on the overall mean (male and female combined) for each strain. These plots could also be displayed in alphabetical order by strain name via the MPD web interface. Additional options for viewing measurements on the MPD web site include y-axis range adjustment; linear vs. log transformation; error bar representation of 1 standard error, 2 standard errors, 1 standard deviation, or min-max range; strain display by sex (males and females on the same plot; males and females on separate plots; or sexes combined as strain mean); data selection conditions (e.g., show only mice that were over 14 wk of age); and other appearance-related details for publication-ready images. Values used for each plot and other statistics are available on the MPD website in tabular form for viewing and downloading (not shown). A: prothrombin time (PT) strain means for males and females. The overall mean is 10.19 ± 0.48. Inset: shows a frequency distribution histogram. These plots are available for all measurements using the numeric order display option. Continuous numeric data are lumped into "bins" each holding a set range (default bin size is based on range of data but can be controlled by the MPD user). B: partial thromboplastin time (APTT) strain means for males and females. The overall mean is 23.14 ± 2.33. C: fibrinogen (Fib) strain means for males and females. The overall mean is 214.45 ± 35.55.

 


View larger version (48K):
[in this window]
[in a new window]
 
Fig. 3. Selected hematologic data for 42 inbred strains of mice. See general information about plots in legend for Fig. 2. A: white blood cell count (WBC) strain means for males and females. The overall mean is 7.45 ± 2.44. B: platelet count (Plt) strain means for males and females. The overall mean is 1,138.70 ± 287.75. C: percentage of neutrophils among circulating white blood cells, expressed as strain means for males and females. The overall mean is 14.63% ± 5.97%.

 
Correlations can be determined between pairwise measurements using MPD analysis tools. As seen in Fig. 4, a highly significant correlation exists between the hemoglobin concentration as directly measured on the Advia hematology analyzer (mHGB) and the calculated hemoglobin (cHGB), which is determined from three independent Advia measurements, the mean of the red cell hemoglobin concentration histogram (CHCM), the red blood cell count, and the mean corpuscular volume (MCV). This correlation is expected in the absence of factors known to interfere with the mHGB (e.g., lipemia, elevated white blood cell count) and thus provides an excellent means to validate our results.



View larger version (13K):
[in this window]
[in a new window]
 
Fig. 4. Scatter plot of measured hemoglobin (mHGB; g/dl) vs. calculated hemoglobin (cHGB; g/dl) for 38 inbred strains. This linear regression plot was generated on the MPD website and downloaded for publication. This analysis option shows graphically the strain means (by sex) between two selected measurements to help identify possible genetic correlations. Each data point represents a strain (male gray, female black). This combination of measurements gives a compelling Pearson correlation coefficient of 0.711 (y = -1.66 + 0.27x). The MPD website provides additional scatter plot options, including labeling each data point with strain name, or highlighting selected strains; plotting male only, female only; adjusting axis range; showing error bars for both dimensions; and plotting one or both measurements in log. In addition to expressing data points as strain means, scatter plots may be generated per individual mouse if the data structure permits. The pairwise correlation function of the MPD is useful for identifying new correlations as well as confirming previously determined genetic overlap.

 

    DISCUSSION
 TOP
 ABSTRACT
 INTRODUCTION
 METHODS
 RESULTS
 DISCUSSION
 References
 
Large-scale, high-throughput phenotyping in mice is critical to the success of ethylnitrosourea (ENU) mutagenesis programs and to extensive strain characterization projects such as the Mouse Phenome Project. The unparalleled genetic resources available make the mouse an extraordinary tool for analysis of both single gene function and complex traits. To truly take advantage of this genetic infrastructure, novel methods to easily screen mice for physiologically relevant phenotypes are needed. Such methods are especially critical to QTL analyses which require phenotyping of large numbers of intercross or backcross progeny.

We have successfully adapted the Dade Behring BCS coagulation analyzer for use in mice. By lowering sample and reagent volume requirements, mice can be tested for several parameters, and reagent costs can be kept to a minimum. The three most basic coagulation tests, i.e., PT, APTT, and Fib, can be performed on 105 µl of plasma (25 µl per test plus 30 µl "dead" volume). The same mice can be re-bled at a later date for FVIII, ATIII, and ProC determinations, which together require 50 µl plasma, including dead volume. (We routinely allow at least 1 wk between bleeds to ensure that the mice have completely recovered.) We have documented that the reduced sample and reagent volumes give valid and consistent results. The reason the decreased total reaction volume does not markedly skew the results is related to the BCS instrumentation. All measurements are taken while the sample rotor is spinning. As long as the reaction volume remains over the light source in the outer cuvette portion of the rotor (Fig. 1C), changes in turbidity/color are reliably detected. We determined that a total volume of 75 µl is sufficient to maintain the fluid level in the cuvette within the path of the light source during centrifugation.

Additional advantages of the BCS assay system include the availability of assayed reference plasma (SHP) and controls. The exception is the ProC protocol, in which an SMP reference is required, as discussed above. However, in some instances, such as relatively short-term studies, use of SMP in assay procedures may be preferred. The ease and flexibility of the BCS system allows such assays to be developed for any desired measurement.

Not all BCS assays are suitable for testing in mice. For example, we were unable to adapt the BCS vWF assay for use in mice. The assay relies upon decreases in turbidity as sample vWF agglutinates a suspension of human platelets. Results in mice proved wildly erratic, suggesting that the use of human platelets was less than ideal.

Sequential testing of mice obviously requires obtaining blood without euthanasia. Hence, we have performed all testing on plasma samples obtained from the retro-orbital sinus. For all tests, consistent results have been obtained over extended periods of time, and these results agree well with previously published data. Hence, valid comparative evaluations of coagulation values obtained by proficient retro-orbital bleeding are possible. This is not meant to imply that values will not vary if the method of procurement varies. Clearly, this was the case with ATIII testing and will likely be the case with other procedures as well, especially those particularly sensitive to tissue trauma.

The current generation of hematology instruments requires small sample volumes and utilizes software suitable for analyzing many different species. Unlike coagulation studies, therefore, automated hematological analysis of large numbers of mice is a relatively simple endeavor. Examples of coagulation and hematologic strain survey data, as displayed on the MPD web site, are provided in Figs. 2 and 3, respectively. Raw data and other information can be accessed through our MPD project web pages (8). In addition to raw data retrieval, summary data can be downloaded graphically from the MPD web site, as shown in measurement profiles and distribution histograms (Fig. 2, inset), as well as in tabular form. Among other functions, options include the ability to directly compare a subset of strains and to identify outlier strains for all measurements. Correlations of strain means are computed between all pairwise measurement combinations in the MPD, pointing to possible genetic correlations that warrant further investigation. The MPD web site also provides access to protocols and information regarding the health status and the environment of the mice. Information for the following environmental parameters, at the vendor site and during acclimation and testing periods, are available: feed and other supplements, bedding, housing system, photoperiod, temperature, and relative humidity.

Our initial goal, the completion of basic coagulation testing (PT, APTT, Fib) and hematological analyses for the 42 top priority inbred strains identified by the Mouse Phenome Project is nearing completion. Future endeavors will include addition of other coagulation parameters to the MPD using our newly developed high-throughput assays (e.g., ATIII, FVIII). It is noteworthy that many phenotypes in addition to coagulation and hematology can be found in the MPD, and all data are freely available, making it a truly remarkable resource. In addition, investigators are encouraged to submit their own phenotyping data. Continued efforts by the scientific community as a whole to submit data, particularly as new phenotyping methods in mice emerge, will ensure that the database remains up to date and useful for a wide variety of disciplines.

The Jackson Laboratory PGA also includes a mutagenesis component. Our goal is to screen up to 4,000 G3 offspring of ENU-mutagenized mice per year for recessive mutations affecting biomedically relevant phenotypes (e.g., blood formation and coagulation, cardiac function, lung function, atherosclerosis, diabetes, metabolic function, and obesity). New heritable animal models generated in this program will be genetically mapped and made available to the scientific community for the cost of shipping. Details of this and all aspects of The Jackson Laboratory PGA are available at http://pga.jax.org/. Notably, 10 other PGA centers are operating in the United States. Each is generating unique resources for the scientific community. Details of these programs can be found at http://www.nhlbi.nih.gov/resources/pga/index.htm.


    ACKNOWLEDGMENTS
 
We thank Beverly Paigen and Carol Bult for critical review of the manuscript.

This research was supported by National Heart, Lung, and Blood Institute Program for Genomic Applications Award HL-66611.


    FOOTNOTES
 
Address for reprint requests and other correspondence: L. L. Peters, The Jackson Laboratory, 600 Main St., Bar Harbor, ME 04609 (E-mail: luanne{at}jax.org).

*This research paper was submitted in response to a Special Call for Papers in Large-Scale ENU Mouse Mutagenesis.

10.1152/physiolgenomics.00077.2002.


    References
 TOP
 ABSTRACT
 INTRODUCTION
 METHODS
 RESULTS
 DISCUSSION
 References
 

  1. Clark AT, Goldowitz D, Takahashi JS, Vitaterna MH, Siepkas SM, Peters LL, Frankel WN, Carlson GA, Nadeau J, and Justice MJ. A summary of ENU mutagenesis activities in North America. Genetica In press.
  2. Denis C, Methia N, Frennette PS, Rayburn H, Ullman-Cullere M, Hynes RO, and Wagner DD. A mouse model of severe von Willebrand disease: defects in hemostasis and thrombosis. Proc Natl Acad Sci USA 95: 9524–9529, 1998.[Abstract/Free Full Text]
  3. Fleming MD, Trenor CC, Su MA, Foernzler D, Beier DR, Dietrich WF, and Andrews NC. Microcytic anaemia mice have a mutation in Nramp2, a candidate iron transporter gene. Nat Genet 16: 383–386, 1997.[ISI][Medline]
  4. Jalbert LR, Rosen ED, Moons L, Chan JCY, Carmeliet P, Collen D, and Castellino FJ. Inactivation of the gene for anticoagulant protein C causes lethal perinatal consumptive coagulopathy in mice. J Clin Invest 102: 1481–1488, 1998.[Abstract/Free Full Text]
  5. Moon BC and Friedman JM. The molecular basis of the obese mutation in ob2J mice. Genomics 42: 152–156, 1997.[ISI][Medline]
  6. Paigen K and Eppig JT. A mouse phenome project. Mamm Genome 11: 715–717, 2000.[ISI][Medline]
  7. Paigen K. Understanding the human condition: experimental strategies in mammalian genetics. ILAR J 43: 123–135, 2002.[Medline]
  8. Peters LL and Barker JE. Hematology, clotting, and thrombosis (accession no. MPD:62). Mouse Phenome Database Web Site [Online]. Bar Harbor, ME: The Jackson Laboratory, 2001. http://www.jax.org/phenome.
  9. Sugiyama F, Churchill GA, Higgins DC, Johns C, Makaritsis KP, Gavras H, and Paigen B. Concordance of murine quantitative trait loci for salt-induced hypertension with rat and human loci. Genomics 71: 70–77, 2001.[ISI][Medline]
  10. Tsakiris DA, Scudder L, Hodivala Dilke K, Hynes RO, and Coller BS. Hemostasis in the mouse (Mus musculus): a review. Thromb Haemost 81: 177–188, 1999.[ISI][Medline]
  11. Vulpe CD, Kuo YM, Murphy TL, Cowley L, Askwith C, Libina N, Gitschier J, and Anderson GJ. Hephaestin, a ceruloplasmin homologue implicated in intestinal iron transport, is defective in the sla mouse. Nat Genet 21: 195–199, 1999.[ISI][Medline]