Thursday, 13 September 2012

Sex- Determination of the Domestic Chicken (Gallus Gallus)

Sex-Determination of the Domestic Chicken

(Gallus Gallus)
 
Abstract 

DNA was extracted from three sample types, destructive tissue sampling, invasive blood sampling and non-invasive feather sampling to determine the quality and quantity of DNA yields between the three types. DNA was then used to determine the sex of each sample.

 
Introduction

When working with wild animals it is often hard to collect and extract DNA therefore, non-invasive methods of collection are required. DNA extraction from animals fall into three categories which are destructive, non-destructive (invasive) and non-invasive. Destructive sampling requires the organism to be killed to be able to obtain the tissues necessary for genetic analysis. This technique is often used when identifying specific enzymes or metabolic requirements.  Non-destructive or invasive sampling requires capture of the organism in order to take a tissue biopsy or blood sample. Non-invasive techniques are the most comfortable and humane way for the animal’s DNA to be extracted and the technique is beneficial when the organism is endangered or hard to capture. Non-invasive sampling does not disturb the animal’s habitat and often DNA that has been left behind in nests such as feathers, skins, bones, eggshells, scats and swabs etc. can be used. Limitations to non-invasive methods have shown that DNA samples degrade overtime which becomes troublesome in analysis. Studies have also shown the amount of DNA extracted is much lower than invasive or destructive means of DNA extraction therefore more research is required to improve the current techniques to increase extracted DNA quality.

Good quality DNA visualisation is a key point to determining whether an experiment was successful or requires more research therefore it is important to reduce variables that may affect results. During DNA extraction the quality of DNA may be affected by the technique used for extraction, ensuring an uncontaminated workspace is available, how the DNA was preserved and the time period between extraction and use as DNA as it degrades over time.  

Storage of DNA is an important factor to ensure the results are conclusive. Freeland (2005) discusses various methods of preservation which are available including expensive freezing with liquid nitrogen or dry ice, use of a deep freezer (<-80°C) or a fridge, or keeping the DNA at room temperature. These later methods have shown that extracted DNA degrades quicker than liquid nitrogen freezing but may be useful when extracting DNA in the field. For optimal results it is best that the time between extraction and analysis is kept to a minimum. Another method of preservation which is useful in the field is the use of ethanol which is useful for DNA preservation when a lab is not available. Colton and Clark (2001) tested various storage conditions over a 21 month period showing the importance of freezing the samples, preferably in ethanol. Freezing the DNA samples at -20°C were used during our experiment therefore DNA degradation was not a variable factor between the samples.

Different DNA extraction and analysis methods use varying chemicals, reagents, buffers, salts or commercial kits therefore to reduce variables it is crucial to check the use by dates throughout the procedure as reagents that have been sitting around for long periods of time affect pH, chemical bonding and structures giving you inconclusive results or lower quality DNA samples.

Good quality DNA extraction displays clear and bright distinct bands on the electrophoresis gel whereas low quality DNA presents a more smeared appearance. DNA determination of gender shows as one or two distinct bands. In mammals the male determines the sex of the offspring and has the heterogametic XY genotype which presents as two distinct bands on a gel while the female has the homogametic XX genotype and displays one distinct band on a gel. In birds the genotypes are essentially opposite to mammals, the male has the homogametic XX genotype showing one distinct band while comparatively the female determines the sex of the offspring and has the heterogametic XY genotype. To save confusion avian chromosomes have been classified in the ZW sex determination system where males have the ZZ genotype and females have the ZW genotype.

Sex determination in chickens is harder than mammals as males and females are monomorphic, meaning they are not phenotypically different in colour, body size or feather type etc. The following experiment followed procedures outlined by Hogan, Loke & Sherman (2012) aiming to extract a domestic chicken’s (Gallus Gallus) DNA from three tissue types to determine the sex and to compare the quality and quantity of DNA from the three samples. DNA was extracted using a commercial DNA purification kit; DNeasy Blood & Tissue Kit produced by Qiagen (2012) which is an expensive procedure however provides fast, easy and high yield silica-based DNA purification utilizing Proteinase K which reduces mechanical and contamination variable risks. The extracted DNA was amplified using PCR to intensify the sex-specific locus CHD1W and CHD1Z, enabling us to determine the gender of the bird using a set of universal avian sexing primers 2250F and 2718R developed by Fridolfsson and Ellegren (1999). Individual results were correlated and compared to the class while molecular markers, negative, male and female controls were used to determine whether the hypothesis was supported that the more destructive and invasive techniques of the tissue and blood samples would extract a higher DNA quality compared to the non-invasive technique of sampling from a feather.  

On completion of the procedure it was determined that Destructive muscle sampling produced a higher yield of DNA compared to Invasive Blood sampling with the lowest DNA yield for non-invasive feather sampling. Horvath, Martinez-Cruz, Negro and Godoy (2005) have improved the extraction of DNA from the feather tip by extracting blood from a blood clot located in the superior umbilicus instead of the basal tip of the calamus which was used in the author’s experiment. Their results indicate clot samples yielded a higher amount of DNA than tip samples and less DNA degradation over time. This procedure could increase the quantity and quality of DNA extracted in feather samples in future experiments.

 Methods

DNA Extraction from Blood, Feather and Muscle Samples

DNA extraction methods compromise of cell lysis which causes a disruption in the cell membrane which allows proteins to be freed from the cell. A protease may be added to remove the proteins and RNA removed with the assistance of an RNAse. The DNA is then precipitated with an alcohol, then amplified using PCR techniques and visualised using electrophoresis.

Following Hogan, Loke and Sherman (2012) in depth procedure all materials and reagents were organised. A sterilized razorblade was used in a sterile petri dish to reduce the muscle tissue down to form a smooth consistency alternatively the superior end of the calamus of the feather was sliced or a hole-punch sized piece cut out of the blood card. The samples were then placed into a sterile microcentrifuge tube. All sharps were safely discarded at this point. PBS (Phosphate Buffered Saline) and RNAse A was added to the blood sample. Buffer ATL was added to the feather and muscle specimens and all samples vortexed. Proteinase K was mixed through the specimens. The blood sample was left to incubate for 30minutes at room temperature then all specimens were incubated for 30 minutes at 56°C. During incubation the muscle sample was vortexed every 10 minutes to aid in dispersion while 200µL of Buffer AL was added to the blood sample. After incubation, RNAse was added to the muscle sample, vortexed then incubated for a further 2min at room temperature and vortexed again. The cells were lysed by this stage.

Buffer AL was added to the muscle and feather samples, vortexed then 200 µL of ethanol was added to all samples which were vortexed again. 20µL of ethanol was added to the blood sample only allowing the solution to be properly vortexed and become a homogenous solution allowing the DNA to precipitate and be ready to be membrane bound.

The precipitated samples were pipetted into a labelled DNeasy Spin Column/collection tube (Qiagen 2012) and centrifuged at 6000 x g (8000rpm) for 1 minute. The guanidine thiocyanate waste which remained in the collection tube was discarded and a new collection tube attached. Buffer AW1 was added to the muscle and feather samples. All samples were centrifuged for 1 minute at 6000 x g (8000rpm) then Buffer AW2 was added and again centrifuged for 3 minutes at maximum speed (13-14,000rpm) to dry the DNeasy membrane. The DNeasy mini spin column was carefully removed so it did not come into contact with the ethanol and guanidine thiocyanate waste and the collection tube replaced with a new one. The sample was finally centrifuged for 1 minute at maximum speed to remove any residual ethanol while the DNA was ready to be eluted from the membrane.

The collection tube was discarded and the DNeasy mini spin column was placed into a labelled, clean, 1.5mL microcentrifuge tube. 100µL of Buffer AE was pipetted directly onto the DNeasy membrane for the blood sample while the feather and muscle samples had Buffer AE added in two 50µL steps the samples were incubated at room temperature for 1 minute then centrifuged for a further minute at 6000 x g (8000rpm) to elute. The elutant was kept aside. A further 50 µL of Buffer AE was added to the membrane, incubated at room temperature for 1 minute then centrifuged at 6000 x g(8000rpm). This elutant was also kept aside. The two elutants were combined measuring 200 µL of DNA. The column was discarded and DNA stored at -20°C until required.

Electrophoresis Method
During electrophoresis negatively charged DNA molecules migrated towards the positive cathode causing smaller protein fragments to move quicker through the gel matrix than larger molecules. The DNA was visualized as fragments in bands on the gel which were stained with an intercalating agent, mutagen and suspected carcinogen, Ethidium Bromide.

An agarose gel and TAE buffer was pre-prepared in the microwave by boiling and allowing the gel to cool to 50 °C in a water bath. Using gloves Ethidium Bromide was added to 150mL of gel to get a final concentration of 0.5µL mL-1. The casting tray was placed into the gel tank with black casting gates at each end. The gel was poured into the tray and comb inserted allowing 30 minutes for the gel to set at room temperature.
10µL of the DNA avian sample extracted from the muscle earlier was mixed with the 6x loading dye in a clean microfuge tube. Wearing gloves the black casting plates and comb was removed and TAE buffer was added until the whole gel was covered by about 5mm.

The molecular weight markers ʎ HindIII and 2-log ladder were added into the first and the last wells. The DNA samples were added into an empty well and the position recorded. The electrophoresis unit was covered and connected to the power unit and run at 120V for 60 minutes. The DNA proceeded to migrate from the negative anode (black cable) to the positive cathode (red cable). Once completed the gel tray was removed and transferred to a plastic container to use the Gel Doc System so the Gel photos could be recorded and visualized (See Figure 1).

PCR method

The Polymerase Chain Reaction (PCR) procedure was used to amplify the DNA into a measureable amount which could then be visualised using electrophoresis. PCR used a number of reagents to cut specific sites of DNA which was then purified and used for biological use. The PCR procedure consisted of cycles of heating and cooling the DNA to allow helix unwinding and binding. Each complete cycle consisted of the following: 1 cycle of 94°C for 5 minutes, 35 cycles of 94°C for 30 sec, 55°C for 30 sec and 72°C for 30 seconds & 1 cycle of 72°C for 5 minutes.

In order for PCR to be successfully carried out the DNA template needed to be within a range of concentrations and be freed of inhibitors. This was done by adding 10-100ng of DNA to a 50µL PCR product. If the band on the gel was very faint or not visible the DNA solution was not diluted, if the band was very bright the DNA was diluted by 1:5 with sterile MilliQ water.

The Mastermix, positive and negative controls were prepared as per instruction in Hogan, Loke & Sherman (2012) and 40µL of the PCR mastermix and 10µL of the DNA sample was mixed into a 0.2mL PCR tube. The prepared tubes were placed into a thermo-cycler and a program which was optimised to amplify the CHD1W and CHD1Z gene variants with the 2550F -2718R primer set.

Once the PCR product had been amplified the DNA was then run for 20 minutes at 300V on a 1% agarose gel in a 1 x SB electrophoresis buffer which had been boiled and allowed to cool to 50°C. Wearing gloves 15µL of 5mg mL-1 Ethidium Bromide stock solution was added to 150mL of SB gel to achieve a final concentration of 5µg mL-1. The DNA samples were prepared by mixing 10µL of PCR product with 4µL of 6x loading dye and pipetted into the gel alongside 5µL of a 100bp molecular marker. (See Figure 2)

The bands were again visualised using the Gel Doc System and PCR product sizes were determined by comparing the position of the PCR band to the molecular marker. (See Figure 2)

 Results

 

Figure 1: (Right click on image to open in new tab) Class results for DNA extraction for various DNA samples. DNA was extracted from blood, feather or muscle and visualised via electrophoresis.
 
DNA was successfully extracted from all tissue types; the muscle presented more distinctive bands, then blood while feather samples extracted a very minute amount of DNA which displayed very faint bands using electrophoresis. (See figure 1). As explained in Hogan, Loke & Sherman (2012), the samples DNA concentration were measured by referencing the comparable brightness of the sample to the 2-log MWM over 10µL of sample DNA (amount of DNA loaded into the well). Figure 1 represents the Gel electrophoresis from the DNA extraction using the commercial DNA purification kit (Qiagen 2012, DNeasy Blood & Tissue Kit)

In Figure 1, DNA can easily be visualized from muscle wells (2, 8, 10, 12, 15 & 16) and from Table 1 and 2 muscle samples had ranging DNA concentrations between 3.2-12.4ng/mL, and an average of 6ng/mL. Figure 1 shows DNA is moderately visualized in blood wells (3, 5, 7, 11 & 14) and from table 1 and 2 blood samples range of DNA concentrations lay between 0-4.5ng/mL with an average of 2.18ng/mL. Blood wells 3 and 14 had unknown concentrations as bands were not visualized. From Figure 1, Feather wells (4, 6, 9 & 13) bands were also moderately visualized while table 1 and 2 shows feather sample DNA concentrations ranged from 0-3.2ng/mL with an average of 2.4ng/mL. Well 6 concentrations were unknown. Some samples within each tissue type yielded a higher quality of DNA than others however on average muscle showed higher DNA quality to blood and feather.

 

Table 1: (Right click image to open up in new tab) shows DNA samples loaded in the wells, µL of sample loaded, referenced MWM of DNA in referenced band and final concentration of DNA (ng/mL)

 
Table 2: shows sample DNA concentration, tissue type and average concentration of tissue type.

Well
Concentration
(ng/mL)
Tissue
 type
Average
(ng/mL)
16
3.2
Muscle
6
8
4.5
15
4.5
2
5.7
12
5.7
10
12.4
 
 
 
 
6
0
Feather
2.4
4
3.2
9
3.2
13
3.2
 
 
 
 
3
0
Blood
2.18
14
0
5
3.2
7
3.2
11
4.5

 
Once DNA was extracted and visualized via gel electrophoresis the samples were diluted to yield similar concentrations. If the band was very faint or not visible it remained undiluted, most blood and muscle samples were diluted by various concentrations. Figure 2 represents the extracted diluted/undiluted DNA being amplified by sex specific primers and visualized using gel electrophoresis to determine the gender of the bird while table 3 represents each sample number, tissue type, sample type and well number from Figure 1 for ease of reference.  There were 7 male genotypes (ZZ) represented in the sample and can be seen with varying DNA qualities by one Z band of 600-700 base pairs in wells (referenced from figure 1); 5, 6, 7, 8, 9, 11 & 15. There were 7 female genotypes (ZW) represented in the sample which can be seen by two bands, one Z chromosome band of 600-700 base pairs and the other W chromosome band of 400-500 base pairs with varying DNA qualities in wells (referenced from figure 1) ; 2, 3, 4, 10, 13, 14 & 16.  The negative controls have been run in Well number 6A, 15A & 7B, male controls have been run in well number 7A, 13A, 17A & 6B, female controls have been run in well number 8A, 14A, 16A & 8B. In the majority of samples the muscle and blood bands appear on the gel as a much stronger and darker band than the feather samples.

In Figure 2, Wells 2-9A, 11A, 12-14A, 18-19A, 2-5B, 9-11B DNA was amplified and visualized successfully. Well 10A no DNA was extracted. Well 2B-8B show DNA samples, however controls were not consistent with others. Well 5B shows bottom band was more preferred than top band. Well 2B shows bands that are not comparable to other blood samples as they are very faint. All samples except 10B correlated with the actual sex of the individual, sample 10B was a known male however bands were not representative of this.

Figure 2: (Right click image to open up in new tab)Class results for various PCR product samples. DNA was amplified from DNA extraction and visualized via electrophoresis to determine sex of bird. Results indicate an error occurred in wells 1B-8B as controls are not as dark as what they were expected to be indicating an error with PCR mastermix. Despite the error the sexes were still able to be determined. PCR result matched expected result for all samples except well 10B which was a known male.

Table 3: wells corresponding to Figure 2, showing each sample number, tissue type, sample type and well number reference from Figure 1.

 

 
                                                                                        Discussion:          

Universal Primers 2550F and 2718R used in Fridolfsson and Ellegren (1999) were replicated in the author’s procedure. Fridolfsson and Ellegren (1999) developed the 2550F primer which binds to 5’-GTTACTGATTCGTCTACGAGA-3’ while the 2718R primer binds to 5’-ATTGAAATGATCCAGTGCTTG-3’ fragments of DNA which were used to amplify sex specific loci; CHD1W & CHD1Z to determine the gender of each sample taken. The advantages of employing universal sexing molecular markers are the ability of the CHD1W and CHD1Z loci to be easily visualized on a gel as they differ by 100-150 base pairs. According to Fridolfsson and Ellegren (1999) the CHD1W fragments varied between 400-450 base pairs compared to the CHD1Z fragments varying between 600-650 base pairs which were comparable to replicated results. The samples were then compared to known male, female and negative controls to view results and inconsistencies. Amplification of the female chromosomes displays one CHD1W and one CHD1Z fragments on a gel compared to the male chromosome of one CHD1Z fragment.

The authors experiment matches that of Fridolfsson and Ellegren (1999) except that a 1% agarose gel was used to visualize the DNA fragments via electrophoresis compared to Fridolfsson and Ellegren utilizing a 3% agarose gel. They also used a phenol extraction according to Manniatis et al. (1982) standard methods compared to our use of a commercial DNA purification kit. (Qiagen 2012). Despite the differences our results were comparable to Fridolfsson and Ellegren (1999).

The results show average sample DNA concentrations vary between tissue types. Possible errors for non-reliable results include pipetting errors, contamination, deviation from the method, a heterogeneous solution when extracting the DNA from the sample and varying DNA quality from samples. The results were also non representative between tissue types as the sample size was small allowing no reliable comparisons to be made between genders within specific tissue samples groups. Storage of DNA could also contribute to varying DNA qualities as there were 6 weeks between the initial extractions of DNA and recording the results from the PCR product. DNA was also stored at room temperature during the practical classes contributing to slight gradual DNA degradation even though ice was provided. This would not have affected the visualised results as all samples were exposed to the same conditions.

Results indicate various samples did not compare to the majority. From table 2, results indicate the average DNA concentration reflects muscle had the highest quality of extracted DNA compared to feather, then blood however the sample sizes were not equally distributed. The class contained 6 muscle samples, 4 feather samples and 5 blood samples. As 1 feather sample and 2 blood samples failed to extract any visible DNA (see figure 1) they bring the averages down. By omitting the non-visible results (as all samples were able to amplified by PCR so non visualized DNA could be due to lower than average DNA quality) the average sample DNA concentrations reflect muscle still had the highest quality of extracted DNA compared to blood then feather having the lowest DNA quality which correlated to expected outcomes.

The three samples that failed to be visualized on the agarose gel during the first round of electrophoresis following initial DNA extraction were Blood well 3 sample B12.23, blood well 14 sample B12.21 and feather well 6 sample F12.20 (Figure 1). Figure 2 shows blood well 3 sample B12.23 was loaded into well 11A where PCR was successfully able to amplify the minute amount of DNA which resulted in a female bird with a more preferred W chromosome which was represented by a darker bottom band (400-500bp) compared to the Z chromosome top band. (600-700bp)

Feather well 6 sample F12.20 from figure 1 shows DNA was able to be amplified to determine the sample was a male via PCR and visually represented in Well 4A in Figure 2. The Z chromosome bands are very faint therefore further PCR amplification would have been beneficial to increase the intensity of the band.

Blood well 14 sample B12.21 from figure 1 shows PCR was also able to successfully amplify the DNA sample but more amplification was required to intensify the bands as Figure 2 shows two very faint bands representing the female ZW genotype.

Other anomalies represented by the results include figure 2 displaying the female feather sample F12.19 bottom W chromosome band (400-500bp) being more preferred by the genotype than the Z chromosome band (600-700bp). This sample is also slightly better visualized than most other feather samples. This could be contributed to the amount of blood available in the tip or the thickness of the stalk.

In muscle sample M12.27 in well 16 from figure 1 and well 5B in figure 2 shows the DNA quality and two chromosome bands which were very faint compared to other muscle samples. The DNA extracted from the first electrophoresis displays the lowest DNA concentration of the 6 samples indicating that the amount of muscle may have been less or the muscle was not completely dissolved into a homogenous solution. In comparison the amount of DNA extracted from M12.27 was comparable to the amount of DNA extracted from the feather and blood samples allowing the DNA to be amplified during PCR and allowing the genotype to be determined. M12.27 banding patterns indicated in figure 2 show a female with a preferred W chromosome than the fainter Z chromosome.

The male, female and negative controls in Well 6B-8B (Figure 2) are not representative of other control samples. The preparation of the PCR Mastermix was done according to the procedure outlined in Hogan, Loke & Sherman (2012) and added to samples in Figure 2 represented in well 3B first then well 4B, 2B and 5B then controls. Due to bubbles forming in the PCR mastermix after addition to the first two samples, the adequate amount was not able to be added to the remaining micro tubes resulting in fainter or non-existent bands. The controls in Wells 6A-8A were used to compare samples loaded in 2B-5B and 9B-11B.

Well 9 (Figure 1) and Well 10B (Figure 2) represents a male feather sample F12.18. Table 1 indicates the most common amount of DNA was extracted from the sample and successfully visualized in Figure 1. Figure 2 displays multiple bands which are not representative of the known male genotype indicating the sample was contaminated.

The above user errors could be avoided in future experiments by adopting Taberlet et al’s (1996) procedure by introducing a multiple tube approach, this would allow lower DNA yields to be rejected in line with Morin et al. (2001) which although would be costly and require more reagents/kits it would eliminate various user errors and produce more reliable results.

The duration of the experiment was held over 6 weeks therefore storage of the DNA was considered. DNA can degrade overtime. Various methods of DNA preservation discussed by Freeland (2005) indicate the importance of preserving DNA to avoid DNA molecular arrangements being rearranged which can affect the results when the DNA is amplified by PCR. Freezing at -20°C was used to preserve the DNA samples in between each practical component. This storage type is one of the most successful according to Colton and Clark (2001). During each practical session the DNA was mostly stored at room temperature which may have contributed to some DNA degradation however this was unlikely to contribute to variations of DNA quality as all samples were exposed to the same environments. During the practical elements ice esky’s were provided to store the DNA samples however it was noted that all students kept the DNA at room temperature when the DNA was not being used, this could be avoided in future experiments by reminding the students of the importance of preserving the DNA for extraction quality.

The quality of DNA extracted varied between tissue samples however all sample types were able to be amplified utilizing PCR indicating non-invasive techniques of using DNA from a feather could be used to extract DNA when alternative methods are not suitable. Destructive muscle samples yielded the highest quality and quantity of DNA compared to the other techniques however destructive means of DNA extraction require the killing of the organism which is not always suitable especially when dealing with endangered species. Invasive blood sampling also produced a good yield and quality of DNA however not as high as tissue sampling but could be used when tissue samples are not available. Disadvantages of blood sampling include the requirement of capturing the organism to extract a blood sample and samples are affected by storage techniques as most samples are taken in the field. Non-invasive feather samples, although yielding a lower DNA quality than the other samples the DNA is easily obtainable in the field through capture and plucking of feathers of the animal (Mundy et al. 1997) or obtaining specimens from the ground or nests. (Beck, Double and Maurer 2010)

References

Cotton, L., and Clark, JB. (2001) Comparison of DNA isolation methods and storage conditions for successful amplification of Drosophila genes using PCR. Drosophila Information Service. 84, 180-182. Retrieved September, 11th 2012 from: www.ou.edu/journals/dis/DIS84/Tec8%20Colton/Colton.pdf

Freeland, J (2005). Molecular Ecology. Wiley. Chichester.

Fridolfsson, A,. and Ellegren, H. (1999) A simple and universal method for molecular sexing of non-ratite birds. Journal of Avian Biology. 30, 116-121

Hogan, F., Loke, S., and Sherman, C. (2012) SLE254 Genetics: Practical Manual 2012~ Sex Determination of the Domestic Chicken (Gallus Gallus). Deakin University. Burwood. 1-46.

Horvath, M., Martinez-Cruz, B., Negro, J., Kalmar, L., and Goday, J. (2005) An overlooked DNA source for non-invasive genetic analysis in birds. Journal of Avian Biology. 36, 84-88

Morin, PA., Chambers, KE., Boesch, C., and Vigilant, L. (2001). Quantitative Polymerase Chain Reaction Analysis of DNA from non-invasive samples for accurate microsatellite genotyping of Wild Chimpanzees. Molecular Ecology. 10, 1835-1844

Mundy, NI., Unitt, P., and Woodruff, DS. (1997). Skin from feet of museum specimens as a non-destructive source of DNA for avian genotyping. Auk 114, 126-129.

Qiagen. (2012). Sample & Assay Technologies: DNeasy Blood & Tissue Kit. Retrieved September, 11th 2012from: http://www.qiagen.com/products/genomicdnastabilizationpurification/dneasytissuesystem/dneasybloodtissuekit.aspx

Taberlet, P., Griffin, S., Goosens, B., Questiau, S., Monceau, V., Escaravage, N., Waits, LP., and Bouvet, J. (1996) Reliable Genotyping of Samples with very low DNA quantities using PCR. Nucleic Acids Research. 24, 3189-94

No comments:

Post a Comment