| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Gamete Biology |
Institute for Biogenesis Research, Department of Anatomy and Reproductive Biology, John A. Burns School of Medicine, University of Hawaii, Honolulu, Hawaii 96822
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
gamete biology, sperm, testis
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Despite its compact nature, sperm chromatin is susceptible to DNA damage. Numerous studies have demonstrated the presence of DNA fragmentation in spermatozoa by the TUNEL (terminal deoxynucleotidyl transferase) assay [9–12]. There was a positive correlation between DNA strand breaks and sperm chromatin with defective protamine packaging. Sperm DNA has endogenous nicks through various stages of spermiogenesis in mice and rats [12, 13]. The endogenous nicks are found during the transition from round to elongated spermatids in the testis, the stage that precedes the protamination of maturing spermatozoa, although no nicks were found once protamination was completed. McPherson and Longo [13] postulated that the reason that chromatin packaging may need an endogenous nuclease is to create nicks that will facilitate the unwinding of the DNA during the addition of protamines. They proposed that topoisomerase was involved because it is endogenous and is capable of both creating nicks to relieve torsional stress during chromatin rearrangement and ligating nicks once the displacement of histones by protamines has occurred. Spadafora and colleagues [14, 15] have suggested that mammalian spermatozoa contain endogenous calcium-dependent nucleases that cause degradation of DNA at hypersensitive sites under certain conditions.
In this study, we tested a model of sperm chromatin structure that we have proposed, termed the donut-loop model (Fig. 1) [8, 16]. Our model predicts that each DNA-loop domain is one protamine-bound toroid. It also proposes that there are DNase-sensitive toroid-linker regions between each toroid, and that these toroid-linker regions are the sites of DNA-loop-domain attachment to the sperm nuclear matrix. We also tested one potential functional consequence of this model, that the DNase-sensitive toroid-linker regions might be a site of endogenous sperm nuclease activity. Analysis of our data supports the existence of such a mechanism in mammalian spermatozoa.
|
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
In a typical experiment, three golden Syrian hamster retired breeder males (Charles River Laboratories, Wilmington, MA) were killed by asphyxiation with carbon dioxide, and the six caudae were dissected. The mature spermatozoa from the epididymal caudae were teased out into 5 ml of ice cold 1x PBS (Biowhittaker, Inc., Walkersville, MD). The hamster spleens were also dissected, cut into smaller sections, and put into another container with 5 ml of ice cold 1x PBS. The sperm cells were counted by using a hemocytometer or counting chamber with ethidium bromide stain under 40x magnification. The suspension was made of 0.25% Triton X-100 (TX) and 20 mM MgCl2 and aliquoted into 2-ml Eppendorf tubes. The cells were then treated with various conditions, as described in the Results. DNase I (Sigma Aldrich, St. Louis, MO) was added to various concentrations from 0 to 300 µg/ml and the tubes incubated at 37°C for 1–20 h. After incubation, the DNase I-treated samples had their nuclease activity deactivated by the addition of 0.5 M EDTA. Dithiothreitol (DTT) was added to some experiments to 20 mM. For measuring the effect of no detergent, spermatozoa were suspended as above but without TX 100 or MgCl2. Spleen cells were homogenized in 1x PBS and counted. They were also made to 0.25% Triton X-100, 20 mM MgCl2 and were incubated with various concentrations of DNase I. All glassware and buffers, when possible, were autoclaved before use.
After the incubation period, the samples were mixed with equivalent amounts of 1% pulse field gel agarose and plugs were made. These pulse-field gel electrophoresis (PFGE) plugs were incubated overnight in lysis buffer adapted from Korzik et al. [17] (10 mM Tris-HCl, pH 8.0, 10 mM EDTA, 100 mM NaCl, 20 mM DTT, 2% SDS, 20 µg/L of proteinase K) and incubated at 53°C overnight. After the lysis buffer incubation, the plugs were washed for 20 min minimum three times in equivalent volume TE buffer (10 mM Tris-HCl, 0.1 mM EDTA, pH 8) and 1 M glycine. Then they were washed with TE buffer alone two additional times. The plugs were stored at 4°C in TE buffer.
Pulse Field Gel Electrophoresis
About 100 ml of a 1% PFGE agarose gel (Pulsed Field Certified Agarose; Bio-Rad Laboratories, Hercules, CA) in TBE (45 mM Tris-borate, 1 mM EDTA) were poured into the PFGE 14 cm x 13 cm casting stand and mold after it had been tempered at around 55°C in a water bath for 10 min. The individual sample plugs were added to the 15-well gel and then an additional 1% PFGE agarose was used to seal the plugs in the well. Two different molecular weight markers were used: a high molecular weight ladder made up of polymers of the 48.5-kb lambda phage genome (size range 48.5 kb–1018.5 kb in increments of 48.5 kb), and a low molecular weight ladder consisting of the HindIII digested lambda phage genome (2 kb to 23 kb) with the lambda phage genome polymers to about 194 kb. These markers are provided in a syringe tube embedded in 1% agarose from New England Biolabs (Beverly, MA). When used in the gel, a small section of this agarose is squeezed out and embedded in the well, as described above.
The agarose gel with the standard casting platform frame and platform was placed into the Biorad PFGE Chef Mapper electrophoresis chamber and approximately 2.2 L of 0.5x autoclaved TBE was added. The temperature was allowed to equilibrate to 14°C and the pump flow was set at 70 (about 1 L/min). The additional conditions for running the gel were as follows: linear ramping, 27 h, 12 min total run time, 4 V/cm voltage, 120 degree angle, range 50 kb to 1 mb; switch time was initial 6.75 sec and final of 33.69 sec. The 1% agarose gel was then stained with ethidium bromide and a Kodak EDAS 290 gel documentation picture was taken under ultraviolet light. Gels were analyzed by the Kodak ID Image Analysis, Version 3.4 (Eastman Kodak, Rochester, NY). Image analysis was performed directly on the image obtained from the camera in the Kodak Gel Doc system. Each PFGE lane was scanned, and the molecular size in base pairs (bp) of the peak was calculated by the software using the two molecular weight markers as standards.
Halo Assay
Hamster spermatozoa treated as above with TX or DNase I were suspended in 10 ml of 0.5% SDS, 50 mM Tris-HCl, pH 7.5 and spun at 2000 rpm for 10 min in a Sorvall Biofuge pico-microcentrifuge. The pellets were resuspended in 15 ml falcon tubes in 15 ml of 50 mM Tris-HCl, pH 7.5, and then spun at 2000 rpm x 10 min in a Sorvall RC 26 Plus centrifuge in an HS-4 swinging bucket. The pellets were resuspended in Eppendorf tubes with 200 µl of 2 M NaCl, 50 mM Tris, and then 10 mM DTT was added. The Eppendorf tubes were incubated in warm water for 5 min and the samples were stained with 100 µg/ml ethidium bromide and examined under a microscope with ultraviolet light.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Our donut-loop model for sperm chromatin structure predicted that DNase treatment of sperm nuclei would cleave the DNA into loop-sized fragments. Implicit in this model is the suggestion that protamine-bound DNA is more resistant to DNase than other types of eucaryotic chromatin. We had previously noticed that sperm DNA could not be degraded completely by DNase I without extracting the protamines, but we needed to compare directly the conditions we would use in this study with somatic cells. We therefore first tested the DNA degradation patterns of spleen cells, in which the DNA is bound to histones, not protamines, under the conditions used in this study. We washed spleen cells with TX and then treated them with increasing concentrations of DNase I, from 0 to 300 µg/ml, for 1 h. The cells were then embedded in agarose and electrophoresed by PFGE (Fig. 2A). We found that, by 1 h, with a small amount of DNase (3 µg/ml), the spleen chromatin was digested into large fragments visible by PFGE (Fig. 2A, lane 2). However, using higher concentrations of DNase for 1 h digested virtually all the spleen DNA (lanes 5–6). Lane 1 shows the classical compression zone that is seen when untreated eucaryotic somatic cell DNA is electrophoresed by PFGE [18].
|
We then treated TX-washed sperm nuclei with the same concentration range of DNase I for 1 h (Fig. 2B). With no DNase I, all the sperm DNA remained in the PFGE well and did not form a compression zone (Fig. 2B, lane 1), as did the somatic nuclei (Fig. 2A, lane 1). We suggest that the lack of a compression zone in the sperm control is due to the greater stability and compactness of protamine-bound DNA in sperm than histone-bound somatic chromatin. In the presence of DNase I, sperm chromatin was digested to sizes that ranged just below the 48.5 kb lambda marker, but no further, even at the highest concentrations (Fig. 2B). When the same concentrations of DNase I were used to digest sperm nuclei for longer periods, all samples eventually reached the same sized fragments (Fig. 2, C and D). Even after overnight digestion in 300 µg/ml DNase I, there was a significant amount of DNA at this size range in sperm chromatin (Fig. 2D, lane 6).
These gels were quantitated using the Kodak ID Image Analysis software to determine the average size of the digested fragments. The average size of the digested DNA was determined by using the program to calculate the molecular size of the peak when the entire lane was scanned. The densitometric scans of PFGE gels of spermatozoa treated with various DNase I concentrations for 1 h (e.g., all the lanes in Fig. 2B) are shown in Figure 3 as an example. For each gel, a clear peak in the intensity was seen. Identical densitometric scans were obtained for all the gels in Figure 2. The molecular sizes of the peaks for each lane were plotted in Figure 4. Spleen chromatin was digested to very large molecular sized DNA (145 kb) at the lowest DNase concentrations, then to roughly sizes in the range of 20–30 kb, then digested completely at the highest DNase concentration. Sperm DNA was digested to sizes that ranged from 43.1 ± 20 kb (mean ± SD, all DNase concentrations averaged together) at the 1-h time point to 35.9 ± 2 kb at the 20-h time point (Fig. 4).
|
|
The results of these experiments suggested that protamine-bound DNA is much less sensitive to DNase I than histone-bound chromatin. They also confirmed the prediction of our donut-loop model that DNase I digestion of sperm chromatin would digest the DNA into fragments with a size similar to that predicted for protamine toroids. The fact that the average size of the DNase I-digested sperm chromatin was smaller than 50 kb suggests that either the toroids are actually smaller or the ends of the toroids are partially digested.
Spermatozoa Contain an Endogenous Nuclease Activity That Also Releases Loop-Sized DNA Fragments
We next tested for the possible presence of an endogenous nuclease that could be activated by treatment with TX and MgCl2. Spermatozoa were washed with 0.25% TX and made to 20 mM MgCl2, then incubated for various times at 37°C. A control aliquot of spermatozoa were not washed with detergent and were incubated in PBS for the same time points. We found that, without TX and MgCl2, the sperm DNA was stable for up to 5 days, but with TX and MgCl2, the sperm DNA was degraded to the same averaged sized fragments that we found with DNase I treatment (Fig. 5). Sperm DNA degradation was not noted in TX-treated spermatozoa until after at least overnight incubation. After 5 days of incubation, the sperm chromatin was not digested further. Image analysis of these gels indicated that the average size of DNA released by DNase treatment was 70.1 ± 9 kb, slightly larger than the DNase I-treated sperm chromatin. This is larger than the fragments released by DNase. Finally, EDTA was found to prevent the fragmentation of sperm DNA by TX treatment (Fig. 6).
|
|
These data suggest that sperm nuclei contain a mechanism that allows them to digest their own DNA under certain conditions to loop-sized fragments. This digestion does not appear to continue past the 70-kb size range.
DNase I Treatment Degrades Sperm Chromatin More Quickly Than the Endogenous Nuclease
We next compared the two treatments by performing a time-course experiment with the endogenous nuclease activated by TX alone versus with DNase I treatment (Fig. 7). The samples were electrophoresed by PFGE after dividing them from the same pool of hamster spermatozoa. The digestion of TX-treated sperm DNA, as seen before in Figure 5, was not complete until overnight incubation, though in this case, there was some digestion visible at 4 h. The digestion of sperm DNA by DNase was apparent by as early as 15 min. After overnight digestion as well as 1–5 days, the digestion of sperm DNA by DNase I resulted in DNA fragments with an average peak size of 55.0 ± 9 kb. The data demonstrate that exogenous DNase treatment works at a faster rate than the endogenous TX-activated digestion mechanism.
|
The effect of DNase I treatment is compared with TX treatment in Figure 8. The average size of the DNA released is smaller in DNase I-treated sperm chromatin, but both treatments result in most of the sperm DNA being digested no further.
|
Dithiothreitol Does Not Increase the Rate of the Endogenous Sperm Nuclease Digestion
Prior work in our laboratory had demonstrated that sperm treated with TX in the presence of DTT for 15 min could cause chromosomal breakage in the paternal chromosomes after intracytoplasmic sperm injection (ICSI) [19]. We tested whether DTT also had an effect on the degradation of sperm chromatin into loop-sized fragments, which might be initial steps in the chromosomal breakage. We treated spermatozoa with TX for 0 h, 4 h, and overnight or with TX with DTT for several hours and overnight. In both cases, only when the treatment had reached the overnight stage did the DNA begin to degrade into loop-sized fragments (Fig. 9). There appeared to be no difference in increasing the rate of endogenous sperm DNA degradation by the addition of DTT.
|
Visualization of Release of DNA-Loop Domains by DNase
Our model predicted that, when sperm nuclei were treated with DNase I, the chromatin would be digested at the sites of DNA attachment to the nuclear matrix, or MARs (matrix attachment regions; Fig. 1). We tested this hypothesis by viewing DNA-loop domains directly under a microscope by halo assays. Spermatozoa that were untreated or treated with DNase I or treated with TX overnight were treated with high salt and DTT to remove the protamines and were stained with ethidium bromide to reveal DNA. Untreated spermatozoa revealed the characteristic hook-shaped hamster sperm nucleus with a visible halo around it (Fig. 10A). This halo is made up of DNA-loop domains attached at their bases to the nuclear matrix. Spermatozoa treated with a low (3 µg/ml) or high (300 µg/ml) concentration of DNase I had no halos at all and the sperm nuclear matrices were condensed (Fig. 10, B–C). Sperm nuclei treated with both concentrations of DNase I resulted in DNA being degraded to the 30–40 kb loop-sized DNA (Fig. 2B). Therefore, these data support our model's predictions that DNase I treatment will release DNA from the sperm matrix by cutting DNA at the loop attachment sites. When we visualized TX-treated sperm, we noticed that most of the DNA remained associated with the nuclear matrix. The nuclear matrices, themselves, appeared slightly distorted, probably as a result of overnight incubation at 37°C.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
DNase Treatment and the Halo Assay Experiments Support the Donut-Loop Model for Sperm Chromatin Structure
Our donut-loop model for sperm chromatin structure suggests that there are DNase I-sensitive toroid-linker regions between the highly stable protamine-bound toroids in sperm chromatin, and our data support this model. Sperm chromatin can be digested into fragments with an average size of 30–55 kb with DNase I, but this DNA is not digested farther into smaller pieces. The variability of the average size of the digested sperm DNA fragments is probably the result of the different experimental conditions used and of the nonspecific nature of DNase I. It also suggests that sperm chromatin contains a limited amount of DNase-sensitive areas. Furthermore, even though the DNA is digested to loop-sized fragments, there is no DNA remaining attached to the sperm nuclear matrices. These data support two predictions of the donut-loop model, that there exist DNase-sensitive regions distributed from each other by 30–55 kb and that these regions are the sites of attachment to the nuclear matrix (Fig. 11). What has yet to be conclusively demonstrated is that one protamine toroid is one loop domain. The DNA fragmentation in these experiments is the same average size as a protamine toroid, but this does not yet prove the model.
|
Another point our data makes is that protamine-bound DNA is much less sensitive to DNase digestion than histone-bound chromatin. Spleen chromatin was digested to completion within 1 h of treatment with 300 µg/ml DNase I, but sperm DNA was still resistant after overnight treatment with this same concentration. This might have been predicted based on how tightly protamines bind to DNA. Finally, the most sensitive region in both sperm and spleen nuclei may be the MARs. After 1 h digestion with the lowest concentration of DNase I, spleen DNA was digested to an average size of 145 kb, then it plateaued between 34 and 22 kb (Fig. 4). This was finally completely degraded, but it suggests that the MARs may be the most sensitive chromatin domain to DNase I digestion.
Sperm Nuclei Have a Mechanism for Chromosomal Degradation
Our data also support the idea that sperm nuclei have an endogenous nuclease activity that can be activated by TX treatment. This suggested to us that it is necessary for the sperm membranes to be compromised by detergent washing in order for the DNA to be degraded into loop-sized fragments. Several laboratories have reported DNA damage in ejaculated human sperm samples [10, 20, 21]. One group has suggested the existence of a sperm-specific nuclease in mouse spermatozoa that is activated by exogenous DNA [15]. Our data suggest that sperm nuclei contain a mechanism to digest their own DNA into loop-sized fragments.
This endogenous digestion of sperm DNA by TX treatment differed from that of exogenously added DNase I in that the DNA loops remained associated with the nuclear matrix, even though they were cleaved into loop-sized fragments. There are two possible explanations for this apparent discrepancy. The first is that, in the TX-treated spermatozoa, the DNA is cleaved somewhere within the DNA-loop domain away from the MARs. This would result in the DNA remaining attached to the nuclear matrix, even though the DNA was cleaved. We view this as unlikely because our data demonstrate that protamine-bound DNA is less sensitive to DNase I than is histone-bound chromatin. The second is that, in the TX-treated spermatozoa, the DNA is cleaved near the MARs but at a specific site that does not digest the entire MAR. This is supported by the fact that the average size of the sperm DNA fragments resulting from TX treatment was larger, about 70 kb, than that of DNase I-treated spermatozoa, from 30 to 55 kb, suggesting that DNase I digested the toroid-linker regions much more completely than did the endogenous nuclease. The degradation of sperm chromatin into loop-sized fragments is similar to a mechanism that exists in somatic cells. At least three different laboratories have shown independently that, when somatic cells undergo apoptosis, they first degrade their DNA into loop-sized fragments with an average size of 50 kb [18, 22, 23]. All three laboratories also suggested that the nuclease was topoisomerase II (Topo II). We are currently investigating this possibility.
While it is possible that the TX-mediated DNA cleavage was an artifact of nucleases from contaminating bacteria, we do not believe this to be the case for several reasons. First, intact spermatozoa were incubated for several days without DNA degradation, and it is likely that bacterial contamination would have affected these as well. Second, all glassware and all buffers possible were autoclaved to prevent such a contamination. Furthermore, as noted above, the DNase I-digested sperm nuclei resulted in DNA-loop domains being released by the halo assay, while the TX-mediated sperm DNA digestion did not. Finally, the fragments released by TX treatment were larger than those released by DNase I. Thus, in these two respects, the TX-mediated DNA degradation did not behave like contaminating DNase.
It is reasonable to predict that spermatozoa would have some sort of mechanism for preventing paternal transmission of damaged DNA to the embryo during fertilization. That is, if a sperm cell encountered a potentially DNA-damaging environment during its transit in fertilization, it may be able to respond by a sperm-specific type of suicide, by digesting its DNA into loop-sized DNA that would be incapable of fertilization.
Relationship of Loop-Sized DNA Breaks to Paternal Chromosome Breakage in Zygotes
Recent experiments from our laboratory have characterized another type of sperm DNA degradation, the breakage of paternal chromosomes in the one-cell zygote after various sperm treatments before ICSI. We demonstrated that, when spermatozoa were treated with TX and DTT for 15 min and then used for ICSI, the paternal chromosomes of the resulting embryo were severely degraded [19]. Neither reagent alone produced this effect. We believe that these two forms of sperm DNA degradation, the paternal chromosome breakage reported earlier and the sperm DNA fragmentation reported here, are related, and we are currently testing this hypothesis. There is an apparent discrepancy between these two forms of DNA degradation in that paternal chromosome breaks could not be caused by TX alone, while the DNA fragmentation reported here can. However, this may only be related to time because, in the previous experiments, spermatozoa were only incubated for 15 min in TX alone, while in these experiments, the spermatozoa were incubated overnight.
Conclusions
We provided evidence that is consistent with, but does not yet prove, our donut-loop model for sperm chromatin structure. We also demonstrated that sperm contains a mechanism for degradation of its chromatin into loop-sized fragments, similar to those seen in apoptotic somatic cells. The data together suggest that the model for sperm chromatin organization that we have proposed may have functional significance in that it predicts a mechanism for the sperm DNA fragmentation we have identified.
|
FOOTNOTES |
|---|
2 Correspondence: Barbara Sotolongo, Institute for Biogenesis Research, John A. Burns School of Medicine, University of Hawaii, 1960 East-West Rd., Honolulu, HI 96822. FAX: 808 956 7316; sotolong{at}hawaii.edu
Received: 21 June 2003.
First decision: 16 July 2003.
Accepted: 18 August 2003.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  Y. Yamauchi, A. Ajduk, J. M Riel, and M. A Ward Ejaculated and Epididymal Mouse Spermatozoa Are Different in Their Susceptibility to Nuclease-Dependent DNA Damage and in Their Nuclease Activity Biol Reprod, October 1, 2007; 77(4): 636 - 647. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Y. Yamauchi, J. A Shaman, S. M Boaz, and W. S. Ward Paternal Pronuclear DNA Degradation Is Functionally Linked to DNA Replication in Mouse Oocytes Biol Reprod, September 1, 2007; 77(3): 407 - 415. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Y. Yamauchi, J. A. Shaman, and W. S. Ward Topoisomerase II-Mediated Breaks in Spermatozoa Cause the Specific Degradation of Paternal DNA in Fertilized Oocytes Biol Reprod, April 1, 2007; 76(4): 666 - 672. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. Miller and G.C. Ostermeier Towards a better understanding of RNA carriage by ejaculate spermatozoa Hum. Reprod. Update, November 1, 2006; 12(6): 757 - 767. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. A. Shaman, R. Prisztoka, and W. S. Ward Topoisomerase IIB and an Extracellular Nuclease Interact to Digest Sperm DNA in an Apoptotic-Like Manner Biol Reprod, November 1, 2006; 75(5): 741 - 748. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. Oliva Protamines and male infertility Hum. Reprod. Update, July 1, 2006; 12(4): 417 - 435. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 O. Mudrak, N. Tomilin, and A. Zalensky Chromosome architecture in the decondensing human sperm nucleus J. Cell Sci., October 1, 2005; 118(19): 4541 - 4550. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B. Sotolongo, T. T. F. Huang, E. Isenberger, and W. S. Ward An Endogenous Nuclease in Hamster, Mouse, and Human Spermatozoa Cleaves DNA into Loop-Sized Fragments J Androl, March 1, 2005; 26(2): 272 - 280. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. Jones Sperm Survival Versus Degradation in the Mammalian Epididymis: A Hypothesis Biol Reprod, November 1, 2004; 71(5): 1405 - 1411. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
, Lambda phage molecular weight markers in 48.5-kb increments; MW, low molecular weight markers
