HELP HELP HELP needed with RNA extraction

Tue Jun 11 10:53:37 EST 1996

Jim Graham wrote:

> Rimantas has suggested that there may be actual data
> relevant to "finger RNAses". Can anyone point out any
> such references?
> Apparently, my own assay, which consited of tipping a
> microcentrifuge tube against my finger tip and shaking it
> -removing 10ul to a tube containing 2 ug of RNA markers
> and incubating 30 min at 37C (which was negative for me
> and a collegue) may not have been sensitive enough to
> detect the proposed low levesl of RNAse acrtivity.
> Any one else care to try? Rimanats suggest putting
> 20ul water on a fingertip directly and testing that for RNAse
> action on an RNA standard. (I'll do it shortly too).
> Jim
> J. Graham PhD
> Biology Department
> Washington University of St. Louis

A reply from Rimantas:

Jim probably didn't notice that my letter he references was sent
personally to him only not to Bionet for world-wide distribution. So
I present it here for other also could figure what's all about:

> > As Jean says, a great deal of effort which goes into avoiding RNAses is
> > probably unecessary, as they will not be found in distilled water,
> > properly cleaned glassware, new plastic, or on most people's fingers. If
> > you are spooked by the preponderance of such claims, test these
> > sources yourself with some RNA markers or rRNA!
> Dear Jim,
> as I remember, you claimed in your post a month ago that you had tested
> the fingers of your colegues and yours and hadn't found any RNases.
> I also tested my fingers (but not my colegues' fingers) and I did find.
> I don't like the hypothesis about difference between people on this
> aspect very much. I think the procedures and their sensitivity are
> different. I think we could compare our measurements if we relate the
> activity to the activity of certain concentration of pancreatic
> RNase. When I apply 20 microliters of buffer solution on the tip of my
> finger and wait for a minute, the activity of RNases in the solution
> aproximates the activity of 10 ng/ml of pancreatic RNase. In the first
> and presumably last so far published paper on skin RNase the authors
> used to wash whole fingers in 10 ml of solution and obtained similar
> activities (1- 10 ng/ml of pancreatic RNase). They tested 6
> individuals and all were found 'positive'.
> What procedure did you use ? What is its sensitivity for pancreatic
> RNase ?

The 'first and last so far' reference mentioned is:

R.W.Holley,  J.Apgar, and S.H.Merrill ( 1961  #7,July)
   JBC  236, PC42-PC43.Evidence for the liberation  of  a
   nuclease  from  human fingers.

The second probably will be mine. I've presented a draft two weeks
ago as a poster in the first meeting of the RNA Society in Madison,
WI, USA. I hope to finish the paper and submit to press in a month (
I'm hoping so since one year). I'll be glad to send preprints for all
interested either on paper or electronicaly as it will be ready. I
add some parts of my poster at the end of this post.

I agree with Jim about distilled water, glasware, plastic but human

I worked long time on the problem of so-called Y-A cleavages. This
thing may be not known for people which use RNA only for Notherns or
RT, but must be well known for those performing direct sequencing of
RNA, processing, investigation of secondary structure by chemical and
enzymatic probes etc. While performing such procedures frequently
small amount of products corresponding to cleavages of CA or UA
sequences appears. The extent of cleavage varies from virtually zero
to complete ruin of one's work. These cleavages are stimulated by
nonionic detergents like Triton X-100, Brij and other. Sometimes in
the presence of such compounds the cleavage may be fast enough to
degrade RNA completely in one hour. Some people claims that these
cases are manifestation of intrinsic RNA self-cleavage reaction.
There are several publications with such claims (see the end of this

I spend a lot of time on these cleavages and I've arrived to the
conclusion that they are caused by RNases. Both skin RNase and
pancreatic RNase are highly specific for UA and CA sequences and
produce the same band pattern on PAGE as the alleged "self-cleavage".
(Textbooks say that pancreatic RNase is specific for pyrimidines.
That's true, but the rate of hydrolysis of YA sequences is aprox 10
fold higher than that of YG and aprox 100-fold higher that of YY (Y
means pyrimidine: U or C). Therefore after low-extent hydrolysis by
this RNase only the cleavage of YA sequences is seen).

I never had a chance to meet any of the researchers claiming that
these cleavages are produced by RNA itself neither could discuss this
issue with them. All RNA people I met didn't believe in RNA "self-
cleavage" but hadn't any substantial evidences. So the people I spoke
with in Madison at my poster were glad that at last one guy had done

Because many people observes the cleavages in YA sequences, I have to
admit that skin RNase is quite widespread in human population if not
universal. What makes difference between them is their RNAs and the
work they do with RNA.

For those who would like to join this discussion and test their
I recommend to use 100 mM of monovalent salt (e.g. ammonium or sodium
chloride) in Tris pH7.5-8  without Mg for testing RNase activities.
These conditions are optimal for RNase activity ( monovalent cations
stimulates RNases, Mg suppresses its activity by stabilizing
secondary structure).

I propose using of pancreatic RNase as a standard to relate the
observations by different people using different RNAs and different
testing conditions because it is very similar to finger's RNase (at
least to mine)


R. Plaipa and B. Juodka
Department of Biochemistry and Biophysics, Vilnius
University, Ciurlionio 21, Vilnius, Lithuania

      Starting from the late 70-ies, when high resolution
PAGE  was  introduced  into RNA chemistry,  investigators
frequently  observe  the  low  level  RNA  self-cleavage,
mostly at UA/CA. There are several reports about the very
effective cleavage of some specific RNAs in the  presence
of  nonionic detergents, also in YA sequences. No kinetic
investigation  neither detailed study of  the  causes  of
this  phenomena ever was published, but people  tends  to
believe  that  this  cleavage is not  caused  by  RNases,
rather  that YA sequence represents some minimal ribozyme
with   the   intrinsic  self-cleavage   capability.   Our
experiments,  however, have shown, that RNase  inhibitors
suppress  this reaction completely. RNases  with  such  a
sequence specificity are present both in humans (the skin
RNase)  and microorganisms. Stable cell RNAs, e.g. mature
tRNAs  are  evolutionary optimized and very resistant  to
these RNases in the native conditions, but e.g. pre-tRNAs
are  much more sensitive. If treated carelessly, they may
appear to have the self-cleavage ability.
      Although researchers started to fight the ghosts of
RNases in the very beginning of the RNA world, the  rites
for  killing them are based mainly on the belief yet.  We
have  reviewed and experimentally tested the most popular
ones (autoclaving, DEPC-ing etc.), and we have found that
its   efficiency  strongly  depends  on  pH  and  on  the
composition  of  the solution and may be insufficient  in
practice.  We have performed the screening for RNases  in
the  laboratory environment, and found some parts of  the
surface  of the workplace heavily contaminated  with  Y-A
specific  RNase. The mode of distribution  of  the  RNase
activity  suggests, that it originates from the  settling
of  dead skin cells rather then from immediate touches by
bare hand, and therefore is very difficult to eliminate.


     In  1961  a  short communication "Evidence  for  the
liberation   of  a  nuclease  from  human   fingers"   by
R.W.Holley, J.Apgar, and S.H.Merrill appeared in JBC [1].
These  investigators  had discovered  that  washing  two
fingers  and  the  thumb  in  10  ml  of  water   leases
ribonuclease activity equivalent to that of 1-10 ng/ml of
RNase  A.  No  more investigations of this RNase  to  our
knowledge  ever  were  published, instead  many  legends,
scientific  folklore, myths and unpublished  observations
by   many   researchers  were  created.   Various   rites
(autoclaving, DEPCing etc.) supposed to scare away RNases
from  laboratories were proposed and are used by  various
scientists  with  different degree of fanaticism.  Taking
care  of  RNases indeed is essential to work successfully
with  RNA  and  using  of special  precautions  helps  to
prevent  severe RNA degradation accidents.  Nevertheless,
the  shortage  of  information about  "home"  RNases  and
procedures for their extermination don't allow to be sure
that  RNases  indeed are completely eliminated  from  RNA
     After  introducing 32P labeling and  high-resolution
PAGE into RNA biochemistry, a new phenomena - instability
of  UA/CA sequences had been observed. Incubation of  RNA
in  mild  conditions (pH7, 37 C)for a quite short  time
(e.g. one hour) without any enzymes frequently results in
the  appearance  of a small amount of cleavage  products,
corresponding  to cleavage of mainly UA or CA  sequences.
Control  samples  of  RNA  without   any  treatment  also
frequently  contains the same cleavage products.  Typical
explanation  for these bands, proposed in  1978  [3],  is
that  these  sequences  are fragile  and  hydrolyze  more
easily  than  other.  Another  less  popular  excuse   is
autoradiolysis  of  32P-labeled  RNAs.  Most   of   these
observations  were  made  on  mature  tRNAs.  After   the
discovery  of ribozymes several papers appeared reporting
that some particular RNAs may self-cleave in the presence
of  some  organic  compounds -  Triton  X-100  and  other
nonionic  detergents, PVP etc. [4,5,6,7].  Most  of  such
cleavages  reported  so far are  located  in  UA  and  CA
sequences. These reactions do not require Mg2+  or  other
divalent  metal cations, but are stimulated by monovalent
salts and/or polyamines.
     One  such  RNA, human pre-tRNATyr, was used  in  our
laboratory  as a substrate for tRNA processing  in  vitro
when  a  report about its ability to self-cleave  in  the
presence   of  Triton  X-100  appeared.  We  decided   to
investigate this reaction in more detail, but  this  work
was hindered by very poor reproducibility of the reaction
rate. We found that the slow cleavage of mature tRNAs  at
UA/CA  sequences  also  is  stimulated  by  Triton  X100.
Unreproducibility of the rate of the cleavage,  sometimes
even  complete disappearance had persuaded us  that  this
cleavage  is not an intrinsic reaction of RNA,  but  most
probably  arises from contamination by RNases.  We  found
that  skin RNase indeed has strong preference for CA  and
UA  sequences and gives nearly identical band pattern  on
PAGE  as  the  alleged "self-cleavage". Pancreatic  RNase
also   has   a  nearly  identical  sequence  specificity.
Moreover,  RNases of the similar specificity are  present
in microorganisms too.
     Nonionic detergents like Triton X-100 stimulates RNA
cleavage  by  trace  amounts  of  RNases  presumably   by
stabilizing   them   and/or  preventing   adsorption   by
surfaces.  Anyway,  these  compounds  do  not  have   any
affinity  for  RNA  and do not cause  RNA  cleavage  when
appropriate RNases inhibitors are used.
     In  the  native conditions, i.e. in the presence  of
Mg2+,  mature  tRNAs are very resistant to these  RNases.
Probably they are intentionally designed so since  RNases
specific  for  YA  are  present  in  cytoplasm  of   both
eukaryotes  and  prokaryotes, though  their  function  is
unclear  yet.  RNA  precursors  need  not  to  be  highly
resistant  since they are short lived in vivo.  They  may
posses  unpaired YA sequences highly sensitive to RNases.
Such  RNAs  may  be  unstable in the presence  of  minute
amounts  of  RNases,  which do not cause  any  noticeable
cleavage of mature tRNAs.
     We  tested  the  most popular procedures  for  RNase
inactivation using pancreatic RNase (RNase A) , RNase  T1
and  barnase (RNase from Bacillus amyloliquefaciens), the
best  studied  RNases, as representatives  of  the  RNase
world. This survey have shown that RNA world persists not
because of these procedures, but due to absence of RNases
in  some very important items like distilled water,  most
of chemical reagents etc. Actually, many researchers have
abandoned   paranoiac   implementation   of   all   those
precautions.  Nevertheless, skin RNase really does exist,
and  it  is  the  main  threat  for  RNA  in  laboratory.
Microbial RNases may appear in more special situations  -
because   of   the  growth  of  microbes  in   solutions,
contamination by E.coli grown in labs etc.
     Gloves helps to prevent contamination by touching by
skin RNase. The outer layer of skin, actually dead cells,
wears  and in the form of invisible dust settles  on  all
human  environment. If RNase is present in  these  cells,
RNase  activity  should be present on  all  things  where
humans lives or work. To test this hypothesis, we checked
the surfaces of the leaned plexiglass shield used for the
work with 32P in our laboratory. These shields are rarely
if  any touched by bare hand. This check was made several
times  with shields used by different workers. All  times
the  top side, turned to face of the worker was heavily
contaminated with YA-specific RNase, while "bottom"  side
essentially  clean.  Because none of  the  workers  whose
shields were tested have a habit to rest the nose against
the  shield while working, we must admit, that RNases  do
not  require  direct touch for transfer  from  humans  to
laboratory equipment. This transfer may be one  cause  of
so  widespread RNA "self-cleavage", and the one very hard
to eliminate.

     RNases  unlike  decent  useful  enzymes  dont   get
inactivated  at elevated temperatures due to irreversible
denaturation.  They  as some other low  molecular  weight
proteins renatures very fast when temperature is  lowered
below the melting point (50- 60   C for RNases A , T1 and
likes).  Inactivation occurs only due  to  slow  chemical
reactions  -  mainly deamidation of Asn and cleavage  and
shuffling  of  disulfide  bridges.  The  rate  of   these
reactions  increases with pH. At  90 C  the  rate  of
inactivation of RNase A is 0.13 (1/h) at pH4, 0.56 (1/h) at
pH6 and 23.4 (1/h) at pH8 [2]. Even if the rate would be 10-
fold  higher at 120 C, it wouldn't be enough to  ensure
complete inactivation of RNase A in 30 min at pH4.
     The  inactivation of RNases in common buffers during
standard  autoclaving cycle (120 C,  30  min)  are  the
following :
Media                       Decrease of the activity
                            of RNase A, times

100 mM NH4Ac, pH4.61        4.3 ( RNase T1 - 4 , barnase - 5)
100 mM NH4Ac, pH4.9         5.5
100 mM NH4Ac, pH5.2         9.1
100 mM MES-NaOH, pH6.12     16.3
100 mM MOPS-NaOH, pH7.5     75
100 mM Tris-HCl, pH7.53     11
100 mM Tris-HCl, pH7.8      69
100 mM Tris-HCl, pH8.1      380

     Tris buffer drops out of the general trend due to  a
particular  pH temperature dependence of Tris(-0.03 (1/K),
which means that pH of Tris buffers drops by aprox. 3  pH
units when temperature rises from 20 to 120 C).
     . Although the stability of RNases A and T1 may look
impressive,  they certainly don't hold the record.  Since
RNases  get inactivated mainly due to degradation of  Asn
and  Cys, their thermal stability depends on the  content
of  these amino acids. Some other mould RNases looks much
better in this aspect.

Sequence   |Asn    |Cys per |Source  of  RNase   (PDB
code     in|per    |molecul |code)
Swiss-Prot |molecu |e       |

RNP_BOVIN   10      8        bovine pancreatic RNase
RNT1_ASPOR  9       4        T1 - Aspergillus oryzae.
RNBR_BACAM  6       0        Barnase    -    Bacillus
RNC2_ASPCL  6       4        Aspergillus clavatus.
RNU2_USTSP  9       6        Ustilago sphaerogena
RNF1_FUSMO  10      4        Fusarium     moniliforme
RNMS_ASPSA  1       4        Aspergillus       saitoi
                             (Aspergillus phoenicis).
                             (1rds, 1rms)
RNSA_STRAU  2       3        Streptomyces
RNST_SACER  2       2        Saccharopolyspora
                             erythraea  (Streptomyces
RNS3_STRAU  1       2        Streptomyces

     Unfortunately, no quantitative data on the stability
of  these  RNases are available, but they are  thoroughly
investigated  in other aspects. Great body  of  available
structural  information on many members  of  this  RNases
family  should be very helpful in engineering more stable
RNases  by  minimizing Asp and Cys content  to  zero  and
creating  bugs  producing RNases of so  far  unimaginable
stability (if they don't exist yet in Nature).
     RNases  are inactivated reasonably fast in  alkaline
media. That may be mostly useful for treatment of plastic
and  metal  tools  used for the work with  RNA.  Alkaline
inactivation  proceeds  by the  same  mechanisms  as  the
thermal  one. The inactivation is suppressed by  tertiary
structure of the protein:

Conditions                         Activity  of  RNase  A
                                   remaining after 4 min,
100 mM NaOH, 18   C                aprox 100 ( 43 % after
                                   30 min).
100 mM NaOH, 0.1% SDS, 18   C      65
100 mM NaOH, 30   C                21
100 mM NaOH+0.1% SDS, 30   C       43

     These  results show, that RNase A is not  completely
denatured  at  18    C in 100 mM OH-, and  therefore  the
inactivation  is  stimulated by SDS,  which  unfolds  the
polypeptide  chain  and  exposes sensitive  residues  for
alkali  attack. SDS is not required at 30   C, presumably
because  this  temperature is  enough  to  melt  RNase  A
without the help of other denaturants. SDS slightly slows
down  the  reaction since the association  of  negatively
charged detergent hinders the OH- attack
     Diethylpyrocarbonate   inactivates   RNases   mainly
through  modification  of  imidazol  rings  of  histidine
residues. Thus, the efficiency of the inhibition  depends
on  the quantity and accessibility of His residues in the
protein and also on pH.
     At pH4 RNase A is not inhibited completely even when
treated  with very high DEPC concentrations  up  to  50mM
(0.1%  DEPC equals 6 mM). Also, half of the activity  can
be  recovered  by  treatment  with  hydroxylamine,  which
cleaves His modification product. This product hydrolyses
in water quite rapidly too (half-time about 25 h at 25
C  pH8.0  )  so one cannot be sure that DEPC  inactivates
RNases   in   all  circumstances  and  inactivates   them
     True  irreversible inactivation  of  RNases  may  be
brought  on only by reaction with Lys (but e.g. RNase  T1
doesn't have Lys at all), Tyr and also by destruction  of
His  imidazol  ring  in  the reaction  analogous  to  the
reaction  of DEPC with Ade. This later reaction is  quite
slow and may be left behind by hydrolysis of DEPC.
     For  a  long  time DEPC is known to be destroyed  by
Tris  and  so unsuitable for treatment of Tris solutions.
We  have  found that other buffers increase the  rate  of
hydrolysis of DEPC too.

solution                       half-time of DEPC  at
                               20   C, min

water pH3-7                    40  (170 min at +4 C)
20 mM NaAc pH5.2               23
80 mM NaAc                     14
3M NaAc                        1.0
100 mM NH4Ac pH7               7
200 mM Na phosphate pH6.8      10
20 mM cacodylate pH6.3         2.7
200 mM MES pH6.2               15
200 mM MOPS pH7.2              6.9
200 mM HEPES pH7.5             3.5

     So  the  DEPC  survival time is quite short  in  the
concentrated  stock solutions of all common buffers,  and
that  should severely hinder its efficiency as the  RNase

     The consequences of DEPCing, the smell etc.
     People  are frequently concerned about DEPC left  in
solution  after  treatment with  it  and  possibility  of
inhibition of enzymatic reactions. Because of  the  short
lifetime  of  DEPC in water and water based solutions  no
DEPC  can survive one day or longer. In Tris buffers DEPC
in  destroyed  completely in few  minutes.  Nevertheless,
everybody   who   use   this  compound   probably   knows
characteristic smell which remains virtually indefinitely
after treatment with DEPC. This smell cannot be caused by
DEPC, but rather by some admixture or hydrolysis product.
     DEPC  reaction with water gives ethanol  and  carbon
     C2H5OCOOCOC2H5+H20 ->2C2H5OH+2CO2

     Ethanol also reacts with DEPC:
     giving  rise  to  diethylcarbonate (DEC)  and  ethyl
ester  of  carbonic  acid, which is unstable  and  decays
rapidly to ethanol and carbon dioxide.
     C2H5OCOOH -> C2H5OH+CO2

     This  way ethanol is regenerated and overall process
is the conversion of DEPC to DEC.
     Therefore water which inevitably gets into DEPC when
it  is  stored and used destroys DEPC catalyticaly rather
then    stoichiometricaly,   converting   it    to    DEC
(diethylcarbonate).  Some  DEC  always  is   present   in
commercial preparations of DEPC.
     We  have  performed analysis of several  samples  of
DEPC    and   smelling   DEPC-treated   water   by   gas-
chromatography with mass-spectrometric detection and  DEC
was the only compound found besides DEPC or ethanol.
     So  very likely DEC causes the smell of DEPC-treated
solution.  DEC  is  rather chemically inert  and  is  not
destroyed by heating and autoclaving. Generally it should
not   inactivate  enzymes  too.  Therefore  DEPC  treated
solutions,  even  smelling, are  generally  suitable  for
molecular  biology. Nevertheless, there are  few  reports
about  the adverse effects of DEPC-ing on certain  subtle
process (in vitro transcription and translation [8,9])

[1]R.W.Holley,  J.Apgar, and S.H.Merrill ( 1961  #7,July)
   JBC  236, PC42-PC43.Evidence for the liberation  of  a
   nuclease  from  human fingers

[2]S.E.  Zale  and  A.M.  Klibanov (  1986  Sep  23  #19)
   Biochemistry  25,  5432-5444.  Why  does  ribonuclease
   irreversibly inactivate at high temperatures?

[3]P.Carbon, C.Ehresmann, B.Ehresmann & J.P.Ebel  (  1978
   #1,Oct  1)  FEBS letters 94, 152-156.The  sequence  of
   Escherichia coli ribosomal 16 S RNA determined by  new
   rapid   gel   methods

[4]R.Kierzek ( 1992 #19) NAR 20, 5073-5077.Hydrolysis  of
   oligoribonucleotides:  influence   of   sequence   and

[5]R.  Kierzek  ( 1992 Oct 11 #19) Nucleic-Acids-Res  20,
   5079-5084.       Nonenzymatic      hydrolysis       of

[6]N. Watson, M. Gurevitz, J. Ford, and D. Apirion ( 1984
   Jan  25  #3) J. Mol. Biol. 172, 301-323. Self cleavage
   of  a  precursor RNA from bacteriophage T4

[7]H.van  Tol, H.J.Gross & H.Beier ( 1989 #1) EMBO J.  8,
   293-300.Non-enzymatic excision of pre-tRNA introns ?

[8]P.Gegenheimer,   <pgegen at>post    to
   bionet.molbio.methds-reagnts, 24 Aug 1994.

[9]O.  Brick, post to bionet.molbio.methds-reagnts,  June

I have neither money nor time to create more stable RNases now. This
field is opened for everyone.

I couldn't found DEC in our laboratory neither get it from chemists I
know in my University. I someone has this stuff please smell it and
inform me if it smells like DEPCed solutions.

Rimantas Plaipa
Dept. of Biochem. and Biophys., Vilnius University, Lithuania
Rimantas.Plaipa at GF.VU.LT

More information about the Methods mailing list