Scientific scrutiny: 2012

Tuesday 4 December 2012

Research and rediscovery?

I have been reading about so many cases of fabrication, respected, well cited scientists who have happily and diligently fabricated papers and research proposals for many years, that it is quite clear that journal editors are only too eager to accept papers from people who have been around for a while.

Of course, what do we believe and what should we expect when we pick up a journal and refer to its impact factor to check its "worth"? What constitutes a good research article and a good study to write up in "good research article(s)"? Are we all really that inventive that we have to keep coming up with new discoveries?

Obviously not. But when we apply existing knowledge to a study, should our hypotheses not take into account the scientific validity of our assumptions? We may re(^n)discover that low pH and urea denature proteins, but should our methodology not take into account these facts and not base a study on such ideas? Apparently not as we see below--not sure why two papers come up with the same title:

1) Shiva Ranjini S. & Vijayalakshmi M.A. Study of catalase adsorption on two mixed-mode ligands and the mechanism involved therein., Journal of molecular recognition : JMR, PMID: 23108613

2) Fleminger G., Ehrlich G., Shiva Ranjini S. & Vijayalakshmi M.A. (2012). Study of catalase adsorption on two mixed-mode ligands and the mechanism involved therein, Journal of Molecular Recognition, 25 (11) 542-548. DOI: 10.1002/jmr.2181

------------------------------------------------------------------------------------------------------------
These are the main sorbents in this study (HEA and PPA HyperCel ligands):

--------------------------------------------------------------------------------------------------------

The forms of catalase being investigated:

---------------------------------------------------------------------------------------------------------

Adsorption and elution buffers (1,2: pH 7; 3,4: pH 4.5 elution; 1,4: adsorption buffers do not contain salt)

---------------------------------------------------------------------------------------------------------

Equilibriation and elution:

-----------------------------------------------------------------------------------------------------------

They elute the PPA sorbent with urea....

-----------------------------------------------------------------------------------------------------------

The abstract from the earlier catalase purification paper from this group also appears below.

In spite of an optimal pH of 7 and stability seen over pH 6-10 for catalase, the authors of this 2012 JMR paper have chosen to elute at pH 4.5. And for some reason, the authors first state that the catalase is unstable below pH 4.5 (to explain why they didn't go below pH 4.5) and then contradict this to state that catalase was stable from pH 5-9.
------------------------------------------------------------------------------------------------------------

-----------------------------------------------------------------------------------------------------------
From the earlier paper:

-----------------------------------------------------------------------------------------------------------

Funny that they ran their SDS-PAGE in non-reducing conditions on BLC (bovine liver catalase) eluted from PPA with UREA.

8M urea (lane 5 above)!!!

Is it not well known that urea denatures proteins? In case, people need references, here's one I found on the Internet. Low pH, urea ... can these be the basis of trying to separate proteins from other components when it is well known that they denature proteins. Don't the proteins of interest also have to be active in the end?

------------------------------------------------------------------------------------------------------------

-----------------------------------------------------------------------------------------------------------

Swing high (pH 8) if you have swung low....

-----------------------------------------------------------------------------------------------------------
The same assumptions about temperature promoting adsorption. Still, Figures 1 and 2 showing protein recovery from HEA for bovine liver catalase at 24 and 37 degrees Celsius, respectively, show that even if pH 4.5 in the elution buffer decreases the difference in protein recovery from retained and non-retained fractions, there is lower recovery of protein in the retained and eluted fractions and also in the non-retained fractions at 37 degrees Celsius.

----------------------------------------------------------------------------------------------------------

Some more re(^n)discoveries: 1) Molecules interact with each other at multiple points, not just at one point.

2) The pi electrons of the phenyl group participates in interactions.

-----------------------------------------------------------------------------------------------------------

3) Chromatography is based on changing pH and conductivity conditions for adsorption and elution of different molecules from different sorbents.

4) Also, the nature of the material being purified influences this process....

----------------------------------------------------------------------------------------------------------

And to stress these points:

3) Chromatography is based on changing pH and conductivity conditions for adsorption and elution of different molecules from different sorbents.

4) Also, the nature of the material being purified influences this process....

Shiva Ranjini S, & Vijayalakshmi MA (2012). Study of catalase adsorption on two mixed-mode ligands and the mechanism involved therein. Journal of molecular recognition : JMR, 25 (11), 542-8 PMID: 23108613

Saturday 3 November 2012

Recyclable CIM disks and papers

Another paper on anti-glycophorin-A MAb "purification" (compare to the one described in the post "Acid Test for antibody function?" by Ambur P. Dhivya's 2010 paper in Chromatographia)

CIM-EDA has been used in both papers. In this latest one (2012), ~ 8 fold purification was seen with elution with 25 mM MOPS, pH 6.5 and 0.2 M NaCl.
In the older one (2010), 20 mM Tris buffer, pH 8, 0.6 M NaCl gave ~74 fold purification.

Images are still difficult to make out; but why change the conditions and report worse "purification" with different conditions?

abstract from Ambur Dhivya et al. 2010 Chromatographia

Dead cells producing antibody?

Side-by-side comparisons of chromatographic purification gels and profiles from the older (blue arrows and underlined in blue) Chromatographia 2010 paper and the recent (red arrows and underlined in red) Hybridoma 2012 paper.

Kumar BP, Rajak P, Vijayalakshmi MA, & Jayaprakash NS (2012). Production of Human Anti-Glycophorin-A Monoclonal Antibodies and Their Purification by Pseudoaffinity Chromatography Using a Convective Interaction Media Monolithic Column Hybridoma DOI: 10.1089/hyb.2011.0104

Dhivya AP, Kumar BP, Prasanna RR, Jayaprakash NS, & Vijayalakshmi MA (2010). Purification of Monoclonal Antibodies from Cell-Culture Supernatant by Use of Anion-Exchange Convective Interaction Media (CIM) Monolithic Columns Chromatographia, 72 (11-12) DOI: 10.1365/s10337-010-1787-3

Rediscovering the basics

Another paper on the purification of anti-HSA IgG1 MAb using CIM-IDA disks immobilized with four different metal ions. This time, the authors report superior purification with Zn (II) and Co (II) ions, with Zn (II)) being supposedly far better.

Again, similar problems as before: images with brightness and contrast adjusted to a point where the marker (especially in the "best" Zn (II) case) itself has lost some bands. In fact, the only ion with which they get one band is with Co (II) but its recovery does not appear to be very good.

Ironically in this paper, the authors make a statement (shown below) about how acidic pH is bad for their antibody. Given that the senior authors appear to have published many papers using pH 4 or pH 5.5 elution or equilibriation conditions (please see other posts in this blog), why this sudden discovery and what about those earlier antibodies and reports?

Poonam Rajak et al. 2012 Ambur P Dhivya et al. 2010

Fairly clean blot compared to other blots from this group with only slight degradation evident in the IgG1 MAb lane.

Rajak P, Vijayalakshmi MA, & Jayaprakash NS (2012). Purification of monoclonal antibodies, IgG1, from cell culture supernatant by use of metal chelate convective interaction media monolithic columns Biomedical Chromatography, 1488-1493 DOI: 10.1002/bmc.2721

Thursday 31 May 2012

Alcoholic extracts of Aloe vera for diabetic rats

This 2008 paper describes the antidiabetic activity of an Aloe vera gel extract with histological data to show the changes in STZ-induced diabetic rats. This critique of the paper highlights the use of important data like extract doses (2004 J.Med. Food paper by Rajasekaran et al.), expected results, key hypotheses from earlier papers and the omission of important details such as the preparation of the different extracts of which the test extract was chosen as the best one. I couldn't get the 2004 J.Med.Food paper but the 2006 paper in Clin. and Exper. Pharmacol. Physiol. by the same authors mentions some of the data in the 2004 paper.

Four extracts made (DATA NOT SHOWN). Extract 2 taken for further study. The gel is refluxed with absolute ethanol and dried in a rotary vacuum evaporator of 80 degrees C. If alcohol can actually extract some of the polysaccharides in the gel, drying at 80 degrees C will cook this extract which is probably why the authors get greenish brown powder.

A 2008 review by C. T. Ramachandra and P. Srinivasa Rao in the American Journal of Agricultural and Biological Sciences (vol. 3, issue 2, pages 502-510) mentions that Aloe vera gel can be heated to 65 degrees Celsius for less than 15 min or to 85-95 degrees Celsius for 1-2 min. Even if oxidation of certain compounds in the gel is important for antidiabetic activity, this heating at 80 degrees Celsius is a long one!

Contrast this with the procedure described by Rajasekaran et al. in a 2006 paper. The pulp (or gel) was lyophilized, extracted with 95% ethanol and dried (no temperature mentioned) to remove the solvent. These authors also published a paper in 2004 in J. Med. Food which has been cited by Noor et al. in the 2008 paper.

------------------------------------------------------------------------------------------------------------

The 2006 paper by Rajasekaran et al. summarizes the reduction in blood glucose in STZ-induced rats receiving a dose of 300 mg/kg Aloe vera relative to control diabetic rats (21 days after administration). The 2008 Noor et al. paper uses the same dose of 300 mg/kg Aloe vera extract (also three weeks) and gets approximately the same effect. In fact, Rajasekaran et al. used 55 mg/kg bw STZ to induce diabetes, Noor et al. used 30 mg/kg STZ.

------------------------------------------------------------------------------------------------------------

Ghannam et al. (1986) used dried sap of Aloe vera and Rajasekaran et al. (2004, 2006) used 95% ethanol to extract Aloe vera pulp.

-----------------------------------------------------------------------------------------------------------

Helal et al. (2003) already showed antidiabetic activity of Aloe vera in alloxan-treated diabetic rats.

-----------------------------------------------------------------------------------------------------------

Would alcohol not extract the same principles water can PLUS other more non-polar molecules? Could these kidney-related changes be due to the higher dose of STZ than due to alcohol used for the extraction?

This 2008 Noor et al. paper represents the increasing trend for peer reviewers to not have much time to read the manuscripts to do a thorough job and for the omission of important experimental details that will leave readers unable to reproduce much of the work described in the paper.

Noor, A., Gunasekaran, S., Manickam, A.S. & Vijayalakshmi, M.A. (2008). Antidiabetic activity of Aloe vera and histology of organs in streptozotocin-induced diabetic rats, Current Science, 94 (8) 1078. DOI:

Sunday 20 May 2012

Solution to ScFv to refolding

<-----------------J. Biotech. 2011 (LEFT SIDE)

Sushma, K., Vijayalakshmi, M.A., Krishnan, V. & Satheeshkumar, P.K. (2011). Cloning, expression, purification and characterization of a single chain variable fragment specific to tumor necrosis factor alpha in Escherichia coli, Journal of Biotechnology, 156 (4) 244. DOI: 10.1016/j.jbiotec.2011.06.039

J. Chrom B 2012 (RIGHT SIDE) --------->

Sushma, K., Bilgimol, C.J., Vijayalakshmi, M.A. & Satheeshkumar, P.K. (2012). Recovery of active anti TNF-α ScFv through matrix-assisted refolding of bacterial inclusion bodies using CIM monolithic support, Journal of Chromatography B, 891-892 93. DOI: 10.1016/j.jchromb.2012.02.011

----------------------------------------------------

This section 3.2 is from the J. Biotech. paper and is an illustration of the rising trend of omitting experimental data which would be relevant to the study.

(J. Chrom B paper on the right)------>

Should reviewers be OK with images of such quality?

J. Biotech paper (Figure 5)

----------------------------------------------------------------------------------------------------------

This paragraph is from the J. Biotech paper. It appears that the ScFv from soluble fraction was less potent than Liu et al's anti-TNF alpha ScFv (already reported in 2007) and probably Geng et al's as well (?).

Why go for mutagenesis to bring something to a level that others have already shown or exceeded? What are the advantages of E.coli over those expression systems used by Liu and Geng and their co-workers?

------------------------------------------------------------------------------------------------------------

Notice that in spite of the very close similarity of the assays and the data, the J. Biotech paper's Fig. 6 legend above states that "control (untreated in figure) cells were without any TNF-alpha and anti-TNF-alpha ScFv treatment." Does this mean that control cells got anti-TNF-alpha treatment and not TNF-alpha?

J. Chrom B's Fig. 4's legend states that "control cells were without any treatment." As the results are so similar, would this mean that the above wording actually means this "without any treatment"?

The results themselves show that untreated cells serve as the control for 100% viability in this MTT assay because they (as the 2012 J. Chrom B paper states clearly) are completely untreated.

Why then are there points all along the X axis in the form of diamonds in the J. Biotech paper and in the form of crosses in the J. Chrom B paper?

What concentrations are these and of what?

-------------------------------------------------------------------------------------------------------------

Duplicate (redundant) publication – refers to repeated publication of a manuscript when there is clear overlap with
a previous manuscript already published. Such overlap is possible when there is distinct similarity in the basic assumptions underlying the research, the sample characterization, identity of patients, research methods, research findings or conclusions.

Salami publication- refers to splitting the research findings from a single study and dispersing them in different publications or at different times, with the intended purpose of increasing the number of publications. (Israel Medical Association)
-----------------------------------------------------------------------------------------------------------

The following two papers cover the expression, purification and characterization of a single chain variable fragment (ScFv) specific to TNF alpha in E. coli: one deals with the soluble fraction (J. Biotech, 2011), the other solubilizes the insoluble fraction (J. Chrom B, 2012).

Analysis of both papers will be presented side-by-side (if space permits) or the figure of the (later) J. Chrom B paper will follow that of the (earlier) J. Biotech. paper to allow comparison.

Would the J. Chrom B paper qualify as a salami or a redundant or just a me-too paper?
-----------------------------------------------------------------------------------------------------------

J. Biotech. 2011 (LEFT SIDE) J. Chrom B 2012 (RIGHT SIDE)