|Which insulin to use? Human or animal?
Madras Diabetes Research Foundation
and M.V. Diabetes Specialities Centre, Gopalapuram, Chennai
600 096, India.
CURRENT SCIENCE, Volume 83, Number
12, 25 December 2002
The introduction of insulin was a breakthrough in the
treatment of diabetes and it produced a remarkable increase
life expectancy of diabetic patients. Animal-derived insulins
have been used to treat people with diabetes since insulin
was first discovered and continuously subjected to various
purification technologies. Genetically engineered human
insulin was introduced in 1982 and now the vast majority
of people requiring insulin treatment worldwide are prescribed
synthetic human insulin. Although there exists a debate
on which insulin to use, the decision of choosing a particular
insulin ultimately falls upon the physician who should
make the right choice depending on the diagnosis, expected
outcome and the affordability of the patients. This brief
document provides an account of historical review of human
versus animal insulins and discusses their relative advantages
and disadvantages. The choice of insulin selection has
be thoroughly weighed by the physician focusing on the
patient's clinical and economic status. However in developing
like India, there is still a role for continuing animal
INSULIN injections are needed in all patients with Type
I (insulin dependent) diabetes right from the time of
of diabetes. In Type 2 (non-insulin dependent) diabetic
patients, insulin is needed in those with primary or secondary
failure to oral hypoglycemic agents and at times of stress
like infection, myocardial infarction, etc. Insulin is
indicated in diabetes complicating pregnancy and gestational
diabetes. The primary goal of treatment of diabetes is
of near-normoglycemia to significantly reduce the risks
for long-term microvascular complications of diabetes'.
Today physicians have the choice of at least 10 to 12
different insulin preparations which are in the Indian
newer insulins available today differ from older insulins
in purity, homogenicity and price. While conventional
preparations, though recrystallized, are often heavily
contaminated with other proteins, the new insulins were
or monocomponent, which means that contamination with proinsulin
is less than 1 part per million (ppm). Again while the
insulins are mostly mixtures of beef and pork insulins,
the new insulins are purified monospecies insulins, either
beef, pork, or human structure. With the emergence of recombinant
technology, biosynthetic human insulin has become the
choice of insulin for many physicians. The issue of whether
to use animal or human insulin has been hotly debated
various parts of the country.
A number of attempts world over have been made to answer
this question, including a recent Cochrane review 2,3 .
This article is an attempt to analyse the scientific data
available for various insulin species available in the country
including the Cochrane review in order to understand their
advantages and disadvantages and to provide clarity on which
species to choose when it comes to choosing an insulin.
Origin and preparation
Insulin has been available for the treatment of diabetes mellitus
for the past 80 years following its momentous discovery by
Banting and Best in 1922. Insulin is effective in restoring
normoglycaemia, suppressing ketogenesis and in delaying or
arresting diabetic complications4. The introduction of insulin
was a breakthrough in the treatment of diabetes and it produced
a remarkable increase in the life expectancy of diabetic patients.
The earliest insulin preparations were obtained from beef
pancreas. They were unstable in neutral solution and were
provided to patients in powder or tablet form, which was suspended
in water or saline immediately before injection. Stable amorphous
preparations in acid solution were then developed. The effects
of these lasted for only a few hours when injected subcutaneously.
Intensive efforts were made to obtain longer acting preparations.
Conventional insulins are significantly contaminated with
other pancreatic hormones and Insulin precursors (Proinsulin,
etc.) and this led to investigations that provided stimulus
for further purification of insulin by gel filtration and
ion exchange chromatography that yielded highly purified bovine
Insulins from porcine source have been available from 1923,
though most patients were treated with bovine or bovine-porcine
mixtures. Arrival of the purer porcine insulins in the Danish
market seems to have abolished insulin resistance in Denmark
by 1970, but the highly purified porcine insulins did not
appear until 1972 (ref. 5). The lack of equivalence of porcine
and bovine insulin preparations has always made it difficult
to test formally the clinical impression of greater antigenicity
of bovine insulins, but comparisons of lente insulins of
equivalent purity have recently shown this to be the case.
The highly purified bovine and porcine insulins available
today may not claim to be non-immunogenic, probably because
of the formation of zinc aggregates and insulin polymers
when they are stored. This is further exacerbated when zinc
is complexed with insulin in the preparation of lente insulins
and these considerations are likely to apply to insulins
of any species.
Around 1976, there was a scare that the demand for insulins
was rising so rapidly that it would outstrip the supply
and by 1992 there was an acute shortage of insulins worldwide
even if measures were taken to raise pigs exclusively for
this purpose. Thus it was clear that an alternative to animal
insulins was urgently needed. By 1980, human insulin produced
by recombinant DNA technology had been introduced, thus
ensuring that the world would never run short of insulin
Two methods of synthesis of human insulin have been developed
to the stage of commercial production. Human insulin was
produced by recombinant DNA technology by using E.coli and
yeast ('biosynthetic' human insulin) and the other way by
conversion of pork insulin into human insulin by an amino
acid substitution ('semisynthetic' human insulin)6,7 . The
semi-synthetic human insulin was initially produced by converting
the porcine insulin into human insulin by replacing alanine
with threonine at the B30 position (Table 1). 'Human' insulin
is so called because structurally and chromotographically
it is identical to the insulin produced by the human body
and not because it is extracted from the human body!
Since 1922, diabetic patients have been treated with insulin
preparations extracted from the pancreas of pigs and
Purity of insulin preparations is generally reflected by
the amount of non-insulin-pancreatic proteins in the
Proinsulin content is usually used to reflect purity. Insulins
are defined as purified when they contain 10 parts per
(ppm) of proinsulin. Clinical problems associated with
impurity of insulin preparations include: (1) local
insulin allergies, (2) lipodystrophy at injection sites,
(3) immunologic insulin resistance, and (4) altered
course of action due to antibodies8. Improved
purification in the past decades has resulted in marked
improvement in purity of commercially available insulin.
Amino acid sequence of human. porcine and bovine insulins.
Insulin formulated by recrystallization was only 92% pure,
while chromatographically purified 'single-peak' insulin
contained up to 10,000 ppm of proinsulin. Further improvements
in the production process progressively lowered impurity
level of 100 ppm and later this was further reduced to 10
ppm. Simultaneously the purified 'monocomponent' or 'single
component' insulins have undergone remarkable improvement
such that the purity of these preparations now is less than
1 ppm of proinsulin. In the early years, insulin allergy
was common, but during the last five decades, production
techniques have become progressively more sophisticated,
ultimately leading to the development of highly purified
insulins containing less than 1 ppm proinsulin and virtually
no other pancreatic peptides. When these preparations are
used, especially those of porcine origin, local or systemic
insulin allergy, lipodystrophy or immunological insulin
resistance occur extremely infrequently9.
The chemical nature and properties
In 1955, insulin became the first protein to be fully sequenced.
That work resulted in a 1959 Nobel Prize for Frederick
Many proteins have more than one chain, joined together
in specific ways. Human insulin has two peptides. The
chain has 21 amino acids and B chain has 30 amino acids.
The two chains are connected by two disulphide bridges,
bonds formed between the sulphur atoms in the amino acid
cystine. The A chain also has a third internal disulphide
bridge. The disulphide bridges hold the molecule together.
Although the amino acid sequence of insulin varies among
species, certain segments of the molecule are highly conserved,
including the positions of the three disulphide bonds,
ends of the A chain and the C-terminal residues of the
These similarities in the amino acid sequence of insulin
lead to a three-dimensional conformation of insulin that
is very similar among species and insulin from one animal
is very likely biologically active in other species. Looking
at the enzyme in more detail, the sequence of porcine (pig)
insulin and human insulin is almost identical, but not
- it differs by one amino acid. Bovine (beef) insulin is
different by three amino acids from human.
It is of interest to point out that the newly developed
insulin analogs like Lispro, Aspart and Glargine; also differ
from 'human insulin' in two or more aminoacids as shown
in Table 2.
|Table2: Amino acid sequence
of human and insulin analogs
* The two argines are
added to the C terminus of 15 C B chain.
Due to the fact that the animal insulins (both bovine and
porcine) are different from the natural human insulin in their
amino acid sequence, it is often argued that this could affect
their clinical efficacy. But it appears that none of the amino
acid changes are at sites crucial to the binding affinity
or action of insulins. Hence there is no significant difference
between insulin species in their ability to bind to the receptors
and their action. Theoretically changes in amino acid could
affect the solubility and diffusion properties of insulin
Indeed, receptor studies have shown that
there is complete identity between porcine and human insulins
employing equal potency. The parameters measured include receptor
number, affinity, association and dissociation kinetics, down
regulation negative, comparative and internalization12.
The purer the insulin the better it is for
clinical efficacy and safety profile. It is generally accepted
that purity of insulin preparations is more important for
immunogenicity and allergenicity than the species specificity13.
Purity abolishes and overcomes insulin resistance. Insulin
impurities, not the insulin itself, was responsible for the
immunogenicity of recrystallized insulin in patients. The
relentless efforts made in the past five to six decades have
resulted in purification techniques that could yield insulin
with less than 1 ppm purity (the monocomponent insulins).
Thus today the purity is hardly an issue and in most clinical
studies, human insulin was shown to be indistinguishable from
porcine insulin of comparable purity with respect to plasma
glucose and glycosylated hemoglobin levels and insulin dose
Essentially all patients who receive insulin for long periods
develop antibodies to insulin (even antibodies to endogenous
insulin is known). Hence antigenicity will exist even if completely
purified insulins are made available, as it is a fundamental
property of polypeptide hormones.
Antigenicity is largely related to purity and site of action.
Indeed even human insulin preparations made by biotechnological
or chemical techniques could be less contaminated by such
derivatives (desamido insulin, arginyl insulin, insulin ethylesters).
These may well be more difficult to separate from the insulin
than the contaminants found in pancreatic extracts. In a clinical
study performed by Larkins et al., they found that the human
insulin was no less antigenic than porcine insulin; significant
IgG associated insulin binding activity was detected in six
of the ten patients in the human insulintreated group and
four of ten patients in the porcine insulin-treated group14.
Clinical efficacy of insulins clearly does not depend on the
species of insulin used. A number of clinical trials have
clearly shown that animal insulins and human insulins are
comparable in their clinical efficacy. It is also claimed
that the doses of human insulin required was also less compared
to animal insulin, but this was only true when the animal
insulins were impure. The duration of action of human insulin
is slightly shorter than animal insulins. This is a slight
disadvantage as patients on twice daily insulin tend to have
higher late evening sugars if on human insulin unless a noon
dose is also introduced.
The Cochrane review compared the effects
of synthetic human insulin and natural animal insulins in
diabetic patients from 1966 to May 2002 (ref. 2). The objective
of the review was to assess the effects of different insulin
species and evaluating their efficacy (in particular glycemic
control) and adverse effects profile (mainly hypoglycemia).
For which a highly sensitive search for randomized controlled
trials combined with key terms for identifying studies on
human versus animal insulin was performed using Cochrane Library
(issue 2, 2002), Medline (1966 to May 2002) and Embase (1974
to Feb 2002) including the reference lists and databases of
ongoing trials. They included randomized controlled trials
with diabetic patients of all ages that compared human to
animal (for the most part purified porcine) insulin. Altogether
2156 participants took part in 45 randomized controlled studies
that were discovered through extensive search efforts. At
the end of the review, the reviewers made the following conclusions:
- A comparison of the effects of human and animal insulin
as well as of the adverse reaction profile did not show
clinically relevant differences.
- No differences were found in metabolic control, and no
differences in HbAlc between 'human' and 'animal' insulins.
- Many patient-oriented outcomes like health-related quality
of life or diabetes complications and mortality were never
investigated in high quality randomized clinical trials.
- Most of the studies comparing the two insulins were of
poor methodological quality.
- None of the studies assessed the costs or socio- economic
It can be concluded that both human and animal
insulins are equally good and the decision to use one or other
of the insulin rests entirely with the physician.
Switching from one source of insulin to another can cause
difficulties in controlling blood sugar level and subsequently
patients should have their doses readjusted. It has been reported
that some diabetic subjects when switching from animal to
human insulin, lose their usual warning symptoms like sweating,
tiredness, etc. associated with low blood sugar or 'hypoglycemia.
This has been termed 'hypoglycemia unawareness'. It has been
claimed by some authors that the clinical symptoms of hypoglycernia
while taking human insulin are significantly less pronounced
when compared to porcine insulin' 5,16 . However, surprisingly
these studies were mainly from UK and despite human insulin
being exclusively used in USA, Australia and other countries
there have been very few reports of hypoglycemia unawareness
in these countries. In the author's view also there does not
appear to be any hypoglycemic unawareness issue with human
insulin given that thousands of patients have been treated
with human insulin.
Since the production of recombinant human insulin, numerous
studies have been performed to compare its efficacy against
that of porcine insulin in treating diabetes. Many of these
studies showed that there were no differences in ability to
transport glucose, and that neither insulin is less reactive
than the other with insulin antibodies. It is unlikely that
an improvement in diabetic control can be achieved merely
by changing to human insulin. In any case, whether animal
or human insulin, insulin therapy should aim to normalize
not only blood glucose levels but also the HbAlc (glycosylated
haemoglobin), which is an index of blood glucose control for
2-3 months and thus prevent long-term complications of diabetes.
Therefore, the decision of choosing a particular insulin ultimately
falls upon the physician who should make the right choice
depending on the need of the hour and the affordability of
Currently human insulins are almost twice
as expensive as porcine insulin. Hence in developing countries
like India, animal insulin should continue to be available.
It would be a great pity if patients could not afford the
insulins and die merely due to socio-economic reasons. There
are moves to produce human insulins in India with the promise
of delivering high quality human insulins at low cost. Until
this becomes a reality there is place for both human and animal
insulins and the choice is entirely left to the physicians
and perhaps more importantly the patient. In situations where
insulin therapy is only for a short period, e.g. pregnancy,
infections, etc. it may be better to use human insulin as
cost implications are not that serious. Whenever long-term
insulin therapy is encouraged, it is better to check the affordability
of the patient and decide which insulin to use.
- Liebl, A., Diab. Metab. Res. Rev., 2002,
- Richter, B. and Neises, G., Cochrane Database
Syst. Rev., 2002, 3, CDO03816.
- Richter, B., Neises, G. and Bergerhoff,
K., Endocrinol. Metab.Clin. North Arn., 2002, 31, 723?749.
- Garg, M. K., J. Indian Med. Assoc., 2002,
100, 194-195, 202.
- Deckert, T., Andersen, O. O. and Poulsen,
J. E., Diabetologia, 1974, 10, 703?708.
- Larkins, R. G., Austr. N. Z. J. Med.,
1983, 13, 647-651.
- Raptis, S. and Dimitriadis, G., Clin.
Physiol. Biochem., 1985, 3, 29-42.
- Home, P. D. and Alberti, K. G., Drugs,
1982, 24, 401-413.
- Markussen, J., Diabetologia, 1983, 25,
- Steiner, D. F., Bell, G. I. and Tager,
H. S., in Endocrinology (ed. DeGroot, L. G.), WB Saunder,
Harcourt Brace Jovanowich, Phila delphia, 1990, p. 1263
- Jacobs, S., Insulin: Its Receptor and
Diabetes, M. Dekker, New York, 1985, pp. 39-51.
- Home, P. D. and Alberti, K. G. M. M.,
Clin. Endocrinol. Metab., 1982, II, 453-483.
- Schernthaner, G., Diab. Care, 1993, 16,
- Larkins et al., N. Engl. J. Med., 1968,
- Roth, C., Landolt, H. P., Achermann,
P., Teuscher, A. and Borbely, A.A.,Sleep, 1998, 21,
- Jakober, B., Lingenfelser, T., Gluck,
H., Maassen, T., Overkamp, D., Renn, W. and Eggstein,
Klin. Wochenschr., 1990, 68, 447-453.